Abstract
An effectual, noninvasive method to identify retinal exudates centered on binary operation is proposed in this research work. A new Histogram equalization technique based on intensity index is utilized for fundus image enrichment. Subsequent to the elimination of Optic Disc (OD) from a fundus image, morphological operation is accomplished to identify the exudate pixels. Ultimately, grouping of hard exudates using a supervised Support Vector Machine (SVM) classifier is realized and calculated using five unique performance factors. Excellent results were obtained and the method can be utilized for pre-processing of retinal images in order to help an ophthalmologist for primary detection of retinopathy symptoms.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have