• A Ramyasree
• M Munisankar

In recent years exudates are emerged as most prominent signs of diabetic retinopathy in diagnostic measurements, and its detection has been investigated in many works for most appropriate diabetic retinopathy CAD system. Here, super pixel segmentation is formulated with most relevant multi-feature set extraction for accurate and fully automated network based classification for the detection of exudates. The process begins with segmentation of super pixels followed by generation of 20 features from two different channels each belongs to different intensity levels namely contextual features. The supervised classification algorithm is used to discriminate the exudates from other classes. Experimental results prove that the proposed models offer improved detection and classification performance in terms of accuracy. Performance validation is carried out over two well known publicly available bench mark datasets namely, DiaretDB1, and e-ophtha EX to prove the effectiveness of the proposed framework.

Diabetic Retinopathy, CAD system, color retina fundus image, exudates, local luminosity, pixel and feature classification etc.

"Hierarchical Model For The Detection Of Exudates Based On Retinal Features Analysis", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.8, Issue 1, page no.278-283, January-2021, Available :http://www.jetir.org/papers/JETIR2101036.pdf

"Hierarchical Model For The Detection Of Exudates Based On Retinal Features Analysis", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.8, Issue 1, page no. pp278-283, January-2021, Available at : http://www.jetir.org/papers/JETIR2101036.pdf