earticle

영어

Image Segmentation is to simplify and/or change the representation of an image into something that is more meaningful and easier to analyze. Pre-processing of the image makes it more clear and visible, while parallelizing of the algorithm optimizes the speed at which the image is processed. This paper presents the performance analysis of parallelizing the video frame segmentation using region growing technique on varying of processors. Region growing Technique is also classified as pixel-based image segmentation method since it involves the selection of initial seed points where determining of whether the pixel neighbors should be added to the region or not is done. The process is iteratively carried out and the region is correspondingly grown for each image. Video is divided into frames, which are passed to different cores. The processing of frames is carried out simultaneously in cores, hence parallelizing the technique. Time analysis for serial and parallel implementation is done. Analysis of time required for processing the frames on different number of cores is carried out.