基于复小波光流法和利用平均算子的二次三帧差法的运动目标检测方法  

A Moving Objects Detection Method Based on Complex Wavelet Optical Flow and Second Three-Difference Method Using Averaging Operator

作　　者：凌滨[1] 邓艳[1] 鞠美玉 于士博[1] 

机构地区：[1]东北林业大学机电工程学院,黑龙江哈尔滨150040

出　　处：《青岛科技大学学报（自然科学版）》2016年第6期693-700,共8页Journal of Qingdao University of Science and Technology:Natural Science Edition

摘　　要：针对机器视觉系统中快速运动目标的检测问题,提出了一种结合复小波光流法和利用平均算子的二次三帧差法的运动目标检测新算法。首先,在考虑算法复杂度的情况下,选择图像中的Harris角点作为特征点计算光流信息;然后运用复小波光流法对运动目标进行更精确的运动估计,为了得到更完整的检测区域,同时引入改进三帧差法作为光流法的补充。改进的二次三帧差法通过平均算子滤波,能较好的抑制噪声,消除背景变化带来的干扰。通过QT和OpenCV搭建的平台,利用该算法对快速运动目标进行检测,结果表明:该算法能检测出动态背景下快速运动的目标,其检测率高于90%,虚警率低于10%,证明了算法的有效性和实用性。In order to solve the problem of machine vision system in fast moving targets detection,new methods of moving target detection are proposed in this paper,which are complex wavelet optical flow and the second three-difference using the average operator.Firstly,in the case of considering the complexity of the algorithm,the Harris corner points in the image are used as the feature points to calculate the optical flow information.Then by using of complex small waves flow method,more accurate motion estimation of moving target is got,in order to get a more complete detection area,while the improved three frame difference method is introduced as a complement to the optical flow method.The improved second three-difference method filtered by averaging operator can restrain the noise and eliminate the interference caused by the background variation.Based on the platform of QT and OpenCV,this algorithm is used to detect the fast moving objects.The results show that the algorithm can detect the fast moving target in dynamic background,the detection rate is higher than 90%,and the false alarm rate is lower than 10%,which proves the algorithm′s effectiveness and practicability.

关 键 词：复小波光流法 二次三帧差法 HARRIS角点 快速运动目标 目标检测 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]