youmetme/rabat-illness-yolov5
Archived
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

dev #1

Merged
youmetme merged 7 commits from dev into main 2024-06-19 15:00:21 +08:00
Conversation 0 Commits 7 Files Changed 27 +1786 -30
16 changed files with 1786 additions and 30 deletions
Download Patch File Download Diff File Expand all files Collapse all files
Showing only changes of commit 14f6261d71 - Show all commits

5
.gitignore vendored
View File

@@ -1 +1,6 @@
database/chestXray8_512/*
database/*
logs/*
!*.md

2
README.md
View File

@@ -5,4 +5,4 @@
# 数据集介绍
[详细见](./database/README.md)
[详细见](./database/Dataset.md)

3
config.ini
View File

@@ -43,3 +43,6 @@ mixup_prob = 0.5
special_aug_ratio = 0.7
; label_smoothing 标签平滑。一般0.01以下。如0.01、0.005。
label_smoothing = 0.01
[dataset]
classes_path = model_data/voc_classes.txt

13
database/Dataset.md Normal file
View File

@@ -0,0 +1,13 @@
# 数据集介绍
数据集来自kaggle[数据集原地址](https://www.kaggle.com/datasets/spritan1/yolo-annotated-chestxray-8-object-detection)
## 标签
1. 肺不张，Atelectasis
2. 心脏肥大，Cardiomegaly
3. 很有效率，Effusion
4. 渗透，Infiltrate
5. 诺依，Nodule
6. 质量，Mass
7. 肺炎，Pneumonia
8. 气胸，Pneumothorax

14
database/README.md
View File

@@ -1,13 +1 @@
# 数据集介绍
数据集来自kaggle[数据集原地址](https://www.kaggle.com/datasets/spritan1/yolo-annotated-chestxray-8-object-detection)
## 标签
1. 肺不张，Atelectasis
2. 心脏肥大，Cardiomegaly
3. 很有效率，Effusion
4. 渗透，Infiltrate
5. 诺依，Nodule
6. 质量，Mass
7. 肺炎，Pneumonia
8. 气胸，Pneumothorax
# 存放数据集

639
model_data/2007_train.txt Normal file
View File

@@ -0,0 +1,639 @@
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000002_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000005_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000005_002.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000005_004.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000005_005.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000006_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000007_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000008_001.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000011_001.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000011_002.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000011_003.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000011_004.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000011_008.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000013_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000013_008.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000013_014.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000013_015.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000013_016.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000013_017.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000013_019.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000014_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000015_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000016_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000017_001.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000017_002.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000018_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000022_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000023_001.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000023_003.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000029_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000032_003.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000032_005.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000032_009.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000032_010.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000032_019.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000032_020.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000032_022.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000032_025.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000032_046.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000032_048.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000032_049.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000032_051.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000034_001.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000035_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000037_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000038_004.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000039_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000039_001.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000039_002.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000039_003.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000040_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000040_002.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000041_003.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000041_004.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000042_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000042_002.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000042_004.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000042_008.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000046_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000047_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000047_001.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000047_002.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000047_005.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000047_007.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000048_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000049_001.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000049_002.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000050_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000050_001.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000050_002.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000050_003.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000052_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000052_001.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000054_001.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000054_002.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000054_003.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000054_005.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000054_008.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000055_000.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000072_000.png 178,283,242,325,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000147_001.png 308,299,359,389,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000149_006.png 301,230,426,275,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000181_061.png 104,284,211,411,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000211_010.png 176,267,227,351,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000211_019.png 170,211,420,414,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000211_041.png 112,208,368,423,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000344_003.png 86,115,135,226,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000377_004.png 193,209,445,393,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000398_003.png 184,194,426,382,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000457_004.png 401,140,448,193,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000468_017.png 74,259,130,282,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000468_033.png 102,251,160,277,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000583_008.png 146,227,251,257,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000643_002.png 409,269,457,307,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000661_000.png 149,218,423,432,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000732_005.png 213,232,420,404,0 306,55,393,107,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000740_000.png 150,192,389,385,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000744_006.png 73,254,144,314,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000756_001.png 144,162,445,377,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000808_002.png 279,192,392,275,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000830_000.png 287,244,428,302,3 154,199,203,259,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000845_000.png 185,183,412,396,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000865_006.png 419,312,458,332,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000902_001.png 175,317,219,359,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00001075_024.png 145,163,222,262,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00001153_004.png 153,126,202,346,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00001170_046.png 317,191,440,288,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00001248_038.png 283,47,410,131,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00001369_000.png 168,138,445,331,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00001373_009.png 206,234,480,484,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00001373_039.png 130,250,411,495,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00001534_005.png 198,171,443,361,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00001555_002.png 307,404,344,442,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00001558_016.png 42,209,156,417,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00001673_016.png 140,274,166,324,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00001836_041.png 313,192,329,205,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00001836_082.png 85,284,166,339,3 168,194,238,251,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00001946_029.png 144,198,181,275,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00002059_008.png 187,232,464,415,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00002106_000.png 162,181,188,207,2 318,214,383,291,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00002176_007.png 129,94,183,152,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00002290_001.png 130,358,161,382,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00002350_001.png 363,240,437,284,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00002533_002.png 140,13,244,280,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00002583_014.png 304,198,463,402,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00002763_031.png 201,177,439,371,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00003072_028.png 12,169,154,394,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00003148_004.png 298,252,326,342,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00003394_006.png 156,187,377,339,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00003400_003.png 304,28,424,142,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00003440_000.png 108,127,170,167,3 295,228,353,285,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00003787_003.png 144,235,228,294,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00003803_010.png 85,295,405,361,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00003945_004.png 34,235,187,319,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00003948_001.png 140,152,192,216,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00004296_000.png 365,376,425,423,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00004342_002.png 100,243,339,483,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00004342_050.png 92,180,160,391,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00004344_013.png 175,173,437,384,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00004344_014.png 164,156,398,381,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00004344_018.png 195,196,448,411,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00004344_022.png 188,254,420,441,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00004344_046.png 184,166,424,353,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00004381_021.png 194,180,421,343,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00004461_000.png 164,196,370,370,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00004533_014.png 225,153,449,349,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00004534_001.png 128,214,403,422,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00004547_003.png 449,345,476,374,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00004578_004.png 216,195,432,381,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00004630_001.png 142,219,398,391,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00004822_051.png 179,184,441,381,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00004893_085.png 192,171,424,366,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00004911_018.png 347,159,398,204,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00004968_003.png 320,342,453,395,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00004968_004.png 317,318,449,380,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00005066_005.png 165,243,419,418,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00005066_030.png 138,229,408,380,0 74,364,88,397,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00005089_002.png 324,287,455,381,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00005089_014.png 89,238,159,274,3 322,93,436,391,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00005089_040.png 309,49,434,389,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00005140_001.png 62,131,132,354,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00005532_000.png 191,131,429,338,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00005532_014.png 169,165,399,338,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00005532_016.png 162,208,369,378,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00005532_019.png 341,117,377,146,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00005827_000.png 196,153,422,389,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00005869_001.png 90,33,213,73,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00006736_000.png 385,236,407,259,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00006751_000.png 87,222,125,255,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00006821_002.png 60,42,251,380,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00006851_033.png 94,312,194,339,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00006851_034.png 125,308,205,369,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00006912_007.png 165,196,385,359,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00007037_000.png 144,243,394,436,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00007043_000.png 145,180,410,399,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00007120_009.png 124,274,192,318,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00007124_008.png 272,244,426,337,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00007471_003.png 270,53,437,294,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00007551_020.png 176,261,405,436,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00007557_026.png 65,353,176,415,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00007676_002.png 180,219,221,288,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00007710_000.png 312,219,420,263,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00007735_040.png 199,238,470,462,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00007882_001.png 75,321,176,350,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00008008_027.png 323,244,371,295,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00008339_010.png 174,173,437,366,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00008365_000.png 168,254,377,456,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00008386_000.png 355,178,385,202,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00008399_007.png 181,189,455,375,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00008522_032.png 157,104,374,301,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00008547_001.png 403,340,474,411,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00008554_009.png 303,247,418,331,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00008716_000.png 353,354,429,406,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00008814_010.png 97,203,432,236,3 97,206,430,259,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00008841_025.png 77,243,210,293,1 292,40,388,83,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00008841_044.png 73,23,235,106,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00009166_004.png 78,121,119,273,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00009218_015.png 96,109,207,405,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00009229_007.png 361,273,396,310,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00009256_005.png 152,196,208,254,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00009342_000.png 429,214,506,340,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00009403_006.png 374,208,412,234,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00009437_008.png 376,297,457,336,3 371,290,465,429,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00009608_024.png 197,201,379,372,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00009619_000.png 150,329,204,385,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00009669_003.png 78,324,209,434,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00009683_005.png 16,179,153,330,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00009705_000.png 169,154,440,339,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00009745_000.png 164,199,392,391,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00009779_001.png 355,247,437,281,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00009889_018.png 62,379,79,399,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00010007_168.png 166,360,194,420,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00010103_014.png 171,206,192,229,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00010125_004.png 73,150,106,194,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00010172_001.png 94,347,118,373,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00010277_000.png 431,346,467,402,1 148,155,418,294,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00010381_000.png 195,156,422,336,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00010478_012.png 106,293,229,394,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00010481_021.png 381,275,461,310,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00010575_002.png 116,302,233,381,3 347,295,471,383,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00010625_014.png 135,221,408,290,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00010767_008.png 101,95,184,166,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00010767_016.png 58,88,214,384,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00010770_000.png 122,185,225,219,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00010815_006.png 155,120,232,242,3 155,120,228,248,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00010828_039.png 294,234,400,348,3 351,223,470,342,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00010936_011.png 135,290,187,389,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00010936_016.png 87,82,203,151,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00010959_010.png 106,245,145,275,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00011151_004.png 165,103,353,286,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00011157_001.png 100,245,163,306,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00011237_094.png 64,114,144,369,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00011263_004.png 156,196,433,437,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00011269_019.png 366,209,475,303,1 286,133,351,204,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00011322_006.png 182,242,423,399,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00011402_007.png 102,240,349,412,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00011450_000.png 367,247,416,295,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00011463_002.png 189,272,410,444,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00011502_001.png 221,158,404,349,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00011576_000.png 103,291,126,312,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00011814_031.png 84,227,178,377,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00011827_003.png 284,235,437,297,3 376,74,440,366,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00011831_010.png 117,41,237,127,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00011832_002.png 372,175,407,212,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00011857_001.png 357,268,462,309,3 347,256,481,339,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00011925_072.png 188,127,269,329,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00011925_076.png 61,53,209,240,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012045_009.png 358,280,408,339,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012048_007.png 143,190,217,266,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012094_040.png 288,55,361,108,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012123_001.png 319,305,456,413,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012174_000.png 84,79,178,349,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012261_001.png 226,216,493,426,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012270_005.png 371,117,428,173,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012291_008.png 116,315,229,353,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012374_000.png 327,128,368,168,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012376_010.png 370,105,433,317,2 187,71,247,99,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012376_011.png 385,240,422,279,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012415_002.png 199,267,236,310,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012505_007.png 127,84,246,350,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012515_002.png 144,319,186,347,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012622_016.png 381,83,422,147,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012637_000.png 364,249,430,291,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012670_000.png 213,210,428,368,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012686_003.png 162,205,430,406,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012741_004.png 221,217,444,421,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012793_000.png 176,189,423,391,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012834_008.png 55,206,154,242,3 2,179,139,390,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012834_034.png 384,113,487,292,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012834_122.png 38,47,167,484,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012892_010.png 187,112,240,179,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012973_005.png 61,31,201,412,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012975_003.png 103,118,206,219,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013031_005.png 9,209,89,406,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013051_000.png 133,275,162,303,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013111_069.png 237,201,309,331,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013118_008.png 112,273,155,313,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013125_000.png 84,203,415,412,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013187_002.png 99,283,468,371,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013249_031.png 191,239,393,368,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013249_033.png 207,201,441,373,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013285_026.png 89,249,108,383,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013337_000.png 30,157,169,381,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013346_015.png 186,211,416,472,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013471_002.png 64,346,199,396,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013508_001.png 161,296,250,341,3 143,258,188,305,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013615_052.png 175,203,407,381,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013635_002.png 181,165,412,350,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013659_019.png 279,83,330,152,2 145,42,199,67,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013674_000.png 105,159,134,188,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013685_028.png 132,246,392,318,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013750_016.png 98,91,195,228,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013751_003.png 338,305,383,350,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013807_009.png 157,61,218,82,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013885_000.png 218,136,240,158,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013911_000.png 59,241,101,292,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013951_001.png 150,133,182,158,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013977_005.png 390,392,452,482,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013991_000.png 190,185,243,232,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013992_006.png 140,263,204,303,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013993_077.png 83,203,184,315,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013993_083.png 68,108,196,457,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014004_038.png 172,198,227,267,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014014_013.png 324,191,409,276,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014083_023.png 320,305,458,371,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014095_003.png 148,248,234,295,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014116_009.png 348,159,375,188,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014177_010.png 94,194,158,310,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014223_009.png 112,193,354,337,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014253_010.png 18,123,141,165,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014346_010.png 76,72,219,379,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014398_031.png 94,231,189,395,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014447_004.png 106,210,188,258,3 77,214,124,348,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014551_010.png 293,205,372,295,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014574_000.png 184,182,403,380,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014607_007.png 100,330,421,437,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014626_017.png 135,67,186,142,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014626_035.png 75,85,183,478,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014687_001.png 363,247,433,275,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014706_018.png 185,223,397,368,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014716_007.png 343,65,435,222,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014731_028.png 386,132,443,203,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014738_000.png 95,60,220,350,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014795_002.png 144,270,218,337,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014822_039.png 322,102,419,414,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014870_004.png 57,306,169,353,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00015018_004.png 402,180,430,209,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00015141_002.png 398,301,423,334,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00015262_005.png 68,267,143,315,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00015300_000.png 85,197,159,271,3 107,237,146,264,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00015304_001.png 145,167,453,414,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00015400_001.png 149,175,389,380,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00015440_000.png 318,152,394,186,3 87,146,241,404,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00015530_147.png 93,33,195,340,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00015583_000.png 339,170,363,188,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00015649_000.png 320,149,363,186,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00015719_005.png 176,189,426,461,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00015770_010.png 181,235,407,370,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00015792_005.png 133,146,173,185,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00015794_000.png 34,324,58,346,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00015799_012.png 180,231,442,430,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016009_008.png 286,288,363,412,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016184_040.png 349,269,371,295,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016191_017.png 341,296,441,378,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016267_000.png 89,299,205,408,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016291_002.png 308,338,471,377,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016291_019.png 76,37,214,219,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016414_000.png 192,215,478,439,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016417_008.png 392,126,483,436,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016429_015.png 73,46,263,331,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016487_002.png 320,346,364,385,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016490_011.png 91,261,175,293,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016522_023.png 68,125,236,447,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016568_010.png 345,200,399,262,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016606_000.png 183,212,427,402,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016837_002.png 70,329,206,376,3 70,352,199,383,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016937_014.png 86,41,218,145,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016964_011.png 364,276,411,306,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016972_019.png 257,317,406,431,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016972_025.png 344,262,404,360,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016987_019.png 132,303,185,342,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016987_022.png 41,244,190,305,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016990_000.png 156,237,385,388,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017028_000.png 72,289,111,319,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017124_004.png 396,284,478,505,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017137_016.png 85,73,228,311,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017151_003.png 312,244,418,343,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017178_007.png 172,150,395,339,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017188_002.png 99,240,214,285,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017199_005.png 386,341,422,378,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017214_015.png 306,174,346,246,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017243_010.png 38,301,82,342,2 10,145,73,428,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017255_001.png 346,299,407,343,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017257_001.png 213,112,372,287,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017346_000.png 436,307,450,324,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017500_002.png 382,307,435,338,3 55,248,154,384,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017514_008.png 173,222,441,390,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017524_028.png 125,217,374,432,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017582_003.png 128,201,228,273,3 53,171,111,326,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017611_002.png 129,224,210,299,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017670_005.png 184,295,244,369,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017714_006.png 406,254,468,418,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017747_008.png 114,87,215,166,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017877_001.png 330,284,430,323,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017893_005.png 193,182,439,386,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017972_026.png 331,123,377,226,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018055_005.png 67,66,238,301,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018055_045.png 44,154,67,204,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018101_012.png 86,92,215,254,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018102_001.png 53,294,73,321,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018187_034.png 240,198,454,368,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018233_057.png 173,192,379,366,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018253_017.png 274,264,355,367,3 67,131,141,397,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018253_059.png 92,202,194,346,3 89,191,190,339,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018366_000.png 95,303,127,338,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018366_029.png 280,235,405,268,1 297,64,399,162,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018387_030.png 194,179,442,326,0 79,220,463,301,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018427_004.png 47,315,135,352,3 385,175,477,449,1 266,164,319,214,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018427_011.png 297,343,449,393,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018496_006.png 40,354,476,464,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018623_001.png 138,249,444,378,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018657_003.png 361,224,446,371,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018686_000.png 176,153,425,380,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018693_004.png 150,217,424,417,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018762_001.png 341,331,382,369,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018762_002.png 301,308,430,349,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018814_000.png 344,257,394,303,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018865_008.png 104,66,177,112,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018980_002.png 123,87,213,197,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018984_000.png 412,154,449,212,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019013_002.png 391,279,416,301,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019058_004.png 175,229,221,299,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019089_004.png 287,312,409,405,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019124_045.png 166,260,229,294,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019124_090.png 328,247,376,284,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019124_104.png 59,69,206,239,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019154_002.png 146,62,212,171,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019187_000.png 198,191,430,384,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019271_030.png 117,348,194,404,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019271_064.png 132,254,429,388,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019271_065.png 105,249,176,305,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019373_036.png 333,203,419,259,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019373_058.png 162,165,224,243,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019399_010.png 391,301,458,425,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019426_000.png 224,220,422,356,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019495_004.png 369,282,443,302,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019499_000.png 361,156,417,204,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019706_002.png 288,44,364,104,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019706_014.png 331,217,417,288,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019766_023.png 367,208,463,366,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019767_016.png 64,318,223,361,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019861_010.png 134,178,402,377,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019892_003.png 285,59,412,109,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019924_020.png 315,314,410,504,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020000_000.png 271,292,382,436,1 135,66,179,80,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020065_008.png 409,158,453,228,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020113_017.png 108,257,448,341,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020124_003.png 87,290,209,341,3 180,142,234,210,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020146_005.png 90,281,239,396,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020213_078.png 92,335,211,385,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020277_001.png 346,214,448,410,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020318_007.png 357,271,467,329,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020318_022.png 275,43,413,163,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020332_000.png 114,311,152,343,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020349_006.png 231,155,371,303,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020393_001.png 118,113,156,148,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020393_003.png 340,270,451,319,3 341,251,467,343,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020405_041.png 55,143,154,332,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020408_037.png 95,316,209,396,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020438_011.png 164,180,407,420,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020482_032.png 241,176,346,357,3 215,212,356,343,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020564_000.png 387,179,444,299,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020673_005.png 281,163,410,248,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020751_003.png 373,77,497,406,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020774_000.png 114,238,145,270,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020810_003.png 79,357,209,414,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020819_002.png 184,170,431,368,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020857_008.png 95,285,376,387,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020986_000.png 209,190,429,336,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021007_000.png 172,234,224,285,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021009_001.png 159,182,389,337,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021024_022.png 359,261,441,356,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021132_000.png 352,102,432,340,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021179_011.png 62,86,239,420,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021181_002.png 75,195,218,298,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021201_010.png 357,95,435,254,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021321_002.png 95,22,197,158,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021374_000.png 373,188,433,241,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021377_016.png 361,336,449,387,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021381_013.png 281,289,422,334,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021420_014.png 393,135,429,174,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021420_020.png 83,345,129,390,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021420_027.png 96,282,149,326,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021443_000.png 143,186,415,393,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021481_014.png 297,200,347,280,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021700_010.png 293,41,394,124,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021703_001.png 68,250,376,356,3 281,255,385,333,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021711_014.png 275,334,451,395,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021748_000.png 275,37,358,64,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021772_016.png 85,96,131,137,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021782_028.png 66,49,177,270,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021796_000.png 93,205,181,295,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021840_016.png 71,268,192,317,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021845_001.png 153,186,347,368,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021860_002.png 50,262,121,367,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021896_003.png 119,68,232,176,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021967_000.png 91,46,273,339,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00022141_023.png 322,48,428,113,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00022155_008.png 342,138,420,177,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00022192_003.png 87,110,218,262,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00022215_012.png 161,176,418,385,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00022237_002.png 134,236,198,287,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00022369_013.png 83,199,170,293,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00022416_004.png 409,234,481,334,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00022416_018.png 178,211,462,388,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00022416_049.png 394,213,436,278,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00022572_063.png 35,258,176,418,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00022572_087.png 10,122,117,353,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00022611_001.png 374,84,459,272,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00022706_001.png 173,157,398,351,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00022707_003.png 377,284,463,320,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00022883_002.png 57,255,423,309,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00022977_000.png 120,321,153,353,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00023058_004.png 19,281,177,382,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00023068_003.png 175,373,193,390,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00023075_033.png 119,267,156,300,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00023078_000.png 370,343,396,367,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00023078_003.png 47,330,112,387,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00023093_007.png 345,206,458,377,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00023093_009.png 216,176,410,312,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00023156_001.png 114,115,192,173,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00023162_025.png 96,39,228,168,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00023168_000.png 107,339,154,400,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00023176_010.png 140,252,478,331,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00023176_017.png 161,127,428,340,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00023325_019.png 172,140,410,315,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025221_001.png 269,291,441,451,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025228_005.png 295,229,443,349,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025228_007.png 59,235,130,386,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025252_032.png 378,340,478,462,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025252_040.png 300,101,431,310,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025252_053.png 7,320,175,400,1 325,319,397,451,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025252_054.png 329,360,401,472,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025270_000.png 134,260,184,294,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025368_014.png 147,342,205,384,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025368_018.png 75,61,235,197,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025368_033.png 379,139,478,236,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025521_003.png 36,260,89,390,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025529_018.png 362,83,454,385,1 351,194,406,258,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025662_006.png 413,319,439,348,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025664_002.png 33,72,171,387,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025686_000.png 298,335,464,387,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025732_004.png 109,195,378,403,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025747_000.png 167,185,415,369,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025769_001.png 350,286,402,318,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025787_027.png 302,49,490,418,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025787_039.png 43,49,134,334,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025787_050.png 305,90,333,215,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025954_025.png 98,46,235,171,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026087_000.png 169,231,378,428,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026136_002.png 275,277,414,332,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026196_001.png 196,245,238,285,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026221_001.png 99,364,151,422,2 68,75,228,328,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026285_000.png 452,412,476,436,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026319_000.png 334,326,394,365,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026338_003.png 176,172,395,330,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026392_005.png 281,55,440,183,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026398_000.png 182,132,206,156,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026451_068.png 147,321,209,368,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026586_009.png 98,93,183,158,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026695_000.png 170,98,229,164,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026753_008.png 299,53,450,393,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026769_010.png 396,208,423,235,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026810_001.png 383,354,442,381,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026886_002.png 112,98,229,437,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026886_004.png 117,258,224,326,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026889_000.png 118,185,415,412,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026911_005.png 173,153,216,201,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026920_000.png 171,223,232,249,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026983_001.png 276,95,310,131,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027028_017.png 75,260,461,330,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027094_003.png 384,280,461,356,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027103_001.png 336,118,438,253,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027213_044.png 285,83,389,489,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027278_007.png 76,23,232,162,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027357_014.png 31,52,162,375,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027441_002.png 60,67,169,348,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027470_006.png 199,250,247,287,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027474_005.png 138,203,188,235,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027577_003.png 47,105,115,238,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027631_000.png 418,356,473,427,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027652_003.png 37,101,110,358,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027685_003.png 170,161,404,352,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027697_001.png 354,192,451,221,3 79,244,234,350,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027797_000.png 215,204,462,365,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027833_022.png 150,152,445,373,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027866_002.png 72,308,373,416,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027875_005.png 348,156,481,357,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027927_009.png 369,195,395,226,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027937_004.png 83,263,138,314,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028012_001.png 280,231,443,331,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028018_000.png 176,226,404,440,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028027_000.png 287,123,346,195,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028173_016.png 282,250,338,319,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028330_003.png 104,219,177,317,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028383_002.png 389,385,433,414,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028452_001.png 314,205,442,279,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028509_007.png 24,169,128,451,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028509_026.png 155,200,408,371,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028518_021.png 92,339,185,392,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028607_000.png 202,151,450,372,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028620_000.png 399,334,476,376,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028625_000.png 169,225,229,273,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028628_008.png 303,49,409,210,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028628_015.png 84,76,170,281,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028628_020.png 313,169,365,211,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028640_008.png 381,271,448,405,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028698_001.png 425,202,455,286,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028861_009.png 290,23,436,173,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028873_009.png 166,140,364,319,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028876_027.png 268,56,384,128,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028876_035.png 95,170,127,206,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028924_009.png 365,112,427,242,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028974_016.png 384,290,467,389,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029075_013.png 328,67,453,182,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029088_023.png 152,147,225,208,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029105_015.png 114,214,174,275,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029200_006.png 66,254,144,311,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029259_027.png 393,193,450,339,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029391_000.png 157,185,400,373,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029464_006.png 45,142,131,239,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029464_015.png 99,176,407,338,3 359,305,455,331,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029469_011.png 362,215,448,441,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029502_006.png 110,294,236,384,3 74,226,202,423,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029532_005.png 64,49,206,280,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029579_014.png 50,260,69,322,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029588_004.png 20,99,153,381,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029631_006.png 59,254,145,307,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029647_002.png 167,203,415,396,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029807_003.png 82,41,260,192,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029817_009.png 110,158,188,267,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029843_001.png 128,224,193,279,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029894_000.png 357,254,472,363,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029906_000.png 165,224,381,382,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029909_003.png 83,84,200,129,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029940_007.png 106,191,178,282,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00030039_008.png 13,60,145,373,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00030106_008.png 112,258,227,393,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00030111_007.png 296,48,375,174,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00030162_029.png 163,190,197,227,2 310,83,412,163,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00030206_013.png 156,169,389,386,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00030260_004.png 184,275,212,386,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00030323_015.png 307,211,420,357,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00030394_001.png 311,91,344,120,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00030408_013.png 359,302,418,374,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00030434_000.png 333,322,371,442,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00030606_006.png 275,15,393,96,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00030636_004.png 305,220,385,295,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00030674_000.png 316,125,430,231,3

72
model_data/2007_val.txt Normal file
View File

@@ -0,0 +1,72 @@
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000005_001.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000005_003.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000013_038.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000032_008.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00000042_005.png
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00001320_003.png 104,272,140,287,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00001437_012.png 41,69,218,377,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00002176_005.png 130,44,223,122,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00002395_007.png 115,224,153,303,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00002435_005.png 181,240,382,389,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00002578_000.png 398,163,437,193,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00002856_009.png 108,318,217,369,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00004342_020.png 158,193,395,409,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00006621_004.png 153,200,261,308,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00007034_016.png 383,288,428,376,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00009368_006.png 184,140,218,194,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00010071_008.png 101,340,137,431,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00011291_003.png 346,73,441,175,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00011322_002.png 173,178,426,420,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00011355_011.png 250,26,464,253,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012045_019.png 211,246,239,274,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012094_047.png 333,71,400,147,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012299_002.png 97,287,427,327,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00012829_004.png 84,269,168,316,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013310_059.png 96,104,190,388,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013673_001.png 75,217,115,253,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013685_000.png 129,218,185,243,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00013911_021.png 49,121,399,390,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014198_000.png 338,256,387,352,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014626_028.png 166,177,385,377,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00014976_003.png 303,340,424,380,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00015090_006.png 319,285,417,374,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00015425_012.png 192,144,415,322,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00015895_017.png 359,177,438,438,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016403_003.png 446,438,501,471,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016564_000.png 192,195,418,348,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016587_069.png 377,203,412,232,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00016624_000.png 276,173,460,339,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017138_037.png 59,143,154,384,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017403_006.png 297,288,393,440,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017448_000.png 177,241,416,437,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00017511_006.png 169,211,368,378,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018187_029.png 171,210,427,441,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018253_054.png 91,154,214,407,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018412_001.png 106,308,392,353,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00018496_007.png 286,62,397,109,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019313_000.png 374,307,423,361,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00019917_004.png 306,388,430,412,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020274_021.png 79,218,130,253,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020408_058.png 80,36,224,232,4
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00020429_020.png 83,295,174,338,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021364_001.png 167,225,415,378,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021409_001.png 156,172,402,365,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00021862_004.png 303,279,363,347,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00022098_006.png 247,288,382,365,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00022726_002.png 344,200,407,260,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00023116_005.png 320,219,410,298,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00025969_000.png 182,142,401,332,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026132_016.png 350,241,390,278,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00026261_001.png 98,140,217,260,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027479_013.png 386,282,479,317,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027556_007.png 86,322,155,380,3 35,98,120,216,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00027817_001.png 380,195,478,378,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028265_007.png 399,303,442,353,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028873_004.png 160,160,380,328,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00028876_060.png 381,194,419,237,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029464_003.png 121,187,224,251,3
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00029808_003.png 160,205,384,402,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00030279_000.png 144,210,418,438,0
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00030413_003.png 127,118,172,157,2
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00030634_000.png 70,305,229,354,1
E:\project\python\rabat-illness-yolov5\database\Train/JPEGImages/00030635_001.png 288,245,408,356,3

5
model_data/coco_classes.txt
View File

@@ -1,2 +1,5 @@
0
1
2
3
4
5

5
model_data/voc_classes.txt
View File

@@ -1,2 +1,5 @@
0
1
2
3
4
5

192
predict.py Normal file
View File

@@ -0,0 +1,192 @@
#-----------------------------------------------------------------------#
# predict.py将单张图片预测、摄像头检测、FPS测试和目录遍历检测等功能
# 整合到了一个py文件中，通过指定mode进行模式的修改。
#-----------------------------------------------------------------------#
import time
import cv2
import numpy as np
from PIL import Image
from network.yolo import YOLO, YOLO_ONNX
if __name__ == "__main__":
#----------------------------------------------------------------------------------------------------------#
# mode用于指定测试的模式：
# 'predict' 表示单张图片预测，如果想对预测过程进行修改，如保存图片，截取对象等，可以先看下方详细的注释
# 'video' 表示视频检测，可调用摄像头或者视频进行检测，详情查看下方注释。
# 'fps' 表示测试fps，使用的图片是img里面的street.jpg，详情查看下方注释。
# 'dir_predict' 表示遍历文件夹进行检测并保存。默认遍历img文件夹，保存img_out文件夹，详情查看下方注释。
# 'heatmap' 表示进行预测结果的热力图可视化，详情查看下方注释。
# 'export_onnx' 表示将模型导出为onnx，需要pytorch1.7.1以上。
# 'predict_onnx' 表示利用导出的onnx模型进行预测，相关参数的修改在yolo.py_423行左右处的YOLO_ONNX
#----------------------------------------------------------------------------------------------------------#
mode = "predict"
#-------------------------------------------------------------------------#
# crop 指定了是否在单张图片预测后对目标进行截取
# count 指定了是否进行目标的计数
# crop、count仅在mode='predict'时有效
#-------------------------------------------------------------------------#
crop = False
count = False
#----------------------------------------------------------------------------------------------------------#
# video_path 用于指定视频的路径，当video_path=0时表示检测摄像头
# 想要检测视频，则设置如video_path = "xxx.mp4"即可，代表读取出根目录下的xxx.mp4文件。
# video_save_path 表示视频保存的路径，当video_save_path=""时表示不保存
# 想要保存视频，则设置如video_save_path = "yyy.mp4"即可，代表保存为根目录下的yyy.mp4文件。
# video_fps 用于保存的视频的fps
#
# video_path、video_save_path和video_fps仅在mode='video'时有效
# 保存视频时需要ctrl+c退出或者运行到最后一帧才会完成完整的保存步骤。
#----------------------------------------------------------------------------------------------------------#
video_path = 0
video_save_path = ""
video_fps = 25.0
#----------------------------------------------------------------------------------------------------------#
# test_interval 用于指定测量fps的时候，图片检测的次数。理论上test_interval越大，fps越准确。
# fps_image_path 用于指定测试的fps图片
#
# test_interval和fps_image_path仅在mode='fps'有效
#----------------------------------------------------------------------------------------------------------#
test_interval = 100
fps_image_path = "img/street.jpg"
#-------------------------------------------------------------------------#
# dir_origin_path 指定了用于检测的图片的文件夹路径
# dir_save_path 指定了检测完图片的保存路径
#
# dir_origin_path和dir_save_path仅在mode='dir_predict'时有效
#-------------------------------------------------------------------------#
dir_origin_path = "application\logs\logData\img"
dir_save_path = "application\logs\logData\save"
#-------------------------------------------------------------------------#
# heatmap_save_path 热力图的保存路径，默认保存在model_data下
#
# heatmap_save_path仅在mode='heatmap'有效
#-------------------------------------------------------------------------#
heatmap_save_path = "model_data/heatmap_vision.png"
#-------------------------------------------------------------------------#
# simplify 使用Simplify onnx
# onnx_save_path 指定了onnx的保存路径
#-------------------------------------------------------------------------#
simplify = True
onnx_save_path = "model_data/models.onnx"
if mode != "predict_onnx":
yolo = YOLO()
else:
yolo = YOLO_ONNX()
if mode == "predict":
'''
1、如果想要进行检测完的图片的保存，利用r_image.save("img.jpg")即可保存，直接在predict.py里进行修改即可。
2、如果想要获得预测框的坐标，可以进入yolo.detect_image函数，在绘图部分读取top，left，bottom，right这四个值。
3、如果想要利用预测框截取下目标，可以进入yolo.detect_image函数，在绘图部分利用获取到的top，left，bottom，right这四个值
在原图上利用矩阵的方式进行截取。
4、如果想要在预测图上写额外的字，比如检测到的特定目标的数量，可以进入yolo.detect_image函数，在绘图部分对predicted_class进行判断，
比如判断if predicted_class == 'car': 即可判断当前目标是否为车，然后记录数量即可。利用draw.text即可写字。
'''
while True:
img = input('Input image filename:')
try:
image = Image.open(img)
except:
print('Open Error! Try again!')
continue
else:
r_image = yolo.detect_image(image, crop = crop, count=count)
r_image.show()
elif mode == "video":
capture = cv2.VideoCapture(video_path)
if video_save_path!="":
fourcc = cv2.VideoWriter_fourcc(*'XVID')
size = (int(capture.get(cv2.CAP_PROP_FRAME_WIDTH)), int(capture.get(cv2.CAP_PROP_FRAME_HEIGHT)))
out = cv2.VideoWriter(video_save_path, fourcc, video_fps, size)
ref, frame = capture.read()
if not ref:
raise ValueError("未能正确读取摄像头（视频），请注意是否正确安装摄像头（是否正确填写视频路径）。")
fps = 0.0
while(True):
t1 = time.time()
# 读取某一帧
ref, frame = capture.read()
if not ref:
break
# 格式转变，BGRtoRGB
frame = cv2.cvtColor(frame,cv2.COLOR_BGR2RGB)
# 转变成Image
frame = Image.fromarray(np.uint8(frame))
# 进行检测
frame = np.array(yolo.detect_image(frame))
# RGBtoBGR满足opencv显示格式
frame = cv2.cvtColor(frame,cv2.COLOR_RGB2BGR)
fps = ( fps + (1./(time.time()-t1)) ) / 2
print("fps= %.2f"%(fps))
frame = cv2.putText(frame, "fps= %.2f"%(fps), (0, 40), cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 255, 0), 2)
cv2.imshow("video",frame)
c= cv2.waitKey(1) & 0xff
if video_save_path!="":
out.write(frame)
if c==27:
capture.release()
break
print("Video Detection Done!")
capture.release()
if video_save_path!="":
print("Save processed video to the path :" + video_save_path)
out.release()
cv2.destroyAllWindows()
elif mode == "fps":
img = Image.open(fps_image_path)
tact_time = yolo.get_FPS(img, test_interval)
print(str(tact_time) + ' seconds, ' + str(1/tact_time) + 'FPS, @batch_size 1')
elif mode == "dir_predict":
import os
from tqdm import tqdm
img_names = os.listdir(dir_origin_path)
for img_name in tqdm(img_names):
if img_name.lower().endswith(('.bmp', '.dib', '.png', '.jpg', '.jpeg', '.pbm', '.pgm', '.ppm', '.tif', '.tiff')):
image_path = os.path.join(dir_origin_path, img_name)
image = Image.open(image_path)
r_image = yolo.detect_image(image)
if not os.path.exists(dir_save_path):
os.makedirs(dir_save_path)
r_image.save(os.path.join(dir_save_path, img_name.replace(".jpg", ".png")), quality=95, subsampling=0)
elif mode == "heatmap":
while True:
img = input('Input image filename:')
try:
image = Image.open(img)
except:
print('Open Error! Try again!')
continue
else:
yolo.detect_heatmap(image, heatmap_save_path)
elif mode == "export_onnx":
yolo.convert_to_onnx(simplify, onnx_save_path)
elif mode == "predict_onnx":
while True:
img = input('Input image filename:')
try:
image = Image.open(img)
except:
print('Open Error! Try again!')
continue
else:
r_image = yolo.detect_image(image)
r_image.show()
else:
raise AssertionError("Please specify the correct mode: 'predict', 'video', 'fps', 'heatmap', 'export_onnx', 'dir_predict'.")

0
test.py
View File

11
train.py
View File

@@ -181,8 +181,8 @@ if __name__ == "__main__":
# Adam可以使用相对较小的UnFreeze_Epoch
# Unfreeze_batch_size 模型在解冻后的batch_size
#------------------------------------------------------------------#
UnFreeze_Epoch = 500
Unfreeze_batch_size = 6
UnFreeze_Epoch = 100
Unfreeze_batch_size = 4
#------------------------------------------------------------------#
# Freeze_Train 是否进行冻结训练
# 默认先冻结主干训练后解冻训练。
@@ -237,14 +237,14 @@ if __name__ == "__main__":
# 开启后会加快数据读取速度，但是会占用更多内存
# 内存较小的电脑可以设置为2或者0
#------------------------------------------------------------------#
num_workers = 20
num_workers = 6
#------------------------------------------------------#
# train_annotation_path 训练图片路径和标签
# val_annotation_path 验证图片路径和标签
#------------------------------------------------------#
train_annotation_path = '2007_train.txt'
val_annotation_path = '2007_val.txt'
train_annotation_path = 'model_data/2007_train.txt'
val_annotation_path = 'model_data/2007_val.txt'
seed_everything(seed)
#------------------------------------------------------#
@@ -261,6 +261,7 @@ if __name__ == "__main__":
print("Gpu Device Count : ", ngpus_per_node)
else:
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
print("\033[1;33;44mRuning on {}\033[0m".format(device))
local_rank = 0
rank = 0

138
utils/get_map.py Normal file
View File

@@ -0,0 +1,138 @@
import os
import xml.etree.ElementTree as ET
from PIL import Image
from tqdm import tqdm
from utils.utils import get_classes
from utils.utils_map import get_coco_map, get_map
from yolo import YOLO
if __name__ == "__main__":
'''
Recall和Precision不像AP是一个面积的概念，因此在门限值（Confidence）不同时，网络的Recall和Precision值是不同的。
默认情况下，本代码计算的Recall和Precision代表的是当门限值（Confidence）为0.5时，所对应的Recall和Precision值。
受到mAP计算原理的限制，网络在计算mAP时需要获得近乎所有的预测框，这样才可以计算不同门限条件下的Recall和Precision值
因此，本代码获得的map_out/detection-results/里面的txt的框的数量一般会比直接predict多一些，目的是列出所有可能的预测框，
'''
#------------------------------------------------------------------------------------------------------------------#
# map_mode用于指定该文件运行时计算的内容
# map_mode为0代表整个map计算流程，包括获得预测结果、获得真实框、计算VOC_map。
# map_mode为1代表仅仅获得预测结果。
# map_mode为2代表仅仅获得真实框。
# map_mode为3代表仅仅计算VOC_map。
# map_mode为4代表利用COCO工具箱计算当前数据集的0.50:0.95map。需要获得预测结果、获得真实框后并安装pycocotools才行
#-------------------------------------------------------------------------------------------------------------------#
map_mode = 0
#--------------------------------------------------------------------------------------#
# 此处的classes_path用于指定需要测量VOC_map的类别
# 一般情况下与训练和预测所用的classes_path一致即可
#--------------------------------------------------------------------------------------#
classes_path = 'model_data/voc_classes.txt'
#--------------------------------------------------------------------------------------#
# MINOVERLAP用于指定想要获得的mAP0.x，mAP0.x的意义是什么请同学们百度一下。
# 比如计算mAP0.75，可以设定MINOVERLAP = 0.75。
#
# 当某一预测框与真实框重合度大于MINOVERLAP时，该预测框被认为是正样本，否则为负样本。
# 因此MINOVERLAP的值越大，预测框要预测的越准确才能被认为是正样本，此时算出来的mAP值越低，
#--------------------------------------------------------------------------------------#
MINOVERLAP = 0.5
#--------------------------------------------------------------------------------------#
# 受到mAP计算原理的限制，网络在计算mAP时需要获得近乎所有的预测框，这样才可以计算mAP
# 因此，confidence的值应当设置的尽量小进而获得全部可能的预测框。
#
# 该值一般不调整。因为计算mAP需要获得近乎所有的预测框，此处的confidence不能随便更改。
# 想要获得不同门限值下的Recall和Precision值，请修改下方的score_threhold。
#--------------------------------------------------------------------------------------#
confidence = 0.001
#--------------------------------------------------------------------------------------#
# 预测时使用到的非极大抑制值的大小，越大表示非极大抑制越不严格。
#
# 该值一般不调整。
#--------------------------------------------------------------------------------------#
nms_iou = 0.5
#---------------------------------------------------------------------------------------------------------------#
# Recall和Precision不像AP是一个面积的概念，因此在门限值不同时，网络的Recall和Precision值是不同的。
#
# 默认情况下，本代码计算的Recall和Precision代表的是当门限值为0.5（此处定义为score_threhold）时所对应的Recall和Precision值。
# 因为计算mAP需要获得近乎所有的预测框，上面定义的confidence不能随便更改。
# 这里专门定义一个score_threhold用于代表门限值，进而在计算mAP时找到门限值对应的Recall和Precision值。
#---------------------------------------------------------------------------------------------------------------#
score_threhold = 0.5
#-------------------------------------------------------#
# map_vis用于指定是否开启VOC_map计算的可视化
#-------------------------------------------------------#
map_vis = False
#-------------------------------------------------------#
# 指向VOC数据集所在的文件夹
# 默认指向根目录下的VOC数据集
#-------------------------------------------------------#
VOCdevkit_path = 'VOCdevkit'
#-------------------------------------------------------#
# 结果输出的文件夹，默认为map_out
#-------------------------------------------------------#
map_out_path = 'map_out'
image_ids = open(os.path.join(VOCdevkit_path, "VOC2007/ImageSets/Main/test.txt")).read().strip().split()
if not os.path.exists(map_out_path):
os.makedirs(map_out_path)
if not os.path.exists(os.path.join(map_out_path, 'ground-truth')):
os.makedirs(os.path.join(map_out_path, 'ground-truth'))
if not os.path.exists(os.path.join(map_out_path, 'detection-results')):
os.makedirs(os.path.join(map_out_path, 'detection-results'))
if not os.path.exists(os.path.join(map_out_path, 'images-optional')):
os.makedirs(os.path.join(map_out_path, 'images-optional'))
class_names, _ = get_classes(classes_path)
if map_mode == 0 or map_mode == 1:
print("Load model.")
yolo = YOLO(confidence = confidence, nms_iou = nms_iou)
print("Load model done.")
print("Get predict result.")
for image_id in tqdm(image_ids):
image_path = os.path.join(VOCdevkit_path, "VOC2007/JPEGImages/"+image_id+".jpg")
image = Image.open(image_path)
if map_vis:
image.save(os.path.join(map_out_path, "images-optional/" + image_id + ".jpg"))
yolo.get_map_txt(image_id, image, class_names, map_out_path)
print("Get predict result done.")
if map_mode == 0 or map_mode == 2:
print("Get ground truth result.")
for image_id in tqdm(image_ids):
with open(os.path.join(map_out_path, "ground-truth/"+image_id+".txt"), "w") as new_f:
root = ET.parse(os.path.join(VOCdevkit_path, "VOC2007/Annotations/"+image_id+".xml")).getroot()
for obj in root.findall('object'):
difficult_flag = False
if obj.find('difficult')!=None:
difficult = obj.find('difficult').text
if int(difficult)==1:
difficult_flag = True
obj_name = obj.find('name').text
if obj_name not in class_names:
continue
bndbox = obj.find('bndbox')
left = bndbox.find('xmin').text
top = bndbox.find('ymin').text
right = bndbox.find('xmax').text
bottom = bndbox.find('ymax').text
if difficult_flag:
new_f.write("%s %s %s %s %s difficult\n" % (obj_name, left, top, right, bottom))
else:
new_f.write("%s %s %s %s %s\n" % (obj_name, left, top, right, bottom))
print("Get ground truth result done.")
if map_mode == 0 or map_mode == 3:
print("Get map.")
get_map(MINOVERLAP, True, score_threhold = score_threhold, path = map_out_path)
print("Get map done.")
if map_mode == 4:
print("Get map.")
get_coco_map(class_names = class_names, path = map_out_path)
print("Get map done.")

32
utils/summary.py Normal file
View File

@@ -0,0 +1,32 @@
#--------------------------------------------#
# 该部分代码用于看网络结构
#--------------------------------------------#
import torch
from thop import clever_format, profile
from torchsummary import summary
from nets.yolo import YoloBody
if __name__ == "__main__":
input_shape = [640, 640]
anchors_mask = [[6, 7, 8], [3, 4, 5], [0, 1, 2]]
num_classes = 80
backbone = 'cspdarknet'
phi = 'l'
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
m = YoloBody(anchors_mask, num_classes, phi, backbone=backbone).to(device)
summary(m, (3, input_shape[0], input_shape[1]))
dummy_input = torch.randn(1, 3, input_shape[0], input_shape[1]).to(device)
flops, params = profile(m.to(device), (dummy_input, ), verbose=False)
#--------------------------------------------------------#
# flops * 2是因为profile没有将卷积作为两个operations
# 有些论文将卷积算乘法、加法两个operations。此时乘2
# 有些论文只考虑乘法的运算次数，忽略加法。此时不乘2
# 本代码选择乘2，参考YOLOX。
#--------------------------------------------------------#
flops = flops * 2
flops, params = clever_format([flops, params], "%.3f")
print('Total GFLOPS: %s' % (flops))
print('Total params: %s' % (params))

16
voc_annotation.py → utils/voc_annotation.py
View File

@@ -3,9 +3,13 @@ import random
import xml.etree.ElementTree as ET
import numpy as np
import sys,os
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
from utils.utils import get_classes
from utils import get_classes
import configparser
conf=configparser.ConfigParser()
conf.read('config.ini',encoding='utf-8')
#--------------------------------------------------------------------------------------------------------------------------------#
# annotation_mode用于指定该文件运行时计算的内容
# annotation_mode为0代表整个标签处理过程，包括获得VOCdevkit/VOC2007/ImageSets里面的txt以及训练用的2007_train.txt、2007_val.txt
@@ -20,7 +24,7 @@ annotation_mode = 0
# 那么就是因为classes没有设定正确
# 仅在annotation_mode为0和2的时候有效
#-------------------------------------------------------------------#
classes_path = 'trainYolov5-v6/model_data/voc_classes.txt'
classes_path = conf.get('dataset', 'classes_path')
#--------------------------------------------------------------------------------------------------------------------------------#
# trainval_percent用于指定(训练集+验证集)与测试集的比例，默认情况下 (训练集+验证集):测试集 = 9:1
# train_percent用于指定(训练集+验证集)中训练集与验证集的比例，默认情况下 训练集:验证集 = 9:1
@@ -32,7 +36,7 @@ train_percent = 0.9
# 指向VOC数据集所在的文件夹
# 默认指向根目录下的VOC数据集
#-------------------------------------------------------#
VOCdevkit_path = 'Data/TrainData'
VOCdevkit_path = r'database/Train'
VOCdevkit_sets = [('2007', 'train'), ('2007', 'val')]
classes, _ = get_classes(classes_path)
@@ -112,9 +116,9 @@ if __name__ == "__main__":
type_index = 0
for year, image_set in VOCdevkit_sets:
image_ids = open(os.path.join(VOCdevkit_path, 'ImageSets/Main/%s.txt'%(image_set)), encoding='utf-8').read().strip().split()
list_file = open('%s_%s.txt'%(year, image_set), 'w', encoding='utf-8')
list_file = open('model_data/%s_%s.txt'%(year, image_set), 'w', encoding='utf-8')
for image_id in image_ids:
list_file.write('%s/JPEGImages/%s.jpg'%(os.path.abspath(VOCdevkit_path), image_id))
list_file.write('%s/JPEGImages/%s.png'%(os.path.abspath(VOCdevkit_path), image_id))
convert_annotation(year, image_id, list_file)
list_file.write('\n')

663
utils/yolo.py Normal file
View File

@@ -0,0 +1,663 @@
import colorsys
import os
import time
import cv2
import numpy as np
import torch
import torch.nn as nn
from PIL import ImageDraw, ImageFont, Image
from nets.yolo import YoloBody
from utils.utils import (cvtColor, get_anchors, get_classes, preprocess_input,
resize_image, show_config)
from utils.utils_bbox import DecodeBox, DecodeBoxNP
'''
训练自己的数据集必看注释！
'''
class YOLO(object):
_defaults = {
#--------------------------------------------------------------------------#
# 使用自己训练好的模型进行预测一定要修改model_path和classes_path！
# model_path指向logs文件夹下的权值文件，classes_path指向model_data下的txt
#
# 训练好后logs文件夹下存在多个权值文件，选择验证集损失较低的即可。
# 验证集损失较低不代表mAP较高，仅代表该权值在验证集上泛化性能较好。
# 如果出现shape不匹配，同时要注意训练时的model_path和classes_path参数的修改
#--------------------------------------------------------------------------#
"model_path" : r'logs-yolov5\1.pth',
"classes_path" : 'trainYolov5-v6\\model_data/coco_classes.txt',
#---------------------------------------------------------------------#
# anchors_path代表先验框对应的txt文件，一般不修改。
# anchors_mask用于帮助代码找到对应的先验框，一般不修改。
#---------------------------------------------------------------------#
"anchors_path" : 'trainYolov5-v6\\model_data/yolo_anchors.txt',
"anchors_mask" : [[6, 7, 8], [3, 4, 5], [0, 1, 2]],
#---------------------------------------------------------------------#
# 输入图片的大小，必须为32的倍数。
#---------------------------------------------------------------------#
"input_shape" : [640, 640],
#------------------------------------------------------#
# backbone cspdarknet（默认）
# convnext_tiny
# convnext_small
# swin_transfomer_tiny
#------------------------------------------------------#
"backbone" : 'cspdarknet',
#------------------------------------------------------#
# 所使用的YoloV5的版本。s、m、l、x
# 在除cspdarknet的其它主干中仅影响panet的大小
#------------------------------------------------------#
"phi" : 's',
#---------------------------------------------------------------------#
# 只有得分大于置信度的预测框会被保留下来
#---------------------------------------------------------------------#
"confidence" : 0.5,
#---------------------------------------------------------------------#
# 非极大抑制所用到的nms_iou大小
#---------------------------------------------------------------------#
"nms_iou" : 0.3,
#---------------------------------------------------------------------#
# 该变量用于控制是否使用letterbox_image对输入图像进行不失真的resize，
# 在多次测试后，发现关闭letterbox_image直接resize的效果更好
#---------------------------------------------------------------------#
"letterbox_image" : True,
#-------------------------------#
# 是否使用Cuda
# 没有GPU可以设置成False
#-------------------------------#
"cuda" : True,
}
@classmethod
def get_defaults(cls, n):
if n in cls._defaults:
return cls._defaults[n]
else:
return "Unrecognized attribute name '" + n + "'"
#---------------------------------------------------#
# 初始化YOLO
#---------------------------------------------------#
def __init__(self, **kwargs):
self.__dict__.update(self._defaults)
for name, value in kwargs.items():
setattr(self, name, value)
self._defaults[name] = value
#---------------------------------------------------#
# 获得种类和先验框的数量
#---------------------------------------------------#
self.class_names, self.num_classes = get_classes(self.classes_path)
self.anchors, self.num_anchors = get_anchors(self.anchors_path)
self.bbox_util = DecodeBox(self.anchors, self.num_classes, (self.input_shape[0], self.input_shape[1]), self.anchors_mask)
#---------------------------------------------------#
# 画框设置不同的颜色
#---------------------------------------------------#
hsv_tuples = [(x / self.num_classes, 1., 1.) for x in range(self.num_classes)]
self.colors = list(map(lambda x: colorsys.hsv_to_rgb(*x), hsv_tuples))
self.colors = list(map(lambda x: (int(x[0] * 255), int(x[1] * 255), int(x[2] * 255)), self.colors))
self.generate()
show_config(**self._defaults)
#---------------------------------------------------#
# 生成模型
#---------------------------------------------------#
def generate(self, onnx=False):
#---------------------------------------------------#
# 建立yolo模型，载入yolo模型的权重
#---------------------------------------------------#
self.net = YoloBody(self.anchors_mask, self.num_classes, self.phi, backbone = self.backbone, input_shape = self.input_shape)
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
self.net.load_state_dict(torch.load(self.model_path, map_location=device),strict=False)
self.net = self.net.eval()
print('{} model, and classes loaded.'.format(self.model_path))
if not onnx:
if self.cuda:
self.net = nn.DataParallel(self.net)
self.net = self.net.cuda()
#---------------------------------------------------#
# 检测图片
#---------------------------------------------------#
def detect_image(self, image, crop = False, count = False):
#---------------------------------------------------#
# 计算输入图片的高和宽
#---------------------------------------------------#
image_shape = np.array(np.shape(image)[0:2])
#---------------------------------------------------------#
# 在这里将图像转换成RGB图像，防止灰度图在预测时报错。
# 代码仅仅支持RGB图像的预测，所有其它类型的图像都会转化成RGB
#---------------------------------------------------------#
image = cvtColor(image)
#---------------------------------------------------------#
# 给图像增加灰条，实现不失真的resize
# 也可以直接resize进行识别
#---------------------------------------------------------#
image_data = resize_image(image, (self.input_shape[1], self.input_shape[0]), self.letterbox_image)
#---------------------------------------------------------#
# 添加上batch_size维度
#---------------------------------------------------------#
image_data = np.expand_dims(np.transpose(preprocess_input(np.array(image_data, dtype='float32')), (2, 0, 1)), 0)
with torch.no_grad():
images = torch.from_numpy(image_data)
if self.cuda:
images = images.cuda()
#---------------------------------------------------------#
# 将图像输入网络当中进行预测！
#---------------------------------------------------------#
outputs = self.net(images)
outputs = self.bbox_util.decode_box(outputs)
#---------------------------------------------------------#
# 将预测框进行堆叠，然后进行非极大抑制
#---------------------------------------------------------#
results = self.bbox_util.non_max_suppression(torch.cat(outputs, 1), self.num_classes, self.input_shape,
image_shape, self.letterbox_image, conf_thres = self.confidence, nms_thres = self.nms_iou)
if results[0] is None:
return image
top_label = np.array(results[0][:, 6], dtype = 'int32')
top_conf = results[0][:, 4] * results[0][:, 5]
top_boxes = results[0][:, :4]
#---------------------------------------------------------#
# 设置字体与边框厚度
#---------------------------------------------------------#
font = ImageFont.truetype(font='model_data/simhei.ttf', size=np.floor(3e-2 * image.size[1] + 0.5).astype('int32'))
thickness = int(max((image.size[0] + image.size[1]) // np.mean(self.input_shape), 1))
#---------------------------------------------------------#
# 计数
#---------------------------------------------------------#
if count:
print("top_label:", top_label)
classes_nums = np.zeros([self.num_classes])
for i in range(self.num_classes):
num = np.sum(top_label == i)
if num > 0:
print(self.class_names[i], " : ", num)
classes_nums[i] = num
print("classes_nums:", classes_nums)
#---------------------------------------------------------#
# 是否进行目标的裁剪
#---------------------------------------------------------#
if crop:
for i, c in list(enumerate(top_boxes)):
top, left, bottom, right = top_boxes[i]
top = max(0, np.floor(top).astype('int32'))
left = max(0, np.floor(left).astype('int32'))
bottom = min(image.size[1], np.floor(bottom).astype('int32'))
right = min(image.size[0], np.floor(right).astype('int32'))
dir_save_path = "img_crop"
if not os.path.exists(dir_save_path):
os.makedirs(dir_save_path)
crop_image = image.crop([left, top, right, bottom])
crop_image.save(os.path.join(dir_save_path, "crop_" + str(i) + ".png"), quality=95, subsampling=0)
print("save crop_" + str(i) + ".png to " + dir_save_path)
#---------------------------------------------------------#
# 图像绘制
#---------------------------------------------------------#
for i, c in list(enumerate(top_label)):
predicted_class = self.class_names[int(c)]
box = top_boxes[i]
score = top_conf[i]
top, left, bottom, right = box
top = max(0, np.floor(top).astype('int32'))
left = max(0, np.floor(left).astype('int32'))
bottom = min(image.size[1], np.floor(bottom).astype('int32'))
right = min(image.size[0], np.floor(right).astype('int32'))
label = '{} {:.2f}'.format(predicted_class, score)
draw = ImageDraw.Draw(image)
label_size = draw.textsize(label, font)
label = label.encode('utf-8')
print(label, top, left, bottom, right)
if top - label_size[1] >= 0:
text_origin = np.array([left, top - label_size[1]])
else:
text_origin = np.array([left, top + 1])
for i in range(thickness):
draw.rectangle([left + i, top + i, right - i, bottom - i], outline=self.colors[c])
draw.rectangle([tuple(text_origin), tuple(text_origin + label_size)], fill=self.colors[c])
draw.text(text_origin, str(label,'UTF-8'), fill=(0, 0, 0), font=font)
del draw
return image
def get_FPS(self, image, test_interval):
image_shape = np.array(np.shape(image)[0:2])
#---------------------------------------------------------#
# 在这里将图像转换成RGB图像，防止灰度图在预测时报错。
# 代码仅仅支持RGB图像的预测，所有其它类型的图像都会转化成RGB
#---------------------------------------------------------#
image = cvtColor(image)
#---------------------------------------------------------#
# 给图像增加灰条，实现不失真的resize
# 也可以直接resize进行识别
#---------------------------------------------------------#
image_data = resize_image(image, (self.input_shape[1], self.input_shape[0]), self.letterbox_image)
#---------------------------------------------------------#
# 添加上batch_size维度
#---------------------------------------------------------#
image_data = np.expand_dims(np.transpose(preprocess_input(np.array(image_data, dtype='float32')), (2, 0, 1)), 0)
with torch.no_grad():
images = torch.from_numpy(image_data)
if self.cuda:
images = images.cuda()
#---------------------------------------------------------#
# 将图像输入网络当中进行预测！
#---------------------------------------------------------#
outputs = self.net(images)
outputs = self.bbox_util.decode_box(outputs)
#---------------------------------------------------------#
# 将预测框进行堆叠，然后进行非极大抑制
#---------------------------------------------------------#
results = self.bbox_util.non_max_suppression(torch.cat(outputs, 1), self.num_classes, self.input_shape,
image_shape, self.letterbox_image, conf_thres=self.confidence, nms_thres=self.nms_iou)
t1 = time.time()
for _ in range(test_interval):
with torch.no_grad():
#---------------------------------------------------------#
# 将图像输入网络当中进行预测！
#---------------------------------------------------------#
outputs = self.net(images)
outputs = self.bbox_util.decode_box(outputs)
#---------------------------------------------------------#
# 将预测框进行堆叠，然后进行非极大抑制
#---------------------------------------------------------#
results = self.bbox_util.non_max_suppression(torch.cat(outputs, 1), self.num_classes, self.input_shape,
image_shape, self.letterbox_image, conf_thres=self.confidence, nms_thres=self.nms_iou)
t2 = time.time()
tact_time = (t2 - t1) / test_interval
return tact_time
def detect_heatmap(self, image, heatmap_save_path):
import cv2
import matplotlib.pyplot as plt
def sigmoid(x):
y = 1.0 / (1.0 + np.exp(-x))
return y
#---------------------------------------------------------#
# 在这里将图像转换成RGB图像，防止灰度图在预测时报错。
# 代码仅仅支持RGB图像的预测，所有其它类型的图像都会转化成RGB
#---------------------------------------------------------#
image = cvtColor(image)
#---------------------------------------------------------#
# 给图像增加灰条，实现不失真的resize
# 也可以直接resize进行识别
#---------------------------------------------------------#
image_data = resize_image(image, (self.input_shape[1],self.input_shape[0]), self.letterbox_image)
#---------------------------------------------------------#
# 添加上batch_size维度
#---------------------------------------------------------#
image_data = np.expand_dims(np.transpose(preprocess_input(np.array(image_data, dtype='float32')), (2, 0, 1)), 0)
with torch.no_grad():
images = torch.from_numpy(image_data)
if self.cuda:
images = images.cuda()
#---------------------------------------------------------#
# 将图像输入网络当中进行预测！
#---------------------------------------------------------#
outputs = self.net(images)
plt.imshow(image, alpha=1)
plt.axis('off')
mask = np.zeros((image.size[1], image.size[0]))
for sub_output in outputs:
sub_output = sub_output.cpu().numpy()
b, c, h, w = np.shape(sub_output)
sub_output = np.transpose(np.reshape(sub_output, [b, 3, -1, h, w]), [0, 3, 4, 1, 2])[0]
score = np.max(sigmoid(sub_output[..., 4]), -1)
score = cv2.resize(score, (image.size[0], image.size[1]))
normed_score = (score * 255).astype('uint8')
mask = np.maximum(mask, normed_score)
plt.imshow(mask, alpha=0.5, interpolation='nearest', cmap="jet")
plt.axis('off')
plt.subplots_adjust(top=1, bottom=0, right=1, left=0, hspace=0, wspace=0)
plt.margins(0, 0)
plt.savefig(heatmap_save_path, dpi=200, bbox_inches='tight', pad_inches = -0.1)
print("Save to the " + heatmap_save_path)
plt.show()
def convert_to_onnx(self, simplify, model_path):
import onnx
self.generate(onnx=True)
im = torch.zeros(1, 3, *self.input_shape).to('cpu') # image size(1, 3, 512, 512) BCHW
input_layer_names = ["images"]
output_layer_names = ["output"]
# Export the model
print(f'Starting export with onnx {onnx.__version__}.')
torch.onnx.export(self.net,
im,
f = model_path,
verbose = False,
opset_version = 12,
training = torch.onnx.TrainingMode.EVAL,
do_constant_folding = True,
input_names = input_layer_names,
output_names = output_layer_names,
dynamic_axes = None)
# Checks
model_onnx = onnx.load(model_path) # load onnx model
onnx.checker.check_model(model_onnx) # check onnx model
# Simplify onnx
if simplify:
import onnxsim
print(f'Simplifying with onnx-simplifier {onnxsim.__version__}.')
model_onnx, check = onnxsim.simplify(
model_onnx,
dynamic_input_shape=False,
input_shapes=None)
assert check, 'assert check failed'
onnx.save(model_onnx, model_path)
print('Onnx model save as {}'.format(model_path))
def get_map_txt(self, image_id, image, class_names, map_out_path):
f = open(os.path.join(map_out_path, "detection-results/"+image_id+".txt"), "w", encoding='utf-8')
image_shape = np.array(np.shape(image)[0:2])
#---------------------------------------------------------#
# 在这里将图像转换成RGB图像，防止灰度图在预测时报错。
# 代码仅仅支持RGB图像的预测，所有其它类型的图像都会转化成RGB
#---------------------------------------------------------#
image = cvtColor(image)
#---------------------------------------------------------#
# 给图像增加灰条，实现不失真的resize
# 也可以直接resize进行识别
#---------------------------------------------------------#
image_data = resize_image(image, (self.input_shape[1], self.input_shape[0]), self.letterbox_image)
#---------------------------------------------------------#
# 添加上batch_size维度
#---------------------------------------------------------#
image_data = np.expand_dims(np.transpose(preprocess_input(np.array(image_data, dtype='float32')), (2, 0, 1)), 0)
with torch.no_grad():
images = torch.from_numpy(image_data)
if self.cuda:
images = images.cuda()
#---------------------------------------------------------#
# 将图像输入网络当中进行预测！
#---------------------------------------------------------#
outputs = self.net(images)
outputs = self.bbox_util.decode_box(outputs)
#---------------------------------------------------------#
# 将预测框进行堆叠，然后进行非极大抑制
#---------------------------------------------------------#
results = self.bbox_util.non_max_suppression(torch.cat(outputs, 1), self.num_classes, self.input_shape,
image_shape, self.letterbox_image, conf_thres = self.confidence, nms_thres = self.nms_iou)
if results[0] is None:
return
top_label = np.array(results[0][:, 6], dtype = 'int32')
top_conf = results[0][:, 4] * results[0][:, 5]
top_boxes = results[0][:, :4]
for i, c in list(enumerate(top_label)):
predicted_class = self.class_names[int(c)]
box = top_boxes[i]
score = str(top_conf[i])
top, left, bottom, right = box
if predicted_class not in class_names:
continue
f.write("%s %s %s %s %s %s\n" % (predicted_class, score[:6], str(int(left)), str(int(top)), str(int(right)),str(int(bottom))))
f.close()
return
class YOLO_ONNX(object):
_defaults = {
#--------------------------------------------------------------------------#
# 使用自己训练好的模型进行预测一定要修改onnx_path和classes_path！
# onnx_path指向logs文件夹下的权值文件，classes_path指向model_data下的txt
#
# 训练好后logs文件夹下存在多个权值文件，选择验证集损失较低的即可。
# 验证集损失较低不代表mAP较高，仅代表该权值在验证集上泛化性能较好。
# 如果出现shape不匹配，同时要注意训练时的onnx_path和classes_path参数的修改
#--------------------------------------------------------------------------#
"onnx_path" : 'model_data/models.onnx',
"classes_path" : 'model_data/coco_classes.txt',
#---------------------------------------------------------------------#
# anchors_path代表先验框对应的txt文件，一般不修改。
# anchors_mask用于帮助代码找到对应的先验框，一般不修改。
#---------------------------------------------------------------------#
"anchors_path" : 'model_data/yolo_anchors.txt',
"anchors_mask" : [[6, 7, 8], [3, 4, 5], [0, 1, 2]],
#---------------------------------------------------------------------#
# 输入图片的大小，必须为32的倍数。
#---------------------------------------------------------------------#
"input_shape" : [640, 640],
#---------------------------------------------------------------------#
# 只有得分大于置信度的预测框会被保留下来
#---------------------------------------------------------------------#
"confidence" : 0.5,
#---------------------------------------------------------------------#
# 非极大抑制所用到的nms_iou大小
#---------------------------------------------------------------------#
"nms_iou" : 0.3,
#---------------------------------------------------------------------#
# 该变量用于控制是否使用letterbox_image对输入图像进行不失真的resize，
# 在多次测试后，发现关闭letterbox_image直接resize的效果更好
#---------------------------------------------------------------------#
"letterbox_image" : True
}
@classmethod
def get_defaults(cls, n):
if n in cls._defaults:
return cls._defaults[n]
else:
return "Unrecognized attribute name '" + n + "'"
#---------------------------------------------------#
# 初始化YOLO
#---------------------------------------------------#
def __init__(self, **kwargs):
self.__dict__.update(self._defaults)
for name, value in kwargs.items():
setattr(self, name, value)
self._defaults[name] = value
import onnxruntime
self.onnx_session = onnxruntime.InferenceSession(self.onnx_path)
# 获得所有的输入node
self.input_name = self.get_input_name()
# 获得所有的输出node
self.output_name = self.get_output_name()
#---------------------------------------------------#
# 获得种类和先验框的数量
#---------------------------------------------------#
self.class_names, self.num_classes = self.get_classes(self.classes_path)
self.anchors, self.num_anchors = self.get_anchors(self.anchors_path)
self.bbox_util = DecodeBoxNP(self.anchors, self.num_classes, (self.input_shape[0], self.input_shape[1]), self.anchors_mask)
#---------------------------------------------------#
# 画框设置不同的颜色
#---------------------------------------------------#
hsv_tuples = [(x / self.num_classes, 1., 1.) for x in range(self.num_classes)]
self.colors = list(map(lambda x: colorsys.hsv_to_rgb(*x), hsv_tuples))
self.colors = list(map(lambda x: (int(x[0] * 255), int(x[1] * 255), int(x[2] * 255)), self.colors))
show_config(**self._defaults)
def get_classes(self, classes_path):
with open(classes_path, encoding='utf-8') as f:
class_names = f.readlines()
class_names = [c.strip() for c in class_names]
return class_names, len(class_names)
def get_anchors(self, anchors_path):
'''loads the anchors from a file'''
with open(anchors_path, encoding='utf-8') as f:
anchors = f.readline()
anchors = [float(x) for x in anchors.split(',')]
anchors = np.array(anchors).reshape(-1, 2)
return anchors, len(anchors)
def get_input_name(self):
# 获得所有的输入node
input_name=[]
for node in self.onnx_session.get_inputs():
input_name.append(node.name)
return input_name
def get_output_name(self):
# 获得所有的输出node
output_name=[]
for node in self.onnx_session.get_outputs():
output_name.append(node.name)
return output_name
def get_input_feed(self,image_tensor):
# 利用input_name获得输入的tensor
input_feed={}
for name in self.input_name:
input_feed[name]=image_tensor
return input_feed
#---------------------------------------------------#
# 对输入图像进行resize
#---------------------------------------------------#
def resize_image(self, image, size, letterbox_image, mode='PIL'):
if mode == 'PIL':
iw, ih = image.size
w, h = size
if letterbox_image:
scale = min(w/iw, h/ih)
nw = int(iw*scale)
nh = int(ih*scale)
image = image.resize((nw,nh), Image.BICUBIC)
new_image = Image.new('RGB', size, (128,128,128))
new_image.paste(image, ((w-nw)//2, (h-nh)//2))
else:
new_image = image.resize((w, h), Image.BICUBIC)
else:
image = np.array(image)
if letterbox_image:
# 获得现在的shape
shape = np.shape(image)[:2]
# 获得输出的shape
if isinstance(size, int):
size = (size, size)
# 计算缩放的比例
r = min(size[0] / shape[0], size[1] / shape[1])
# 计算缩放后图片的高宽
new_unpad = int(round(shape[1] * r)), int(round(shape[0] * r))
dw, dh = size[1] - new_unpad[0], size[0] - new_unpad[1]
# 除以2以padding到两边
dw /= 2
dh /= 2
# 对图像进行resize
if shape[::-1] != new_unpad: # resize
image = cv2.resize(image, new_unpad, interpolation=cv2.INTER_LINEAR)
top, bottom = int(round(dh - 0.1)), int(round(dh + 0.1))
left, right = int(round(dw - 0.1)), int(round(dw + 0.1))
new_image = cv2.copyMakeBorder(image, top, bottom, left, right, cv2.BORDER_CONSTANT, value=(128, 128, 128)) # add border
else:
new_image = cv2.resize(image, (w, h))
return new_image
def detect_image(self, image):
image_shape = np.array(np.shape(image)[0:2])
#---------------------------------------------------------#
# 在这里将图像转换成RGB图像，防止灰度图在预测时报错。
# 代码仅仅支持RGB图像的预测，所有其它类型的图像都会转化成RGB
#---------------------------------------------------------#
image = cvtColor(image)
image_data = self.resize_image(image, self.input_shape, True)
#---------------------------------------------------------#
# 添加上batch_size维度
# h, w, 3 => 3, h, w => 1, 3, h, w
#---------------------------------------------------------#
image_data = np.expand_dims(np.transpose(preprocess_input(np.array(image_data, dtype='float32')), (2, 0, 1)), 0)
input_feed = self.get_input_feed(image_data)
outputs = self.onnx_session.run(output_names=self.output_name, input_feed=input_feed)
feature_map_shape = [[int(j / (2 ** (i + 3))) for j in self.input_shape] for i in range(len(self.anchors_mask))][::-1]
for i in range(len(self.anchors_mask)):
outputs[i] = np.reshape(outputs[i], (1, len(self.anchors_mask[i]) * (5 + self.num_classes), feature_map_shape[i][0], feature_map_shape[i][1]))
outputs = self.bbox_util.decode_box(outputs)
#---------------------------------------------------------#
# 将预测框进行堆叠，然后进行非极大抑制
#---------------------------------------------------------#
results = self.bbox_util.non_max_suppression(np.concatenate(outputs, 1), self.num_classes, self.input_shape,
image_shape, self.letterbox_image, conf_thres = self.confidence, nms_thres = self.nms_iou)
if results[0] is None:
return image
top_label = np.array(results[0][:, 6], dtype = 'int32')
top_conf = results[0][:, 4] * results[0][:, 5]
top_boxes = results[0][:, :4]
#---------------------------------------------------------#
# 设置字体与边框厚度
#---------------------------------------------------------#
font = ImageFont.truetype(font='model_data/simhei.ttf', size=np.floor(3e-2 * image.size[1] + 0.5).astype('int32'))
thickness = int(max((image.size[0] + image.size[1]) // np.mean(self.input_shape), 1))
#---------------------------------------------------------#
# 图像绘制
#---------------------------------------------------------#
for i, c in list(enumerate(top_label)):
predicted_class = self.class_names[int(c)]
box = top_boxes[i]
score = top_conf[i]
top, left, bottom, right = box
top = max(0, np.floor(top).astype('int32'))
left = max(0, np.floor(left).astype('int32'))
bottom = min(image.size[1], np.floor(bottom).astype('int32'))
right = min(image.size[0], np.floor(right).astype('int32'))
label = '{} {:.2f}'.format(predicted_class, score)
draw = ImageDraw.Draw(image)
label_size = draw.textsize(label, font)
label = label.encode('utf-8')
print(label, top, left, bottom, right)
if top - label_size[1] >= 0:
text_origin = np.array([left, top - label_size[1]])
else:
text_origin = np.array([left, top + 1])
for i in range(thickness):
draw.rectangle([left + i, top + i, right - i, bottom - i], outline=self.colors[c])
draw.rectangle([tuple(text_origin), tuple(text_origin + label_size)], fill=self.colors[c])
draw.text(text_origin, str(label,'UTF-8'), fill=(0, 0, 0), font=font)
del draw
return image