Comparison of feature extraction algorithms for mammography images

Mammography image classification is a very important research field due to its domain of implementation. The aim of this paper is to compare feature extraction methods and to test them on a variety of classifiers. Five feature extraction methods were used: LBP, GLDM, GLRLM, Haralick and Gabor texture features. Three classification algorithms were used during the experiments, namely, support vector machines, k-nearest neighbor and c4.5 algorithm. The experiments were conducted on the MIAS database. The results show that GLDM is the most appropriate feature extraction method for images from this database.