- author is leilei
- reflective-clothes-detect qq群: 980489677
- 如果此项目对您有所帮助，请给个star，您的star是对我的鼓励！
- reflective-clothes-detect-dataset (with xml annotations 1083) download: BaiDuYunPan 提取码->(dooh)
yolov5s's weight is in reflective-clothes-detect-dataset !
- download BaidDuYunPan's data and weight file
- put yolov5s's weight file into yolov5 folder
- cd yolov5, and excuting an order:
python detect.py --source ***/aaa.jpg --weights ./best.pt - convert VOC2021 to YOLO format：
call yolov5's voc_label_Re.py
-
We annotate the reflective-clothes-detect-dataset as Pascal VOC format:
--VOC2021(反光衣) --Annotations (xml_num: 1083) --ImageSets(Main) --JPEGImages (image_num: 1083) --images (yolov5 need, the same with JPEGImages) --labels (yolov4-5 need) --txt_yolov5 (yolov5 need 2021_train.txt) --2021_train.txt (yolov4 need) --yolov5_weight (yolov5s's weight) --label_name: reflective_clothes、other_clothes -
If you want to crawl some images
Please refer to this crawler code on github: https://github.com/gengyanlei/fire-detect-yolov4 -> crawl_baidu.py
- 基于SHWD数据集进行安全帽-反光衣-整体人 5类标注训练yolov4-yolov5，实现施工区域or危险区域检测
- Based on the SHWD data set, perform the five-category labeling training yolov4-yolov5 for helmet-reflective clothing-holistic people to achieve detection of construction areas or dangerous areas.
- ./result:
 |
|---|
 |
| ---- |
 |
| ---- |
- Based on the trained yolov4 model of 1083 reflective clothing images, expand the reflective clothing category of the SHWD data set;
- Based on the data set expanded by 0 steps, train yolov4.
- 基于1083张反光衣图像训练好的yolov4模型，对SHWD数据集进行反光衣类别扩充；
- 基于0步骤扩充得到的数据集，训练yolov4.