Detection of MAs in Retinal Images Based on SVM and ASOBoost Classifiers |
Author(s): |
| Prof. S. B. Patil , Department of Electronics and Telecommunication Sinhgad College of Engineering, Pune; Prof. S. B. Takale, Department of Electronics and Telecommunication Sinhgad College of Engineering, Pune |
Keywords: |
| Diabetic Retinopathy, Segmentation, Microaneurysms, Classification, Fundus Image |
Abstract |
|
Diabetic Retinopathy (DR) is the major cause of blindness in diabetic patients. Regular screening of retina using fundus images can stop or reverse the progress of this disease. Microaneurysms are common and early medical signs of DR. Automatic detection of DR lesions in retinal images can help the diagnosis and treatment of the disease. Proposed system focused exclusively on MA detection which is earliest sign of DR. Retinal Image is pre-processed using AHE and Gaussian filter followed by enhancement and extraction of blood vessels in order to reduce false positives. Then MA's are identified and features are extracted which are applied as an input to classifier such as ASO BOOST and SVM are analysed for classifying retinopathy lesion from non-lesion. |
Other Details |
|
Paper ID: IJSRDV6I30955 Published in: Volume : 6, Issue : 3 Publication Date: 01/06/2018 Page(s): 2237-2240 |
Article Preview |
|
|
|
|
