Features Recognition on Retinal Fundus Image – A Multi- Systemic Comparative Analysis

This Article describes the perspective analysis and study on Pattern Recognition of the Retinal Nerve Fibers. The articles published in recent years are considered for observing the analytical techniques as well as approaches used for implementing the image-based processes of our Proposed System. The various Process Implementation Systems play important role for obtaining the accuracy in the performance and time-complexity based processes. Various Technical Systemic processes, approaches, methodologies, techniques are used in various articles related to retinal images published recently. The Systems such as Support Vector Machine (SVM), Linear Discriminant Analysis (LDA), Artificial Neural Networks (ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS) are mostly used for implementing Image-based tasks. In this article the Back Propagation Technique is applied on ANN. The efficient results of the proposed processes are produced by the system named as ANFIS1 with Grid Partitioning Technique and ANFIS2 with Subtractive Clustering Technique. The Retinal Images, which are extracted from Original Retinal Fundus Image DB Set-1 and Original Retinal Fundus Image DB Set-2 have been processed by using all these systemic techniques for obtaining the Percentage of Average Classification Error, which is the core value for identifying the efficiency of a System among multiple systems.