An approach of color image segmentation based on fuzzy clustering

摘要

Considering the disadvantage that the classical 2D Otsu threshold method is time-consuming, this paper proposed an improved algorithm. By calculating two 1D Otsu threshold method instead of the traditional 2D Otsu threshold method, it reduced the complexity of the algorithm. Simultaneously in order to improve the segmentation performance, this algorithm adopted fuzzy C-means method. The experimental results show that this improved algorithm can take the advantages of both. It is better than the traditional 2D Otsu not only in the computation time, but also in the quality.