Smoking target detection based on Yolo V3

The current smoking recognition and detection system in my country is mainly based on an intelligent analysis of the video, that is, target detection and recognition of smoking behavior in the monitoring area, and early warning in the fastest way when abnormal conditions are found, therefore, the advantages and disadvantages of the detection algorithm directly affect the safety factor. The Yolo V3 model used in this article is based on DarkNet53 as the backbone. After the feature map is output, it is decoded by the Bounding Box, and finally the prediction result is obtained. The prediction results can be used as the evaluation results of the model. The prediction results show that the smoke detection effect based on Yolo v3 is good, reaching a mAP of 0.76.