Segmentacão de imagens de câncer de pele utilizando superpixels

Abstract

Nowadays, there is a wide variety of computational techniques for the processing of digital images that can be applied to a wide range of fields, such as industry, commerce, education and the field of medicine. Highlighting with great importance, the field of medicine, which through the analysis of digital images, some diseases that are difficult to detect and identify can be diagnosed and treated properly. Among the diagnosed diseases, the skin lesions are the most disturbing representative of cutaneous melanoma. Because melanoma presents difficult identification when compared to other skin lesions or nevus (moles), technologies based on digital imaging techniques may be used to characterize the disease. In this work a digital image segmentation method is applied for the detection of melanoma skin cancer, using the superpixel technique, based on the Simple Linear Iterative Clustering - SLIC algorithm. The superpixel technique presents, as a basic characteristic, the grouping of pixels in regions significantly similar, in which the characteristics of the study images will be analyzed, which will facilitate the detection of skin lesions.