Illustration of Redundancy Removal in Run Length Coding and Its Implication in Surveillance Video

Abstract

Surveillance video and its implication in image processing have paved the newlinefootprint of enormous changes in the vehicular ad-hoc networks and traffic newlinemonitoring. Obtaining the characteristics and the parameters of the video newlinefrom a signal is a challenging issue. The parameters such as pixel connectivity, newlinedistance metrics, redundancy, and compression are the key aspects. In this research the work uses picture metrics in both Redundancy Removal newlinebased Enhanced Run Length Coding Scheme (RR-ERLC) and (RR-PRLC) newlineRedundancy removal based Proficient Run Length Coding. The proposed newlinework uses vector processing in colour domain followed by histogram analysis newlineusing compression and segmentation. Initially, the RR-ERLC algorithm comprises of encoding the video frames by newlineremoving the redundancies using the texture information which is dependent newlineon pixel value. The workflow of the algorithm involves the conversion of the newlineinput video into frames which are subsequently processed with color space on newlineindividual layers. Then based upon their similarity measure the redundancy is newlineremoved. The metrics of similarity used relies on Euclidean distance. The newlineprotocol RR-ERLC achieves better performance metrics than the conventional newlinetechniques. Similar to RR-ERLC algorithm, RR-PRLC algorithm encodes the frames newlineinitially by removing the redundant component and the resultant video newlinesequence is converted to individual frames. Then processed with frames newlinewherein the color space for individual layers has been incorporated with newlinesimilarity measure of redundancy removal using cosine angle distance which newlineis correlation based. It achieves better compression metrics when compared newlinewith the conventional techniques. Regression and Wilcoxon signed rank test are the two standard forms of newlinestatistical analysis carried out in the proposed model with the data obtained newlinefrom the Matlab simulation results. Regression signifies the relationship newlinebetween dependent variable PSNR and two independent variables such as newlineexecution time and video size.