Improved algorithm of image segmentation based on maximum classes square error

摘要

In this paper, a binary search algorithm is used to search optimal threshold value for image segmentation. It overcomes the shortcoming of huge calculation by using Otsu algorithm to search optimal threshold value traversing the whole grayscale, enable which to be applied in some occasions requiring for high real time. Experimental results show that this method can find reliable optimal threshold value, conduct image segmentation, and improve the processing speed greatly. The improved Otsu algorithm becomes more practical and wider application.