A comparison of filtering and enhancement methods in malignant melanoma images

Abstract

Skin cancer defined as the uncontrolled growth of cells in the skin. Malignant melanoma is a type of skin cancer formed due to the abnormal growth of the melanocytes that produce melanin pigments. The main aim is to analyze the preprocessing techniques for enhancement of the malignant melanoma images from the authorized PH2 database. Image enhancement techniques such as histogram equalization, adaptive histogram equalization, contrast stretching, and unsharp masking for the adjustment of the pixels in the image and the filtering technique in which different filters such as median, adaptive median, Gaussian and wiener are used for the removal of the unwanted details from the image are comparatively studied with the help of the histogram and peak signal to noise ratio. Histogram defines the pixel variations or alterations in the resultant image after pre-processing. Statistical parameters like mean and standard deviation are derived for the PSNR values of the enhancement and filtering methods. From the results obtained, a comparative study of the enhancement and filtering methods in malignant melanoma images is performed.