Diabetic retinopathy is a common complication of diabetes and is a leading cause of vision loss in adults. Early detection and management of diabetic retinopathy is crucial to prevent further vision loss. In recent years, image processing techniques have been widely used for automated detection and diagnosis of diabetic retinopathy. In this study, this paper investigates the use of the Gray Level Co-occurrence Matrix algorithm for feature extraction from diabetic retinal images with BRISK algorithm. The GLCM algorithm is a popular method for feature extraction in computer vision, and its use for diabetic retinopathy detection has shown promising results. Results demonstrate that the Gray Level Co-occurrence Matrix (GLCM) algorithm can effectively extract features from diabetic retinal images and BRISK can be used for automated detection feature extraction and robotics and autonomous systems diagnosis of diabetic retinopathy. Finaly, the top 80% of the feature points with high reaction esteem were separated as the main element focuses. Evaluate the performance of the GLCM algorithm and BRISK using various performance evaluation units, including accuracy, precision, recall, and F1 score.

GLCM, BRISK, MISSIDOR, DRIVE, Kaggle dataset, F1 score