The Study Of The Code And Decode Algorithm About The Focus Video In Traffic Surveillance Environment

Traditional video codec algorithms has some shortage in traffic surveillance system which request the coded video data has high compression ratio and high quality in some area.This paper studies the codec algorithm about focus video in traffic surveillance environment.Firstly, it analyse the characters of the traffic surveillance environment,and presents a system of focus video codec.Then it analyse the characters of video image about vehicle license plate, presents an algorithm of license plate location based on the energy filter method and figure analyse method.Then it presents an algorithm of focus video codec based on DCT transform,which coding the video data of focus area only in frame without quantification and coding the video data out of focus area between frames.Validated by many experiments,the vehicle license plate location algorithm can meet the requirement of fast license plate location in traffic surveillance environment,and the video data coded by the video codec algorithm has high compression ratio and high quality in focus area.