Shadow Detection and Removal for Traffic Surveillance System

Shadow detection and removal is an important task when dealing with outdoor images. Detection of foreground objects from traffic video sequences is one of the most important parts in many computer vision applications. In traffic surveillance system, detecting vehicles require an accurate method to detect foreground objects. Background subtraction is a common method for detecting foreground objects but it is difficult to obtain pure vehicles because of shadow which is often mistaken as part of vehicles. Shadow removal is identified as a critical step for improving the accuracy of vehicle detection and tracking. This paper proposed to detect shadows by analysing the image pixel values of shadows and vehicles based on HSV color spaces and remove shadows by replacing the pixels of background image at the detected shadow region. Aye Aye Win "Shadow Detection and Removal for Traffic Surveillance System" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-3 | Issue-5 , August 2019, URL: https://www.ijtsrd.com/papers/ijtsrd26732.pdf