A survey on classification techniques in biometric retinal system

Biometric system is a steady and accurate model for retina based recognition and identification. Biometric recognition provides natural and stable solution to the problem for person identification. Various biometric attributes have been produced and utilized to examine each person. In This application performs the following process. Initially Adaptive Histogram Equalization (AHE) and Gaussian Filter preprocessed the captured image. After that small elements are extracted by top hat transformation and Binary morphological reconstruction. KNNRF (K-Nearest Neighbour Random Forest) classifier is utilized for clustering the nearest image pixels. At last, minutiae technique is used for finding bifurcation point angle and width of the blood vessel. After completing all process the image is compared with the database. The image verification process is valid then only user can access the smartphones. It enables high security, good performance and greater accuracy. Also it provides better FAR, FRR and decreases the error rate. Furthermore, in this work, a Geographical User Interface (GUI) with biometric verification framework has been proposed. It comprises of different MATLAB capacities identified with image processing and utilizing the same to make an essential image processing editorial manager having various features.