Image thresholding based on maximum mutual information
Lulu Fang, Yaobin Zou, Fangmin Dong, Shuifa Sun, Bangjun Lei · 2014
Thresholding segmentation is a critical preprocessing step on many image processing applications. However, most of the existing thresholding methods can only deal with an image with some special histogram patterns. To automatically determine the robust and optimum thresholds for the images with various histogram patterns, this paper proposes a new thresholding segmentation method based on maximum mutual information. The optimal threshold value is determined by maximizing the mutual information between a series of binary images and a reference image. The reference image is generated by a multi-scale gradient multiplication transformation on the original gray level image. Experiments on synthetic images and real images show the effectiveness and the accuracy of the proposed segmentation method.