| Popis: |
Diabetic retinopathy and diabetes hyperglycemia are becoming more common over the global. Presently, otolaryngologists are having a difficult time distinguishing between the different phases of eye problems. Microaneurysm is the first step of such phases. A new texture-based machine microaneurysm diagnosis technique is described. The textured descriptive show that participants the textural properties of every picture, greatly increasing MA detection capability over pattern characteristics. When applying a Logistic regression classifier to differentiate the tumors, the retrieved characteristics from the Local Binary Pattern contributes significantly. The suggested project schedule is depicted in Figure 1 and contains six phases: (1) spatial measurement, (2) preprocessing, (3) optic region, (od) elimination, (4) candidate retrieval, (5) extraction of features, and (6) categorization. The output of proposed is simulated in matlab and compared with existing approaches, proposed it performs better than the existing methodologies. |
