五帧差分和边缘检测的运动目标检测  

Five Frame Difference and Edge Detection of Moving Target Detection

作　　者：舒欣[1] 李东新[1] 薛东伟[1] 

机构地区：[1]河海大学计算机与信息学院,南京211100

出　　处：《计算机系统应用》2014年第1期124-127,共4页Computer Systems & Applications

摘　　要：为了改进传统三帧差分的运动目标检测方法易出现的空洞和与虚假边缘的现象,提出了一种基于五帧差分和边缘检测相融合的运动目标检测方法.该算法首先利用优化的Canny边缘检测算法快速提取五帧连续图像的边缘图像,然后对五帧连续的边缘图像进图像的五帧差分运算,最后通过阈值分割和形态学处理完成对目标的提取.该方法计算简单,实验结果表明,该算法准确率高,连通性好,能满足实时性检测的要求.To improve the traditional three frame difference of moving target detection method in which the hole edge and false edge tend to appear, this paper puts forward a method based on the fusion of five frame difference and edge detection, moving object detection method. The algorithm firstly uses Canny edge detection algorithm of optimized fast extraction of five consecutive images edge image, and then to the edge of the frame of five consecutive image into five frames difference operation. Finally, by threshold segmentation and morphological processing is completed on the extraction of target. The method is simple in calculation and the experimental results show that the algorithm has high accuracy, good connectivity, and can meet the requirements of real-time detection.

关 键 词：五帧差分 边缘检测 运动目标检测 动态阈值 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]