MEDICAL IMAGES ENHANCEMENT USING MULTISCALE TOP-HAT TRANSFORM BY MORPHOLOGICAL RECONSTRUCTION

Medical images help doctors provide more rapid and ecient diagnosis to their patients. The quality of the medical image directly influences its diagnostic. However, at the time of acquisition of the medical images, these usually present degradations such as poor details or low contrast. This work presents an algorithm that allows to improve the contrast and the details of the medical images. This algorithm is based on the multiscale top-hat transform by reconstruction. The proposal extracts multiple bright and dark features from the medical image. To enhance the medical image, the maximum bright regions are added and the maximum dark regions are subtracted. To quantify the performance of the proposed method, 100 medical images from a public database were used. The experiments show that the proposal enhances the contrast and detail of the medical image, and introduces less distortion into the enhancement process than state-of the-art algorithms.