Shadow Removal from Images Using the Concept of Chromaticity

Abstract: A shadow basically is a sort of image produced when light is blocked. A shadow generally takes up most of the space behind an opaque item with light right in front of it. Shadows often introduce errors in the performance of computer vision algorithms, such as object detection and tracking. For better performance, there is need to remove it to get shadow free image. This paper proposes a simple method to remove shadow from images. In this work, in order to objective evaluate the performance of shadow removal system, the statistical parameters were compared using information mean and average gradient. Average gradient can express the ability of small details and can be used to evaluate the clarity of the image, the greater its value, the more clear that the image. In addition, to some extent, the mean can be used to evaluate the image contrast.

Keywords: Shadow removal, Chromaticity, Mean, Average gradient, Statistical Parameters.