基于FPGA的视频跟踪系统目标检测方法研究  

Study of target detection methods in video tracking system

作　　者：冯亮[1] 刘紫燕[1] 祁佳[1] 

机构地区：[1]贵州大学电子信息学院,贵州贵阳550025

出　　处：《微型机与应用》2014年第12期34-36,39,共4页Microcomputer & Its Applications

基　　金：贵州省科学技术基金(黔科合J字[2011]2193);贵州大学自然科学青年基金(2010(035))

摘　　要：为了提高视频跟踪系统中目标检测的精度,提出了一种以FPCA为核心的目标检测方法。利用形态学对S0bel目标边缘检测的结果进行优化,在此基础上,改进新的目标检测算法弥补形态学优化存在的不足。FPcA实现结果表明,改进的目标检测方法可以实现任何阈值情况下目标边缘的深度优化,使得图像边缘更清晰、噪声更小。In in order to improve object detection accuracy in video tracking system , a target detection scheme with FPGA as the core is proposed . It uses morphology method to improve the results of sobel target edge detection , and proposes a novel method for target detection to compensate the problem of morphology optimization . The experiment results on FPGA demonstrate that the proposed method can achieve depth optimization of target edge in any threshold case and make clearer image edge with lower noise .

关 键 词：形态学 SOBEL FPGA 目标检测 

分 类 号：TP216.1[自动化与计算机技术—检测技术与自动化装置]