Remote Sensing Image Segmentation Based on Region-Split and Graph Cut

According to the features of HVS, the algorithm uses region-split method to segment the remote sensing image into a large number of small regions. By integrating gray feature and spatial location of each region, NC is used to segment the image among regions from global view, by which the final segmented image can be generated. Experimental results show that comparing with the traditional NC, operating speed is significantly improved as getting close segmentation quality, and this is a kind of effective method of image segmentation.

[1]  A.A. Alatan,et al.  Region-based image segmentation via graph cuts , 2008, 2008 IEEE 16th Signal Processing, Communication and Applications Conference.

[2]  Wang Ji-cheng,et al.  Multi-resolution image segmentation based on Normalized Cut , 2009 .

[3]  Tian Zheng,et al.  Optimum cut-based clustering , 2007, Signal Process..

[4]  Hai Jin,et al.  Color Image Segmentation Based on Mean Shift and Normalized Cuts , 2007, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[5]  Keechul Jung,et al.  Graph Cuts-Based Automatic Color Image Segmentation , 2008, 2008 Digital Image Computing: Techniques and Applications.

[6]  Avinash C. Kak,et al.  On combining graph-partitioning with non-parametric clustering for image segmentation , 2004, Comput. Vis. Image Underst..

[7]  Keechul Jung,et al.  Graph Cut-based Automatic Color Image Segmentation using Mean Shift Analysis , 2009 .