Biomedical images enhancement based on the properties of morphological spectra

The method enhancing distinctiveness of the micro-morphological structures, developed using the properties of morphological spectra of their monochromatic 2D images, is presented and its effects on the bone section image are statistically compared with enhancements by Sobel, Roberts and Laplace high-pass filters. Comparison of different filters based on statistical parameters of the classes of selected image details is presented. The preferable method for choosing filtering weight coefficients is described and illustrated by an example of processing an electron-microscope image of a biotechnological specimen. The applicability of this approach and possible development directions are discussed.