A Rapid Moving Human Detection Algorithm On Mobile Phone Platform

In recent years, the mobile industry in China has been rapidly developed by leaps and bounds,and the third generation of mobile communication is very near. And with the computing power of smart phones enhancing and the mobile data network bandwidth increasing, using smart phones as the core of the monitoring system becomes more and more possible. It can greatly facilitate the people to keep the safety of their homes.Therefore, at first this paper carries out a detailed analysis on many kinds of current monitoring systems, especially on their complexity and reliability. Then it divides them into time segmentation algorithm and space segmentation algorithm, and studies the domestic and international researches in this field from this point of view.On the basis of learning the characteristics and principles of existing algorithms on moving object detection, and combining the special characteristics of the mobile platform, this paper presents a rapid moving object detection algorithm. Basically when the background changes slowly, this algorithm uses the background upgrating method,and when there are dramatic light changes it uses the difference between the two consecutive frames to segment the object. At last, this paper designs a simple human detection algorithm in the indoor environment. This paper first designs many experiments on PC using C++ to solve a series of problems about moving object detection ,and then implements the algorithm on NOKIA N72. The results show that the algorithm can effectively detect moving objects in the indoor environment, and can effectively cope with the situation of light changes.