基于线性内插权值人群密度估计的方法研究  

Research on crowd density estimation method based on linear interpolation weights

作　　者：张朋[1] 温宏愿[1] ZHANG Peng;WEN Hongyuan(Taizhou Institute of Sei.＆ Teeh.,NUST,Taizhou 225300,Chin)

机构地区：[1]南京理工大学泰州科技学院,江苏泰州225300

出　　处：《现代电子技术》2018年第14期79-82,共4页Modern Electronics Technique

基　　金：江苏省"333工程"科研资助项目:江苏省高校"青蓝工程"项目(BRA2016177);泰州市科技支撑指导项目(SSF20160033)~~

摘　　要：针对密集人群环境中行人相互遮挡造成人数难以准确统计问题,提出一种基于线性内插权值的人数密度统计方法。系统从视频监控设备获取图像后,首先对图像进行预处理以减少原始图像的干扰,通过混合高斯建模提取前景图像,然后通过线性内插权值进行透视矫正,最后通过构造和提取能量、对比度、熵和相关性四个特征参数进行人数统计。结果表明,在不同的测试环境下所提方法的准确率在90%以上,能够达到人群密度监控的基本要求。In allusion to the problem that it is difficult to accurately count the number of pedestrians in a crowded environment due to people′ s blocking from each other,a population density statistical method based on linear interpolation weights is proposed. The image obtained from the video surveillance equipment of the system is preprocessed to reduce the interference of the original image and extract the foreground image by using the mixed Gaussian modeling,and then the perspective correction is performed by using the linear interpolation weights. Population statistics are obtained by constructing and extracting the four characteristic parameters of energy,contrast,entropy and correlation. The results indicate that the method in this paper has over 90% of accuracy rate in different test environments,which can meet the basic requirement of population density monitoring.

关 键 词：人群密度估计 混合高斯建模 透视矫正 能量 对比度 相关性 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]