基于深度学习的车辆目标检测算法综述  

A Review of Vehicle Target Detection Algorithms Based on Deep Learning

作　　者：杨伟[1] 杜学峰 张勇[1] 高越 YANG Wei;DU Xuefeng;ZHANG Yong;GAO Yue(School of Automobile,Chang'an University,Shaanxi Xi'an 710064)

机构地区：[1]长安大学汽车学院,陕西西安710064

出　　处：《汽车实用技术》2022年第2期24-26,共3页Automobile Applied Technology

摘　　要：车辆目标检测是自动驾驶环境感知的重要组成部分。近年来随着深度学习在目标识别领域取得重大突破,基于深度学习的车辆目标检测算法逐渐成为该领域的研究热点。论文对当前主流的两阶段车辆目标检测算法和单阶段车辆目标检测算法进行简要介绍,分析了其中几种具有代表性的卷积神经网络算法的优缺点,最后总结目前车辆目标检测存在的问题以及未来的发展方向。Vehicle target detection is an important part of automatic driving environment perception.In recent years,with the great breakthrough of deep learning in the field of target recognition,vehicle target detection algorithm based on deep learning has gradually become a research hotspot in this field.This paper briefly introduces the current mainstream two-stage vehicle target detection algorithm and single-stage vehicle target detection algorithm,analyzes the advantages and disadvan-tages of several representative convolution neural network algorithms,and finally summarizes the existing problems and future development direction of vehicle target detection.

关 键 词：深度学习 卷积神经网络 车辆检测 目标检测算法 

分 类 号：U495[交通运输工程—交通运输规划与管理]