二维直方图区域斜分Otsu阈值分割的快速迭代算法

A Fast Iterative Algorithm of the Otsu Threshold Based on Two-Dimensional Histogram Oblique Segmentation

鉴于现有的基于二维直方图区域直分的阈值选取方法中存在明显的错分,提出了一种新的二维直方图区域斜分方法,导出了基于二维直方图区域斜分的Otsu法的快速迭代算法。在实验结果中给出了分割结果和运行时间,并与基于二维直方图直分的Otsu原始算法及其他两种快速算法进行了比较。结果表明所提出的快速迭代算法,使分割后的图像内部区域均匀,边界形状准确,有稳健的抗噪性,同时运行时间大幅减少。

In view of the obvious wrong segmentation in the existing threshold selection method based on two-dimensional histogram vertical segmentation, a new two-dimensional histogram oblique segmentation method is proposed. The formulae and its fast iterative algorithm are deduced. The thresholded images and processing time are given by the experiment results, which are compared with original Otsu algorithm and other two fast algorithms based on two-dimensional histogram vertical segmentation. The results show that this new algorithm makes inner part uniform, edge accurate, better capability to anti-noise, and running time reduced significantly.