{"references": ["G. Argenziano, H. P. Soyer, S. Chimenti, and R. T. et al., \"Dermoscopy\nof pigmented skin lesions: Results of a consensus meeting via the\nInternet,\" Journal of the American Academy of Dermatology, vol. 48,\npp. 679-693, 2003.", "P. Braun, H. Rabinovitz, M. Oliviero, A. Kopf, and J. Saurat, \"Dermoscopy\nof pigmented lesions,\" Journal of the American Academy of\nDermatology, vol. 52, no. 1, pp. 109-121, 2005.", "A. Perrinaud, O. Gaide, L. French, J.-H. Saurat, A. Marghoob, and\nR. Braun, \"Can automated dermoscopy image analysis instruments\nprovide added benefit for the dermatologist? A study comparing the\nresults of three systems,\" British Journal of Dermatology, vol. 157, pp.\n926-933, 2007.", "M. E. Celebi, H. Iyatomi, G. Schaefer, and W. V. Stoecker, \"Lesion border\ndetection in dermoscopy images,\" Computerized Medical Imaging\nand Graphics, vol. 33, no. 2, pp. 148-153, 2009.", "H. Iyatomi, H. Oka, M. E. Celebi, M. Hashimoto, M. Hagiwara,\nM. Tanaka, and K. Ogawa, \"An improved internet-based melanoma\nscreening system with dermatologist-like tumor area extraction algorithm,\"\nComputerized Medical Imaging and Graphics, vol. 32, no. 7,\npp. 566-579, 2008.", "M. Hintz-Madsen, L. K. Hansen, J. Larsen, and K. T. Drzewiecki, \"A\nprobabilistic neural network framework for the detection of malignant\nmelanoma.\" Artificial neural networks in cancer diagnosis. Prognosis\nand Patient Management, pp. 141-183, 2001.", "R. Melli, C. Grana, and R. Cucchiara, \"Comparison of color clustering\nalgorithms for segmentation of dermatological images,\" in SPIE Medical\nImaging, vol. 6144, 2006, pp. 3S1-3S9.", "G. Hance, S. Umbaugh, R. Moss, and W. V. Stoecker, \"Unsupervised\ncolor image segmentation:With application to skin tumor borders,\" IEEE\nEngineering in Medicine and Biology Magazine, vol. 15, pp. 104-111,\n1996.", "P. Schmid, \"Segmentation of digitized dermatoscopic images by twodimensional\ncolor clustering,\" IEEE Transactions on Medical Imaging,\nvol. 18, no. 2, pp. 164-171, 1999.\n[10] M. E. Celebi, Y. A. Aslandogan, W. V. Stoecker, H. Iyatomi, H. Oka,\nand X. Chen, \"Unsupervised border detection in dermoscopy images,\"\nSkin Research and Technology, vol. 13, pp. 454-462, 2007.\n[11] M. E. Celebi, H. A. Kingravi, H. Iyatomi, Y. A. Aslandogan, W. V.\nStoecker, R. H. Moss, J. M. Malters, J. M. Grichnik, A. A. Marghoob,\nH. S. Rabinovitz, and S. W. Menzies, \"Border detection in dermoscopy\nimages using statistical region merging,\" Skin Research and Technology,\nvol. 14, pp. 347-353, 2008.\n[12] T. Lee, , V. Ng, R. Gallagher, A. Coldman, and D. McLean, \"Dullrazor:\nA software approach to hair removal from images,\" Computers in\nBiology and Medicine, vol. 27, pp. 533-543, 1997.\n[13] L. Lucchese and S. Mitra, \"Color image segmentation: A state-of-theart\nsurvey,\" in Proceedings of Indian National Science Academy Part A,\nPINSA2001, 2001, pp. 207-221.\n[14] K. N. Plataniotis and A. N. Venetsanopoulos, Color Image Processing\nand Applications. Springer, 2000.\n[15] N. Otsu, \"A threshold selection method from gray-level histograms,\"\nIEEE Transactions on Systems, Man, and Cybernetics, vol. 9, no. 1, pp.\n62-66, 1979.\n[16] R. M. Haralick and L. G. Shapiro, Computer and Robot Vision.\nAddison-Wesley, 1992, vol. 1.\n[17] M. E. Celebi, G. Schaefer, H. Iyatomi, W. V. Stoecker, J. M. Malters,\nand J. M. Grichnik, \"An improved objective evaluation measure for\nborder detection in dermoscopy images,\" to appear in Skin Research\nand Technology.\n[18] T. Sorensen, \"A method of establishing groups of equal amplitude in\nplant sociology based on similarity of species and its application to\nanalyses of the vegetation on danish commons.\" Royal Danish Academy\nof Sciences and Letters, vol. 5, pp. 1-34, 1948.\n[19] J. Davis and M. Goadrich, \"The relationship between precision-recall\nand roc curves,\" in Proceeding of 23rd International Conference on\nMachine Learning (ICML), vol. 148, 2006, pp. 233-240."]}

Automatic segmentation of skin lesions is the first step towards development of a computer-aided diagnosis of melanoma. Although numerous segmentation methods have been developed, few studies have focused on determining the most discriminative and effective color space for melanoma application. This paper proposes a novel automatic segmentation algorithm using color space analysis and clustering-based histogram thresholding, which is able to determine the optimal color channel for segmentation of skin lesions. To demonstrate the validity of the algorithm, it is tested on a set of 30 high resolution dermoscopy images and a comprehensive evaluation of the results is provided, where borders manually drawn by four dermatologists, are compared to automated borders detected by the proposed algorithm. The evaluation is carried out by applying three previously used metrics of accuracy, sensitivity, and specificity and a new metric of similarity. Through ROC analysis and ranking the metrics, it is shown that the best results are obtained with the X and XoYoR color channels which results in an accuracy of approximately 97%. The proposed method is also compared with two state-ofthe- art skin lesion segmentation methods, which demonstrates the effectiveness and superiority of the proposed segmentation method.