Research on Deep Reinforcement Learning-based Pedestrian Detection
To address the current problems of fixed single initialization training window of intelligences in target detection by deep reinforcement learning(DRL),high error detection rate and leakage rate of multi-target and small target images,a pedestrian detection method combining YOLOv5s and DQN algorithm is proposed in this paper.The method is able to search the number and area containing targets by YOLOv5s,set the initial enclosing frame of regression as the initialization window of the intelligent body,and improve the scale adaptation.The reward function of the traditional reinforcement learn-ing model is improved to make the reward and punishment feedback more accurate and improve the model detection ac-curacy and speed.Comparison experiments with existing target detection models based on deep learning and deep rein-forcement learning are conducted to obtain empirical results showing that the proposed pedestrian detection method can effectively improve the detection accuracy.
pedestrian detectiondeep reinforcement learningYOLOv5reward function
万方数据
维普
