使用彩色直方图均衡法改进的Mean Shift行人跟踪算法  

Improved Mean Shift Algorithm for Pedestrian Tracking Based on Color Histogram Equalization

作　　者：付丽梅[1] FU Limei(Department of Software Engineering,Dalian Neusoft University of Information,Dalian 116023,China)

机构地区：[1]大连东软信息学院软件工程系,辽宁大连116023

出　　处：《软件工程》2019年第2期17-19,共3页Software Engineering

摘　　要：视频监控中的行人跟踪是一个具有实际应用价值的研究领域。针对城市中智能视频监控中行人跟踪问题,提出了一种改进的基于彩色直方图均衡化的MeanShift行人跟踪算法,实验的图像序列通过监控摄像头获得。这种行人跟踪算法是基于均值移位算法,它使用颜色直方图均衡化来改进原始算法。通过使用CAVIARproject/IST200137540数据集来进行试验,试验结果表明,改进后的算法在某些情况下比原算法表现得更好。The pedestrian tracking in video monitoring is a research field with practical application value.An improved pedestrian tracking algorithm is proposed with image sequences acquired by surveillance cameras to solve the problem of pedestrian tracking in in smart video monitoring in cities.This pedestrian tracking algorithm is based on mean shift algorithm,using color histogram equalization to improve the original algorithm and applying CAVIAR project/IST 2001 37540 dataset to evaluate the algorithm.Experimental results show that the improved algorithm performs much better than the original one in some situations.

关 键 词：计算机图像处理 MEANSHIFT 行人跟踪 直方图均衡化 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]