Video syntactic coding for data compression

In this paper we present our new approach to video data preparation for a data compression enhancement named video data syntactic coding. Video data standard encoders MPEG, H.26x use data preparation (preprocessing) based on either data spatial and temporal filtering or on video sequence complexity analysis (by auxiliary encoder use) that is exploited for video filtering. However, preprocessing efficiency is only 10% up to 15%. Our video data syntactic coding presented here is based on both video content complexity assessment in accordance with the named standard encoding algorithm (MC-DPCM) and 'data filtering adopted to the video complexity and a digital visual perception. The complexity is estimated by data entropy calculation based on averaging of histogram frame-to frame changes. In addition, the histogram change is used for detection of new details appearance and their inframe position determination. The video data model is based on a detail visual perception, and the above video content analysis results provide effective video data prefiltering. The developed video syntactic coding used together with H.264 standard encoders presents up to 25% data compression enhancements.