Reduction of color variability in color image segmentation

Processes of color image segmentation are crucial for the image interpretation. Thus, many techniques give very good results. However it is relatively difficult to standardize the segmentation algorithm by considering the variability between images: a limit of segmentation performance. In this paper, our goal is to present a method of inter-images variability reduction, by using the colorimetric context in the image. For this purpose, we propose a method of segmentation based on transformation and conversion in various colorimetric spaces.