BP神经网络下的限速交通标志实时检测识别  

Real-time detection and identification of speed limit traffic signs under the BP neural network

作　　者：张兴国[1] 刘晓磊[1] 李靖[1] 王环东[2] ZHANG Xingguo;LIU Xiaolei;LI Jing;WANG Huandong(Chinese Flight T'cst Establishment,Xi’an 7 10089,China;School of Rcmiote Sensing and Informiation Engineering,Wuhan Univ.,Wuhan 430072,China)

机构地区：[1]中国飞行试验研究院,陕西西安710089 [2]武汉大学遥感信息工程学院,湖北武汉430072

出　　处：《西安电子科技大学学报》2018年第5期136-142,共7页Journal of Xidian University

基　　金：国家自然科学基金资助项目(F050105)

摘　　要：目前已有很多基于交通标志图片进行道路交通标志识别的系统研究,但存在误识别率较高的问题.因此文中基于视频数据对限速交通标志进行检测与识别,实现了限速交通标志的自动检测定位,并采用反向传播神经网络来进行道路标志的识别;同时,还采用连续自适应的均值漂移算法和光流法进行视频加速.实验表明,新提出的算法在耗时上缩短50%以上,检测识别准确度在90%以上,适用于相关智能识别领域.There are a lot of traffic signs based on the picture systemi of traffic sign recognition,and based on the video data to detect and identify the speed of traffic signs,However,there is a higher error rate.So this paper realizes the automiatic detection and localization of speed limit traffic signs,and uses the BP neural network to identify the road signs.Meanwhile,the CamiShift micthod and the optical flow micthod are used to speed up video,Experiment shows that the algorithmproposcd in this paper can shorten the time by miore than 50%,and the detection and recognition accuracy is over 90%,which is suitable for the related field of intelligent recognition.

关 键 词：限速交通标志 自动检测定位 人工神经网络 视频加速 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]