Segmentation of Pedestrian Video Using Thresholding Algorithm and its Parameter Analysis

Abstract

Video segmentation is the field of video processing that deals with dividing a video into useful segments. Video segmentation plays an important role in real-time encoding systems. This paper deals with analyzing and segmenting a pedestrian video using the thresholding algorithm. By using the k-means algorithm, which uses clustering technique, the parameters of the pedestrian video such as direction of movement of the pedestrian and the distance of the pedestrian from the capturing device is calculated and the result is displayed. Furthermore, this can be implemented in real-time systems, which is used for various applications.