Automatic mobile segmentation of dermoscopy images using density based and fuzzy c-means clustering

Accurate identification and extraction of region of interest (ROI) in dermoscopy images play crucial role in diagnosis, and treatment of melanoma and other skin diseases. Human interpretation of dermoscopy images is not only tedious and time consuming task but also subjective. This fact has attracted numerous attentions for developing automated assessment tools. In this paper, we present lesion detection schemes in dermoscopy images on mobile platforms. The systems are based on density based clustering (DBSCAN) and fuzzy c-means (FCM) clustering and developed for Windows Phone and Android environments. We tested the systems on dermoscopy images and ROIs are successfully extracted. The proposed systems may improve the management of melanoma by providing automatic early monitoring of skin lesions that will assist clinical investigation.