Title: Image-Based Eye Classification Using Deep Learning
Authors: Taif Jamal Abu Musabeh, Samy S. Abu-Naser
Volume: 10
Issue: 2
Pages: 66-77
Publication Date: 2026/02/28
Abstract:
Eye image analysis plays an important role in medical diagnosis and health monitoring. Traditional methods for analyzing eye conditions rely heavily on expert knowledge, which can be time-consuming and may not scale well when large numbers of images need to be examined. With the increasing availability of digital eye images, automated image-based analysis has become a promising alternative. In this paper, a deep learning-based approach is presented for classifying eye images using a labeled dataset containing images captured under different conditions. Convolutional neural networks (CNNs), which have shown strong performance in image recognition tasks, were employed to automatically learn discriminative features from eye images. Transfer learning was utilized to improve classification performance, especially given the limited size of the dataset. The experimental results demonstrate that the proposed approach achieves high classification accuracy on a held-out validation set, highlighting the effectiveness and feasibility of using deep le rning techniques for image-based eye classification.