结合细粒度特征与注意力机制的行人检测  

PEDESTRIAN DETECTION COMBINING FINE-GRAINED FEATURE AND ATTENTION MECHANISM

作　　者：肖顺亮 强赞霞[1] 李丹阳 刘卫光[1] Xiao Shunliang;Qiang Zanxia;Li Danyang;Liu Weiguang(School of Computer Science,Zhongyuan University of Technology,Zhengzhou 451191,Henan,China)

机构地区：[1]中原工学院计算机学院,河南郑州451191

出　　处：《计算机应用与软件》2025年第4期166-173,207,共9页Computer Applications and Software

基　　金：国家自然科学基金项目(61802454);河南省科技攻关项目(182102210126)。

摘　　要：由于行人易被遮挡、尺度差异大等原因,导致行人漏检率高。针对这种情况,对基于Anchor-free思想的行人检测算法进行改进。针对卷积神经网络提取特征时对目标尺度变化敏感的问题,提出细粒度特征融合策略,获取丰富的行人特征信息。采用空间域注意力机制对特征图不同区域进行权重学习,提高模型的表达能力。利用多尺度检测方法,使模型自适应检测不同尺度的行人,增强模型检测时的鲁棒性。实验结果表明,改进算法在Cityperson数据集的Reasonable、Bare、Partial和Heavy子集上分别取得11.33%、6.81%、11.52%和50.09%的MR-2,性能优于其他行人检测算法。The pedestrian is easy to be blocked and the scale is different,so the pedestrian missed detection rate is high.In view of this,the pedestrian detection algorithm based on Anchor-free idea is improved.Aimed at the problem that convolutional neural network was sensitive to target scale changes when extracting features,a fine-grained feature fusion strategy was proposed to obtain rich pedestrian feature information.The spatial attention mechanism was used to study the weight of different regions of the feature map to improve the expression ability of the model.Using multi-scale detection method,the model adaptively detected pedestrians of different scales and enhanced the robustness of model detection.The experimental results show that MR-2 of 11.33%,6.81%,11.52%and 50.09%are obtained on Reasonable,Bare,Partial and Heavy subsets of Cityperson dataset,respectively,which is better than other pedestrian detection algorithms.

关 键 词：行人检测 无锚框 特征融合 多尺度 注意力机制 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]