Oliveira, Roberta B. [UNESP]Caldas Jr., Carlos Roberto D. [UNESP]Pereira, Aledir S. [UNESP]Guido, Rodrigo C. [UNESP]Araujo, Alex F. deTavares, João Manuel R. S.Rossetti, Ricardo B.2014-05-272014-05-272012-11-19Iberian Conference on Information Systems and Technologies, CISTI. Information Systems and Technologies. New York: IEEE, p. 2, 2012.2166-07272166-0735http://hdl.handle.net/11449/73741Due to the increased incidence of skin cancer, computational methods based on intelligent approaches have been developed to aid dermatologists in the diagnosis of skin lesions. This paper proposes a method to classify texture in images, since it is an important feature for the successfully identification of skin lesions. For this is defined a feature vector, with the fractal dimension of images through the box-counting method (BCM), which is used with a SVM to classify the texture of the lesions in to non-irregular or irregular. With the proposed solution, we could obtain an accuracy of 72.84%. © 2012 AISTI.porbox-counting methodfractal dimensionintelligent systemmachine learningsupport vector machineBox-counting methodFeature vectorsSkin cancersSkin lesionDermatologyFractal dimensionImage retrievalInformation systemsIntelligent systemsLearning systemsSupport vector machinesTexturesImage textureTrabalho apresentado em eventoWOS:000319285900159Acesso aberto2-s2.0-8486903870465420862268080670000-0002-0924-8024