Texture analysis for classification of cervix lesions

This paper presents a generalized statistical texture analysis technique fo r characterizing and recognizing typical, diagnostically most important, va scular patterns relating to cervical lesions from colposcopic images. The c ontributions of the research include: 1) the introduction of a generalized texture analysis technique based on the combination of the conventional sta tistical and structural textural analysis approaches by using a statistical description of geometric primitives; 2) the introduction of a set of textu ral measures that capture the specific characteristics of cervical textures as perceived by human. Experimental study with real images demonstrated th e feasibility and promising of the proposed approach in discriminating betw een cervical texture patterns indicative of different stages of cervical le sions.