基于自适应阈值的Surendra背景提取方法研究  

Research on Surendra Background Extraction Method Based on Adaptive Threshold

作　　者：单玉刚[1] 汪家宝[1] 郝峰[1] 

机构地区：[1]湖北文理学院教育学院,湖北襄阳441053

出　　处：《计算机技术与发展》2017年第7期91-95,共5页Computer Technology and Development

基　　金：河南省基础与前沿技术研究计划项目(142300410295);湖北文理学院博士基金(2015B002)

摘　　要：针对使用OTSU算法分割的Surendra背景提取算法实时性和精度的不足,在分析差分图像特征和阈值对图像分割影响的基础上,提出了一种自适应阈值的图像分割和背景提取及更新算法。通过采用切比雪夫不等式确定阈值可选范围,引入模糊隶属度改进基于灰度直方图的最大类间方差法,寻找区域最优分割阈值,准确分割出差分图像背景和目标。背景更新时,利用相邻帧差和自适应阈值法得到当前帧的二值图像,再根据前景目标更新得到新的背景,并适应光照强度的变化和场景变化。仿真实验结果表明,所提出的自适应阈值差分图像分割算法要优于OTSU算法的图像分割算法,应用在Surendra中,提取的背景精度高,效果好,可快速有效地对背景进行更新,并显著降低动态背景更新模型的计算量。In view of the problem of real-time and accuracy in Surendra background extraction algorithm by using OTSU, based on the a- nalysis of the features of the differential image and the effect of threshold on the image segmentation, an adaptive threshold algorithm for extraction and update of background has been presented. For a series of frames in video, the threshold range by using Chebyshev inequali- ty has been determined in which the final threshold could be acquired according to the gray histogram OTSU method based on fuzzy membership. During background updating, two adjacent frame difference and adaptive threshold method has been adopted to generate the binary image of the current frame and then the new background by considering the moving foreground targets in the frame and has been adapted to highlight intensity changes and scene changes. The experiment results show that the adaptive threshold algorithm for Surendra background extraction is better than the OTSU algorithm, which is applied to the Surendra and can update the background quickly and ef- fectively, and has reduced the cost of computing of dynamic background update model significantly.

关 键 词：自适应阈值 Surendra背景提取 切比雪夫不等式 模糊OTSU 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]