Foreground and background interactive segmentation method of large-size color image and system thereof

The invention relates to a foreground and background interactive segmentation method of a large-size color image and a system thereof, belonging to the image segmentation technology field. The method comprises the following steps: selecting a plurality of foreground reference points and background reference points in a color image; segmenting the color image into a plurality of image blocks with a set size, carrying out area growth on each image block, and taking each grown area as a characteristic point, wherein, color of the characteristic point is a mean value of all pixel point colors of the area; calculating similarities between each characteristic point and foreground, background and neighborhood; according to the similarities, establishing a digraph, segmenting a pixel point belonging to a candidate area in the digraph, and generating a candidate area border. According to the method and the system in the invention, the large-size color image can be segmented, segmentation speed is high, and an effect is good.

[1]  Olga Veksler,et al.  Graph Cuts in Vision and Graphics: Theories and Applications , 2006, Handbook of Mathematical Models in Computer Vision.

[2]  Marie-Pierre Jolly,et al.  Interactive graph cuts for optimal boundary & region segmentation of objects in N-D images , 2001, Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.

[3]  Jian Sun,et al.  Lazy snapping , 2004, SIGGRAPH 2004.