Exploration and improvement of Ostu threshold segmentation algorithm

As a classical image segmentation method, Ostu threshold algorithm had been applied widely in image processing. This paper had a comparison of the following two-dimensional Ostu thresholding method. One-dimensional Ostu method considered only grayscale information of the pixel, 2D Ostu algorithm considered both the gray value of a pixel and the average gray value of its neighborhood, thus is more robust to noise. But because calculating of the two-dimensional Ostu threshold method demands a long time, so restricted its use, by constructing look-up tables recursively, its fast algorithm reduces its complexity from O(L⁴) to O(L²), based on the decomposition of 2D Ostu adaptive algorithm. When the hypothesis of original 2D Ostu algorithm holds, the method can get the same segmentation threshold as the original two-dimensional method, while the computational complexity is reduced further. In the paper, the algorithm is improved on the basis of the original one, one-dimensional threshold average decomposition of two-dimensional Ostu algorithm is proposed. The algorithm not only maintained advantages of less time and smaller space of calculating the threshold of 2D Ostu adaptive algorithm, of the threshold shorter and space advantages of a smaller, but also better segmentation results can be found to get from experiment.