k-Gabor: A new feature extraction method for medical images providing internal analysis

This paper proposes the k-Gabor method, a new image feature extractor that captures texture information from medical image regions without a costly segmentation usually associated to texture extractors. It employs Gabor filters, thus, the k-Gabor method can quantify texture information from specific regions, tissues and internal structures of the images providing a succint representation for a richer image analysis. The feature vectors generated describe the images more precisely than other methods from the literature, as shown in the experiments. Besides providing meaningful information from the images, the cost to obtain it is very small, since the total time to extract the k-Gabor features was always only fractions of seconds.