Image segmentation based on the minimum spanning tree with a novel weight

Image segmentation is a foundation for the image understanding and advanced application. However, the ill-posedness of image segmentation is a main challenge. To tackle this challenge, this paper proposes a novel minimum spanning tree (MST) based image segmentation algorithm. In the algorithm, we suggest a difference formula of RGB with respect to angular distance color as the weight of judge standard of the segmentation. To improve the stability and applicability of the segmentation, the judge standard not only considers the different sensitivities of the human eyes in regard to the variations of RGB values, but also involves the spatial distance and vector relationship between two pixels. The experiment verified the effectiveness of our proposed MST based image segmentation algorithm, the advantage of the algorithm was also illustrated by comparisons.