A fuzzy approach to object segmentation using depth image

In this paper, we proposed a fuzzy approach to segmenting objects using spatial location information from depth image. The optical and location information are combined by proposed fuzzy rules tale which base on K-means clustering. The segmentation of our framework is an efficient graph-based image segmentation algorithm. This framework combined obvious changes in color and physical location to segment reality scenes into uniform regions. The performance of our proposed framework is demonstrated in a series of reality-scene images using experimental data from the Middlebury stereo image data.