Melanoma recognition using extended set of descriptors and classifiers

Abstract

The paper presents a novel method of melanoma recognition on the basis of dermoscopic images. We use color images of skin lesions, advanced image processing, and different classifiers to distinguish melanoma from the other non-melanoma lesions. Different families of descriptors are used for generation of the image diagnostic features for final pattern recognition. To increase the efficiency of the system, we apply different selection procedures to find the best set of features and different solutions of classifier. The numerical results concerning the accuracy of the proposed recognition system have confirmed good accuracy of the proposed method and high sensitivity in melanoma recognition.

An Overview of Classification Methods From Dermoscopy Images in Skin Lesion Diagnostic

“…Nowadays, various modifications are becoming more and more common. They are used to classify images [15,17,23,31,33].…”

Segmentation of skin lesion using Cohen–Daubechies–Feauveau biorthogonal wavelet

“…A fully automated segmentation method is based on threshold for dermoscopic images is described in Kruk et al. ( 2015 ). This method utilizes histogram based thresholding on all three RGB colors.…”

Bagged textural and color features for melanoma skin cancer detection in dermoscopic and standard images

“…In (Vasconcelos et al, 2015), color features have been derived from the ABCD rule where the authors proposed a clustering approach to adjust the system to different datasets and image types. In (Kruk et al, 2015), different texture and statistical features were adopted, including the numerical descriptors based on the Kolmogorov-Smirnov (KS) statistical distance, the classical Haralick descriptors and fractal texture analysis-based descriptors. In (Giotis et al, 2015), physician annotations for skin lesions, referred to as visual diagnostic attributes, were combined with lesion color and lesion texture for melanoma skin detection in non-dermoscopic images.…”