Many real-world and man-made objects are line symmetry. To detection the line-symmetry objects from an image, in this paper, a new cluster validity measure which adopts a non-metric distance measure based on the idea of "line symmetry" is presented. The thresholding technique is first applied to extract the objects from the original image; and the object pixels are transferred to be the data patterns. Then the fuzzy clustering algorithm is applied to label the object pixels; and the proposed validity measure is used in determining the number of objects. Simulation results are used to illustrate the performance of the proposed measure.
2013 2nd International Conference on Information Computer Application (ICICA 2013), 3p.