Y.-P. Chu, S.-F. Yang-Mao, Y.-F. Chen, Y.-K. Chan, C.-S. Tsai, and W.-C. Chiang (Taiwan)
Retinal Fundus Image, Optic Disk, Texture, Statistic parameters, Eigenface
In this paper, we propose a texture-based method for automatically locating the optic disk on the retinal fundus images. Totally 46 fundus images were used for the experiment, in which 23 images were randomly chosen and used for training and the others for testing the performance of the proposed method. In the training step, 3 blocks containing the 1% brightest pixels were collected from each image, which accounted for 69 blocks, and used as the training set. The weighted function was defined by linearly combining three regression lines obtained by fitting three pairs of statistic parametric relations, which were then used to precisely detect the optic disk regions from the retinal fundus images. The proposed method shows that it can effectively eliminate incorrect optic disk detection caused by extending bright areas around the optic disk and significantly improves the performance compared to previous studies.
Important Links:
Go Back
