Detection of Basal Cell Carcinoma (Bcc) using Surf

Abstract

Detection of Basal Cell Carcinoma (BCC) using Speeded-Up Robust Features (SURF).The
skin diseases, Melanocytic nevi(Nv),Melanoma(Mel),Benignkeratosislikelesions(
bkl),Basalcellcarcinoma(bcc),Actinickeratos es(alkies), Vascular lesions(vas),
Dermatofibroma(df). The Statistical Region Merging (SRM) technique is used to segment the
lesion. Based on scale invariant Under texture analysis, the SURF (Speeded up Robust
Features) technique is used for feature point detection and descriptions. For feature point
detection, SURF uses Hessian matrix approximation, and for feature descriptions, surf uses
haar-wavelet response. Color intensities are described using the lab colour space. The
detected patterns are then classified using a multi-SVM classifier.