Computing Orthogonal Moments in Biomedical Imaging

A novel methodology that permits the computation of the orthogonal moments of biomedical images in high rates is presented in this paper. The proposed methodology is making use of the inherent property the biomedical images have, the limited number of intensity values in the gray scale. This morphological characteristic facilitates the application of the ISR (Image Slice Representation) algorithm, proposed by the authors, to decompose a gray-scale image to a finite number of nonoverlapping slices. The image´s pixels belonging to a specific slice can be organized as homogenous rectangular blocks in order to simplify the moments´ computation formulas. Appropriate experiments on biomedical images have shown that the proposed technique can accelerate the moment´s computation significantly, by reducing the overall time about 50%.