Abstract:
Cataract is one of the leading causes of blindness, especially among older people. By the age of 80,
almost half of the older population of India either have cataract or have already had a surgery for it. Surveys by WHO/NPCB suggest that there are over 12 million blind people in India out of which 80.1% of these people are blind because of cataract. It is crucial that cases of cataract are detected in its earlier stage to avoid complete blindness. In this research paper, digital image processing is used on
eye images to determine the presence of cataract along with its severity. Two different image processing algorithms are applied on a dataset that consists of images of eyes having varying
degrees of cataract. The first approach uses a feature extraction based automated detection algorithm. It accepts the image, preprocesses it, extracts information regarding intensity and makes classification decisions- healthy eye, mild cataract or severe cataract. Histograms are plotted for analysis. The second method determines the degree of cataract using an area based approach. Hough transform and contour detection are used to detect shapes in the image, inversion using logical operations is performed and percentage of cataract is successfully calculated. The algorithms are compared for efficiency and accuracy.