基于YOLOv5的行人检测系统研究  

Research on Pedestrian Detection System Based on YOLOv5

作　　者：焦天文 田秀云 JIAO Tianwen;TIAN Xiuyun(School of Electronics and Information Engineering,Guangdong Ocean University,Zhanjiang 524088,China)

机构地区：[1]广东海洋大学电子与信息工程学院,广东湛江524088

出　　处：《现代信息科技》2025年第2期33-37,45,共6页Modern Information Technology

基　　金：2022年广东海洋大学校级课程-数学物理方法(010301112202)。

摘　　要：针对计算量冗余和精度低的问题,提出一种改进的YOLOv5行人检测模型。系统采用Ghost卷积结合Transformer自注意力机制,结合双向金字塔结构以及EIoU损失函数,将INRIA行人检测数据集按照7∶2∶1的比例分配训练集、验证集和测试集,采用SGD优化器对模型进行300个Epochs的训练,并利用训练好的权重模型对测试集进行检测,结果表明:改进模型检测的平均精度值增加了1.5%,且计算量显著降低。Aiming at problems of the amount of calculation with redundancy and low accuracy,an improved YOLOv5 pedestrian detection model is proposed.This system uses Ghost convolution to combine with the Transformer Self-Attention Mechanism,then combines with the BiFPN structure and EIoU loss function.The INRIA pedestrian detection data set is divided into the training set,the validation set,and the test set according to the ratio of 7∶2∶1.The SGD optimizer is used to train the model for 300 Epochs,and the trained weight model is used to test the test set.The results show that the average accuracy of the improved model detection is increased by 1.5%,and the amount of calculation is significantly reduced.

关 键 词：行人检测 YOLOv5模型 Ghost卷积 双向金字塔结构 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]