Image information-based client flow volume identification statistic method

作者: Fang Kangling , Fu Xiaowei , He Peng

DOI:

关键词:

摘要: The invention relates to an image information-based client flow volume identification statistic method. method is improved self-adapting movement target detecting After targets existing in a area detected, the sequence of processed according following steps firstly, predicting search areas images on basis Kalman filtration; secondly, matching cost function, creating chain, and determining corresponding relation among images; finally, association between each frame image, whether enter or depart from area, are still continuously moved in/out so that aim statistics fulfilled. has characteristics high real-time property, rate, full data, low hardware requirement like.

参考文章(2)
V. Srikrishnan, Subhasis Chaudhuri, Tadinada Nagaraj, Visual object tracking with scale and orientation adaptation ,(2009)
Li Yiyue, Tian Qiuchang, Hu Bin, Wang Feiyue, Liu Siyi, Xiong Gang, Tian Bin, Method and system for passenger flow statistics based on cameras with intelligent analysis function ,(2014)