Melasma Image Segmentation Using Extreme Learning Machine

This paper introduces an image segmentation method based on the extreme learning machine (ELM) to detect melasma in human face images. In this work, skin texture features are extracted and fed into ELM classifier for segmentation. The results obtained by the proposed method show better segmentation performance visually than that obtained by a conventional threshold based image segmentation method. The proposed method could potentially lead to the development of a more reliable computerized melasma severity assessment system.