The Gray-Level Image Thresholding Method Based on the Optimal One-Dimensional Projection of Two-Dimensional Histogram
基于二维灰度直方图最佳一维投影的图像分割方法

The features of the clusters on 2-dimensional histogram of images are analyzed firstly in this paper. It is shown that there is an optimal 1-dimensional projection of the 2-dimensional histogram. If we use this optimal projection to replace the 1dimensional histogram in the threshold division of an image, the probability of error caused by noise can be reduced greatly. Theoretic analysis and experiments show that this new method is much better than the methods that merely make use of histograms of the image.