关键词: 肠癌, 灰度行程长度, 方向梯度直方图, 最大相关最小冗余

Abstract: The recognition rate of colon cancer histopathological images using single feature set is difficult to improve. In this paper, a colon biopsy image classification system based on HOG-GLRLM features is proposed. Considering the rich texture and edge information of the histopathological image, two feature types, namely gray level run length matrix and histogram of oriented gradients based features have been proposed. Moreover, the minimum Redundancy Maximum Relevance（mRMR）feature selection method has been employed to select meaningful features from individual and hybrid feature sets. The outcome of this experiment indicates that the algorithm achieves good results in classification accuracy.

Key words: colon cancer, gray level run length matrix, histogram of oriented gradient, minimum redundancy maximum relevance