Caracterização de textura em imagens de lesões de pele por máquina de vetor de suporte

Abstract

Due to the increased incidence of skin cancer, computational methods using intelligent systems have been developed to aid dermatologists in the diagnosis of skin lesions. This paper proposes a method to classify the texture, considering that it is an important feature in identification of lesions. For this is defined a feature vector with the fractal dimension of images through the box-counting method (BCM), which were used by SVM to classify the texture of the lesions in non-irregular or irregular, where it obtained 72.84% of accuracy.