| With the development of modern science and technology, especially thepromotion of computer processing and broadband network and the improvement ofpeople’s safety awareness, video surveillance technology is more and more widelyused in various fields. However, the traditional video surveillance system still has alot of limitations, the main problems are as follows: one is the traditional videosurveillance adopts the wired transmission leads t to the front-end wiring difficultiesand lack of flexibility; The second is network transmission rate slow, make themonitoring terminal for receiving images difficult to achieve real-time and imageclarity lacking; Three is the surveillance front intelligence poor, lead to need to spenda lot of manpower guard terminal screen, and it is difficult to do real time warning.To solve these problems, we propose a video surveillance system based onHi3510processor as the core and Huawei EM770W3G wireless module. Combiningthe advantages of3G network and H.264video compression, this system can improvethe quality of the entire network of video surveillance.In this paper we designed the intelligent front-end system based on the Hi3510as the video processor and EM770W3G wireless module. Then completed the wholedesign of intelligent system and optimized the hardware of power and memory parts.We have completed the software environment in the hardware platform, includingBoot and linux kernel transplantation, startup parameters configuration and the crosscompile environment installation. Further more, we have designed the intelligentarithmetic in the front-end, is that licence plate image preprocessing, licence platearea location and character segmention. And then such feature data sent to thebackend server through3G wireless. The backend server uses the BP neural networkto realize character recognition after receiving such data. Finally, the recognitionresult is sent to the user terminal to complete the entire video surveillance algorithmdesign. |
PDF Full Text Request