Classification of melanoma based on feature similarity measurement for codebook learning in the bag-of-features model.

•We propose a codebook learning algorithm using feature similarity measurement (FSM).•We develop a novel melanoma classification method based on the FSM algorithm.•The RGB color histogram and SIFT descriptors are combined to describe skin lesions.•The evaluation is performed on both the dermoscopy and non-dermoscopic datasets.•Results show that the proposed method outperforms some state-of-the-art approaches.