Robust extraction of moving objects based on hue and hue gradient

Abstract

This paper presents a new method for robustly extracting moving objects in an environment with varying illuminations. The proposed method applies background subtraction scheme based on hue and hue gradient to minimize the effect of the illumination change. First, we train the background images in the HSI color space and build the Gaussian background model with respect to the hue and hue gradient. Next, image subtraction is performed between the trained background image and the current input image based on the Gaussian background model. Finally, the morphological operations are applied to remove the background noise. In this paper, we compare the previous subtraction schemes to our method applied to both the hand and body tracking in order to prove the robustness of the proposed method in sudden illumination changes. © Springer-Verlag Berlin Heidelberg 2007.