DETECTION AND CLASSIFICATION OF BRIGHT LESIONS IN COLOR FUNDUS IMAGES (MA-L4)

Author(s) : Xiaohui Zhang (School of EEE, NTU, Singapore) Opas Chutatape (School of EEE, NTU, Singapore)

Abstract : Bright lesions including exudates and cotton wool spots are main symptoms in diabetic retinopathy. Early detection and classification of these evidences is essential for an effective treatment. In this paper, a three-stage approach is applied to detect and classify bright lesions. After local contrast enhancement preprocessing stage, Two-step Improved Fuzzy C-Means is applied in Luv color space to segment candidate bright lesion areas. The results are shown to be effective in dealing with the inhomogeneous illumination of the fundus images while reducing the influence of noises. Finally, a hierarchical support vector machines (SVM) classification structure is successfully applied to classify bright non-lesion areas, exudates and cotton wool spots. Finally, a hierarchical support vector machines (SVM) classification structure is applied to classify bright non-lesion areas, exudates and cotton wool spots.

Menu