An Improved Detection Method of Human Target at Sea Based on Yolov3
Dongjin Li, Yu Liu, Jin Wang, Rufei Zhang, Feng Jiang, Fu Niu · 2021
In the mission of searching and rescuing, it is often faced with the situation that the area to be searched is large and the target to be searched is small. Combined with the object detection technology, this paper proposes a method for searching drowning people. At first, we make a dataset, which contains a large number of human targets at sea. Then, we improve the Yolov3 algorithm: In the feature extraction network, we use the residual module with channel attention mechanism. In the feature fusion network, we add a bottom-up structure to the FPN structure. Moreover, in terms of loss function, we use the CIoU loss function. Finally, on the settings of the anchor box, we use a linear transformation method to deal with the anchor boxes generated by clustering algorithm. The detection accuracy of the improved algorithm for human targets at sea is 72.17%, which has a good detection effect.