A network intelligent video analysis system based on multimedia DSP

Abstract

In this paper, we designed a hardware and software system using TI's new multimedia technology, Davinci based on multimedia DSP. The objective is to develop a practical Internet based intelligent video analysis product. This system can perform real-time processing of the video data from IP camera. And at the same time, it executes intelligent video analysis algorithms, provides users quick, accurate and intelligent video analysis service. In this system, we implement a multiple objects tracking approach for objects labeling being able to preserve objects identity even when an overlapping occurs on the image plane between different objects. And we use histogram matching techniques for real-time objects tracking under heavy occlusions and disappearance. We also compare our improved method with another objects tracking algorithm based on Mean Shift and Kalman filter. Experimental results show that the system is stable and fits real-time requirement, and the performance of our method is better.