Boat rotation angle detection method based on improved Yolo v8s-seg
Intelligent control lock switching system for water toll station needs to be able to accurately detect the rotation angle of boats in the channel in order to be able to automatically recognize the movement of boats inside and outside the lock.Traditional target detection algorithms have problems such as limited accuracy and poor real-time performance.To address the above problems,a model Yolo v8s-seg-boat is proposed to detect the rotation angle of a boat by adding CA(Coordinate Attention)mechanism on the basis of Yolo v8s-seg model.The model adopts the instance segmentation algorithm to extract the boat's contour points,and judge the boat's center of gravity accordingly,and finally calculate the rotation angle of the boat.The experimental results show that the model improves the segmentation evaluation index mAP by 1.8%compared with Yolo v8s-seg on the boat dataset captured at the water toll station,and the segmentation precision rate reaches 97.6%,and the error between the acquired boat rotation angle and the actual angle is smaller than that of the Yolo v8s-seg model.
instance segmentationrotation angleattentional mechanismcenter of gravity
万方数据
维普
