A Shadow removal method for a mobile robot localization using external surveillance cameras

Abstract

Shadow detection and removal is important to deal with vision based mobile robot localization. The shadow cast by objects within environment which a mobile robot moves can lead to inaccurate object feature extraction and an erroneous scene analysis. To remove the shadow of an object, the difference image between the original and background image is usually used. However, as the shadow area caused by the light source comes out, it should be removed to obtain the accurate object image. In this paper, we used the surveillance cameras installed at the room ceiling for vision based robot localization. Time-consuming algorithm is difficult applied for the mobile robot localization. Therefore, we proposed a fast shadow removal algorithm based the HSV color space. Generally, the shadow removal method by the HSV color space is speedy but not enough to find shadow property compared with other methods. Shadow mask and HSV color space are used together to detect the shadow region. Experimental results show that the proposed method is simple to understand, can detect and remove shadow effectively for a mobile robot localization. © 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license.