Comparative Analysis of Exudates Detection from Retinal Image with Effective Preprocessing Method
The most primal sign of Diabetic Retinopathy is exudates. Detecting the exudates at an earlier stage can prevent the vision loss. In this paper, we propose an efficient preprocessing algorithm, to enhance the contrast between background and exudates area. For feature extraction demonstrate Kirsch and Linde-Buzo-Gray (LBG) clustering method. The proposed algorithm is tested on publically available DIARETDB1 database. Due to its distinguishing performance measures, the proposed method has been successfully applied to images of variable quality. Receiving operating curve (ROC) and weighted error rate (WER) which shows that the performance of our method is effective to detect exudate in retinal image.
- There are currently no refbacks.
This work is licensed under a Creative Commons Attribution 3.0 License.