Research On Surveillance Videos Comparession Algorithm Based On X264

With the wide application of surveillance video in scientific research,big data analysis,transportation,public safety and other fields,the amount of data for surveillance video has increased dramatically.The growth rate of surveillance video compression rate has lagged far behind the increase of video data.Compared to general video,surveillance video is captured by a surveillance camera deployed in a fixed location,which contains a large amount of redundant background information.By removing this redundant information,the compression ratio of the surveillance video can be further improved.The research on coding technology for surveillance video features has become a research hotspot of experts and scholars at home and abroad.On the basis of investigating the existing surveillance video coding schemes,the background-based scheme is deeply studied.The research content of this paper mainly includes:Firstly,real-time surveillance video coding based on dual background coding scheme.In order to achieve the video acquisition speed to meet the design requirements,the coding algorithm used in the video surveillance system needs to meet the real-time requirements.Although the dual background coding scheme can effectively compress the surveillance video,it does not consider the real-time problem in the design and implementation process.In this paper,a fast algorithm for dual-background coding scheme is proposed,which can optimize the scheme in real-time from two aspects: on the one hand,when classifying blocks,it reduces the calculation times of the maximum absolute difference;on the other hand,only one block classification is carried out in the iteration process of background updating.In addition,the algorithm is implemented in x264 to meet the real-time requirements.The experimental results show that the coding speed of the fast algorithm is slightly lower than that of the original x264 algorithm,but its compression rate is increased by 12.8%.Compared with the original dual-background coding scheme,the coding frame rate of the proposed algorithm is increased by 21% when the compression performance is comparable.Secondly,surveillance video coding based on adaptive background updating.At present,most of the existing Background-Based coding schemes only perform background coding and reconstruction once in the initial stage to avoid the increase of bit rate.However,with the passage of time,the background data will change greatly,which leads to the background can not provide effective reference in the subsequent coding process.In order to solve the above problems,this paper proposes an efficient background updating scheme: adaptive background updating scheme.When updating the background,the scheme classifies the updated background macroblocks,and adaptively chooses different coding schemes according to the classification results to realize the adaptive updating of the background.The experimental results show that the compression rate of the proposed scheme is improved by 1.7% compared with that of the non-updating background.