Skip to content

juseok1729/BOHOL

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

84 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Custom-YOLO with Tensorflow 2.x

#BOHOL #Philippines #Scuba diving #Sea ecology #App #Flask

My Custom weights

Download : https://drive.google.com/file/d/1-L5U2JCJG73HMfHJNIBsK9123N1yaHqf/view

# weights → tf
python save_model.py --weights ./data/yolov4-obj_best.weights --output ./checkpoints/bohol-416 --input_size 416 --model yolov4

Dataset Statistics

created a 16 category saltwater fish dataset with roughly 200 images for each class.
16개의 클래스, 클래스 별 약 200장의 이미지 라벨링.
거북이 클래스 불균형..

Class_id Name Count
0 Green_sea_turtle 630
1 Anthias 287
2 Blonde_naso_tang 210
3 Blue_damsel 208
4 Brown_tang 248
5 Domino_damsel 259
6 Foxface_rabbitfish 227
7 Jackfish 254
8 Lionfish 315
9 Regal_angelfish 320
10 Scissortail_sergeant_major_fish 221
11 Sergeant_major_fish 183
12 Snowflake_moray 342
13 Three_stripes_damsel 271
14 Tomato_clown 193
15 Yellow_boxfish 308
  Total 4476

mAP (mean average precision)

Test images

# image
python detect.py --weights ./checkpoints/bohol-416 --size 416 --model yolov4 --images ./data/images/img1.jpg

Test videos

# video
python detect_video.py --weights ./checkpoints/bohol-416 --size 416 --model yolov4 --video ./data/video/vid1.mp4 --output ./detections/best2/vid1.avi

About

⚡ Object-detection, Tensorflow2.x

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published