Color image segmentation using steerable filters

In this paper we propose a color image segmentation algorithm based on steerable filters. By their directional response, this type of filters has been previously used for performing edge-based image segmentation. Our approach to image segmentation is novel, as it uses steerable filters to determine uniform regions. In a first step, uniform regions are found separately on intensity and hue-saturation components. The segmentation results on the two components are then combined. The obtained segmentation is finally refined using classical techniques.