An approach for Shadow Detection and Removal based on Multiple Light Sources

Abstract

Shadows in images are essential but sometimes unwanted as they can decline the result of computer vision algorithms. A shadow is obtained by the interaction of light with objects in an image surface. Shadows may letdown the image analysis processes and also cause a poor quality of information which in turn leads to problems in execution of algorithms. In this paper, a method has been proposed to detect and remove the shadows where multiple sources of light is been estimated, as we can take an example of playground stadium where multiple floodlights are fixed, multiple shadows can be observed originating from each of the targets. To successfully track individual target, it is essential to achieve an accurate image of the foreground. Also, an effort has been done to list some of the very crucial techniques related to shadow detection and removal. Many times, the shadow of the background information is merged with the foreground object and makes the process more complex.
DOI: 10.17762/ijritcc2321-8169.1505174