A Novel Framework for Iris Disease Detection Using MobileNetV2 and CNN with Streamlit Interface

“Iris Disease Detection” project is a comprehensive approach to iris health assessment that combines deep learning techniques with image processing methodologies. The system consists of a Streamlit-based web application and a Mobile Net V2 and CNN Classifier, which use OpenCV and KMeans clustering for effective image analysis. The dataset, derived from medical imaging platforms, includes hundreds of high-resolution photos of the iris, annotated with detailed diagnostic information about a wide range of iris health problems. The project includes both normal and diseased irises, as well as those with cataracts, glaucoma, and iritis. The Streamlit web application provides a user-friendly interface for uploading iris images, which undergo preprocessing, grayscale conversion, and Gaussian blur before applying the Hough Circles method to detect the iris region. The system generates visualizations, including the original image, identified iris, and segmented representation achieved through KMeans clustering. The machine learning component enhances the accuracy of 95 percentage and reliability of iris detection, offering a holistic solution for iris disease detection.