Video Compression Based on Motion Compensation and Contourlet Transform

Abstract

The extensive use of multimedia technology by different communication devices increases the data flow over the transmission media; The researchers try to find an efficient way to compress data to preserve a space in storage or to ease transmitting it over media. Video compression techniques are developed in many areas such as (educational and medical, …etc.). Contourlet transform (CT) is used in this paper with a motion compensation technique (MC) for video compression purposes. CT has a good ability to retain image details because it avoids downsampling to a high-frequency sub-band, and the coefficients are also very sparse, by exploiting these features CT is used along with MC which overcomes the disadvantage of not synthesizing the current frame from the previous frame accurately such as in simple frame differencing technique. In this paper CT is applied to two frame’s prediction techniques of the video: simple frame differencing (classical method) and motion compensation. The results show the excellence of CT with MC, where the compression efficiency is improved. Results have obtained with these techniques in terms of Peak Signal to Noise Ratio (PSNR), Compression Ratio (CR) and a Bit Per Pixel (BPP).