Aiming at the detection accuracy and speed of pedestrian detection in complex road environ-ment,a lightweight pedestrian detection algorithm based on multi-scale information and cross-dimension-al feature guidance is proposed.Firstly,based on the high-performance detector YOLOX,a multi-scale lightweight convolution is constructed and embedded in the backbone network to obtain multi-scale fea-ture information.Secondly,an end-to-end lightweight feature guided attention module is designed,which guides the model to focus on the visible region of pedestrian targets by fusing spatial information and re-lated information through cross-dimensional channel weighting mehod.Finally,in order to reduce the loss of feature information in the process of lightweight of the model,a feature fusion network is con-structed by depthwise separable convolution with increasing the depth of the receptive field.The experi-mental results show that compared with other mainstream detection algorithms,the proposed algorithm on the KITTI dataset reaches 71.03％detection accuracy and 80 FPS detection speed,which has better robustness and real-time performance in scenes with complex background,dense occlusion and different scales.
关键词
行人检测/多尺度/跨维特征引导/特征融合/轻量化模型
Key words
pedestrian detection/multi-scale/cross-dimensional feature guidance/feature fusion/light-weight model
维普
万方数据