Application of SVM based on genetic algorithm in classification of cataract fundus images

Cataract is one of the most common diseases that might cause blindness. Previous research shows that cataract occupies almost 50% in severe visually impairments. Fundus image is a significant reference for the diagnosis of the cataract disease. The classification of fundus images mainly consists of four parts: pre-processing of fundus images, features extraction, features weighting and classification. In this paper, firstly a whole fundus image is divided into 17 images evenly, secondly features are extracted features from each sub-image, then the feature vectors are weighted with the result of genetic algorithm and finally support vector machine is used to train and classify the fundus images. The experimental result shows that the accuracy of four-class classification can reach 87.52%.