基于帧间差分和金字塔光流法的运动目标检测  

Algorithm for moving object detection based on inter-frame differencing and pyramid optical flow methods

作　　者：郝慧琴[1] 王耀力[1] 

机构地区：[1]太原理工大学信息工程学院,山西太原030024

出　　处：《电视技术》2016年第7期134-138,共5页Video Engineering

基　　金：山西省自然科学基金项目(2013011015-1)

摘　　要：针对用于运动目标检测的光流算法存在处理复杂、计算量大等问题,提出一种帧间差分算法和金字塔LK光流法相结合的运动目标检测方案。该方法先对视频图像进行帧间差分处理,得到图像的运动区域,再对该运动区域进行金字塔LK光流计算,减少了计算区域,提高目标检测的速度。最后在搭建的视觉避障平台上使用Lab VIEW语言进行算法程序验证,实验结果证明了算法的有效性。A method to detect moving objects using Inter-frame differencing and pyramid LK optical flow is proposed, as optical flow algorithm has a high computational complexity and large computation, which make the limitation certain. Firstly, the motion area of the video image can be detected by the inter-frame differencing method, and then the optical flow of this area is calculated using pyramid LK flow method. The computed area is reduced, and therefore the detection speed is improved. Finally, this im- proved method is verified on the visual obstacle avoidance platform by LabVIEW programs and the experiment results show the ef- fectiveness of this algorithm.

关 键 词：帧间差分法 金字塔LK光流法 运动目标检测 视觉避障平台 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]