Improved YOLOv5-based sleep post detection method

The invention relates to an improved YOLOv5-based sleep detection method, which is characterized in that a suspected sleep detection model for detecting suspected sleep persons is constructed based on an improved YOLOv5 model structure combined with an improved CBAM attention mechanism, and tracking detection of continuous sleep suspected persons is realized by combining two-stage detection design of feature matching and detection frame matching in sequence. And finally, based on the continuity of the suspected sleep person, accurate detection of the person in the sleep state is realized. In the design scheme, the Yolov5 model and the Deepsort network are improved and simultaneously applied, so that the accuracy and the robustness of sleep detection are effectively improved, and the method has a wide application prospect in the aspect of improving the working safety and the enterprise revenue. 本发明涉及一种基于改进YOLOv5的睡岗检测方法，基于结合改进CBAM注意力机制的改进型YOLOv5模型结构，构建用于检测疑似睡眠人物的疑似睡眠检测模型，并结合依次所进行特征匹配、检测框匹配的二级检测设计，实现连续睡眠疑似人物的跟踪检