colinlin1982 / yoloncv

test yolo object detector model with opencv4.2 dnn

yoloncv

(YOLO ON OPENCV) Test yolo object detector model with opencv4.2 dnn

Requirements

c++

1. opencv >= 4.2 (with dnn module)
2. cmake >= 3.13 python
3. opencv>=4.2

Usage

test on a image:
detect --image=/your/path/to/image.jpg --cfg=/your/path/to/config.data --thresh=0.5 --nms=0.6
detect --image=/your/path/to/image.jpg --cfg=/your/path/to/config.data
detect -i=/your/path/to/image.jpg -c=/your/path/to/config.data -t=0.5 -n=0.6

test on jpg images in a dir:
detect --dir=/your/path/to/dir --cfg=/your/path/to/config.data --thresh=0.5 --nms=0.6
python testoncv.py -d /your/path/to/dir -c /your/path/to/config.data -t 0.5 -n 0.6

About config.data

names : path to your classes name list, such as coco.names, voc.names
cfg : path to your darknet network cfg file, such as yolov3.cfg, yolov3-tiny.cfg
weights: path to your darknet network weights file, download from web or trained with your own dataset
width : network input width
height : network input height

Contact me

QQ群：695860125

Thanks

https://github.com/spmallick/learnopencv/tree/master/ObjectDetection-YOLO https://blog.csdn.net/reasonyuanrobot/article/details/89451198