Fuzzy treatment of surroundedness based object detection

In this paper, we present a real-time salient object detection system based on the Fuzzy Treatment of Surroundedness. Since background regions are typically connected to the image boundaries, salient objects can be extracted by computing the distances to the boundaries. However, it is a challenging problem to measure the image boundary connectivity efficiently. To improve the processing efficiency, super-pixel segmentation will be utilized in our method. Firstly, SLIC is employed to segment the image into super-pixels, and the distance of each super pixel to the boundary is calculated and transformed to generated the minimum tree, which can extract the salient object preliminarily. Then, the contour confidence map is obtained by a fast contour detection method. Finally, the unfixed contours are fuzzed by the fuzzy color difference histogram to generate a saliency map which can be incorporated into the final saliency map. Extensive experiments on benchmark data sets validate the effectiveness and superiority of the proposed approach over state-of-the-art methods.