Multiple vehicle detection and segmentation

摘要: Vision based system are widely used in the field of Intelligent Transportation System (ITS) to extract a large amount information analyze traffic scenes. Previously, this burdensome task was performed by human operator monitoring centre. Nevertheless, increasing number vehicles on road as well significant increase cameras dictated need for surveillance systems. The research undertaken thesis is mainly concentrated developing multiple vehicle detection and segmentation focusing through Closed Circuit Television (CCTV) video. proposed able automatically segment extracted from heavy scene. In work, optical flow estimation alongside with blob analysis technique order detect moving vehicle. Since there no reference background image, distinguish between video scene Prior segmentation, will compute area interest region corresponding which be create bounding box that particular Experimental validation algorithm demonstrated various set