基于风险决策与混合高斯背景模型的前景目标突变快速检测  

FAST DETECTION OF FOREGROUND TARGET MUTATIONS BASED ON RISK DECISION AND GAUSSIAN MIXTURE BACKGROUND MODEL

作　　者：闫英战[1] 杨勇[1] 

机构地区：[1]广东科技学院计算机系,广东东莞523083

出　　处：《华南师范大学学报（自然科学版）》2011年第2期56-59,共4页Journal of South China Normal University(Natural Science Edition)

基　　金：国家基础科学研究基金(A1420060159)

摘　　要：将风险决策引入前景目标的突变判断中,通过设计一个时序计数器函数来记录图像上某一像素点被划为前景的次数,当次数大于某一阈值时,将该像素从前景点改判为背景点,可以估计该像素点为背景点的概率,做出风险决策,以便及时更新混合高斯背景模型参数,减少多个高斯模型的高额计算量.最后通过实验验证了算法在目标检测率和实时性方面的改进.Risk decision is introduced to judge mutations of foreground target.A timing counter function is designed to record the times of a pixel judged as foreground pixel.When the times is greater than a certain threshold value,the pixel will be judged as a background pixel instead of a foreground pixel.So the probability of background pixel can be evaluated for making risk decision,in order to update the GMM parameters and reduce the large computation of the multiple Gaussian model.Finally,the algorithm is verified by experiments for the target detection rate and real-time improvements.

关 键 词：目标检测 混合高斯模型 背景建模 风险决策 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]