改进的Otsu算法及在运动目标检测中的应用研究  

Research on Improved Otsu Algorithm and Its Application in Moving Objection Detection

作　　者：杨胜辉[1] 李太君[1] 肖沙[1] 周浩理 徐宁敏 

机构地区：[1]海南大学信息科学技术学院,海南海口570228 [2]海南省公安厅,海南海口570311

出　　处：《电视技术》2015年第24期100-103,112,共5页Video Engineering

基　　金：海南省社会发展科技专项(SF201455)

摘　　要：针对Otsu自适应阈值分割算法中阈值搜索精准度较低、效率不高的问题,结合视频图像序列的帧间相关性,利用遗传算法的全局寻优优势,及模拟退火算法较好的爬山性能,提出一种改进的Otsu算法,并与三帧差分法相结合应用于视频运动目标检测。实验证明,该算法相对Otsu算法和对比算法减少阈值分割中的寻优尝试次数,使最优阈值的选取更精确,并提高了目标检测效果。Considering the problem of the Otsu adaptive threshold segmentation algorithm which the accuracy of the threshold search is lower and the efficiency is not high in threshold search precision, in this paper, combined with the correlation between frames of video sequences, genetic algorithm is used with the advantage of global searching, and annealing algorithm is simulated with the better performance of mountain climbing, to improve Otsu algorithm. Then three frame difference method are contacted to detect moving objection in video sequences. Experimental results show that, the method reduced segmentation threshold optimization attempts, selected more accurate threshold compared with the traditional Otsu algorithm and comparison algorithm, and improved the target detection effects.

关 键 词：OTSU算法 遗传算法 模拟退火算法 三帧差分 阈值分割 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]