The diagnosis of brain white matter lesions is an important aspect of medical research.White matter segmentation of brain MR image plays a key role in diagnosis,and the accuracy of segmentation directly influences the following analysis and diagnostic processing.This paper proposes an algorithm for white matter segmentation of brain MR images based on local Walsh transform and non-negative matrix factorization.The algorithm first operates a local Walsh transform on the pixels of a brain image,and then selects the features with good discrimination to get the feature matrix,and finally gets the results of segmentation by non-negative matrix factorization on the feature matrix.The experiments show that the proposed method is simple with high precision,and can gain reasonable results.