AUTOMATED GLAUCOMA DETECTION SYSTEM BASED ON TWIN STAGE SEGMENTATION AND MACHINE LEARNING

Abstract

Glaucoma is a serious threat and it causes blindness and it ranks third in India. Early identification of glaucoma sickness prevents eye disorders from worsening. Due to the need for early illness detection tools to aid in screening and management, retinal image analysis has attracted keen interest. This article presents an automated glaucoma detection system, which is based on machine learning. The retinal images are denoised and contrast enhanced, followed by which the Optic Disc and Cup are extracted. The Complete Local Binary Pattern (CLBP) and contourlet features are extracted to train the Extreme Learning Machine (ELM) classifier. The ELM differentiates between the glaucomatous and non-glaucomatous images. The experimental findings are evaluated with the existing methods, and it is found that the proposed work is superior in terms of standard performance measures.