Labeling watershed algorithm based on morphological reconstruction in color space

Abstract

The watershed algorithm is one of the most powerful morphological tools for image segmentation, but the traditional watershed algorithm always exists serious over-segmentation, and can be easily affected by specularities and shadows. A new image segmentation based on reconstruction labeling watershed algorithm in color space is proposed to solve these problems. First, the RGB color image is converted into a new color space image, and the gradients are calculated by extracting the directions which won't be affected by specularities and shadows. Second, object regions are extracted to compose a binary marked image through the use of morphological opening and closing reconstruction, and gradient image is introduced for substituting the marked image. Finally, the watershed transformation is carried out in the modified gradient image. The new algorithm not only can overcome over-segmentation produced by texture details and noises, but also suppress over-segmentation caused by specularities and shadows. Furthermore, the segmentation algorithm is used in the primitive gradient image instead of filtered and simplified images, so that the edge information of the object can be retained as much as possible. Theory analysis and experimental results have shown that our method has obtained obvious improvement.