Design and implementation of Optimization based video compression Techniques for multimedia applications

Abstract

Nowadays, video compression plays an important role in every newlinewalk of life. Video compression plays a vital role to recreate a video without newlineaffecting the quality with reduced size. The video transmission area endures newlineseveral failures owing to the limited number of cutting-edge techniques to newlinestore large sized videos. Extracting the semantic data from a large amount of newlinevideo-based applications is the necessary feature. The existing methodologies newlineare insufficient and need a high amount of optimization cost. The videos newlinebeing searchable on the web also have drastically increased. Two video newlinecompression approaches are proposed in this thesis to mitigate the above said newlinedrawbacks. newlineThe first proposed methodology is utilized to improve the video newlinecompression and increase the quality of the video. Initially, the video newlinesequences are converted into a number of video frames. This method newlineexplores by utilizing a diamond search pattern with a secure number of newlinesearch points at every lattice. It uses the association of bordering pixels newlinewithin the frame. So as to diminish the computative intricacy, this newlinemethodology uses a hybrid diamond search and teaching learning based newlineoptimization algorithm. This method additionally decreases the newlinecomputational unpredictability of the block matching procedure. The image newlinequality has been confirmed through 3D reconstruction using structured light newlinetechniques. newline