Color image segmentation using edge detection and seeded region growing approach for CIELab and HSV color spaces

Image segmentation is one of the important tasks in image processing. The paper presents a simple and efficient method for segmentation of color images. Segmentation is performed for CIELab and HSV color spaces. The paper presents an integrated approach which uses edge detection and region growing for segmentation of color image. The proposed method involves automatic seed selection based on edge information. According to the selected seed pixels, region growing based on color similarity and connectivity of pixels is performed. This process is followed by region merging based on similarity and size to reduce over-segmentation. The proposed method provides efficient and effective results for many color images.