Abstract: Diabetic eye disease is one of the major problems worldwide. That can cause major eye impairment, including a permanent loss of vision. Early detection of eye diseases increases the survival rate by successful treatment. The blood vessels are the primary anatomical structure visible in retinal images. The segmentation of retinal blood vessels has been accepted worldwide for diagnosing both diabetic and retinal diseases. Thus, it requires an appropriate vessel segmentation method to detect retinal diseases such as diabetic retinopathy and cataracts automatically. Detecting retinal diseases using computer-aided diagnosis (CAD) can help people avoid the risks of visual impairment and save medical resources. Existing methods practically gave good testimony to understanding the genetics of retinoblastoma. Innovative development of low-cost application-based intelligent systems integrated with a microscopic lens allows patients in remote and remote areas to be regularly screened and diagnosed. The diagnostic system uses artificial neural network algorithms to analyze retinal images captured to detect retinal disease status. The algorithm is first trained using a personal computer with infected and normal retinal photos and then developed in an atmospheric diagnostics application. The proposed methodology is to explore machine learning techniques to detect diabetic diseases using thermography images of an eye and to introduce the effect of variation of abnormality in the eye structure as a diagnosis imaging modality, which is useful for ophthalmologists to make the clinical diagnosis. Images are pre-processed, and then Gray Level Co-occurrence Matrix (GLCM) based texture features from grey ideas, statistical features from RGB and HSI images are extracted and classified using a classifier with various combinations of features. They detect diabetic diseased eyes, a CNN classifier is used for classification, and their performance is compared. A different-fold cross-validation scheme is used to enhance the generalization capability of the proposed method.Keywords: Retinal disease Neural Networks, Retinal image analysis, Blood Vessel Detection, Image Segmentation, Fundus Image Classification, Feature Selection