Hybrid cosine and Radon transform-based processing for digital mammogram feature extraction and classification with SVM.

A new methodology to automatically extract features from mammograms and classify them is presented. It relies on a hybrid processing system that sequentially uses the discrete cosine transform (DCT) to obtain the high frequency component of the mammogram and then applies the Radon transform to the obtained DCT image in order to extract its directional features. The features are subsequently fed to a support vector machine for classification. The approach was tested on a database of one hundred images and shows improved classification accuracy in comparison to using the discrete cosine transform or the Radon transform alone, as done in others works.