Diabetic Retinopathy (DR) is a widespread stage of diabetes that influences the retina and, if untreated, can lead to blindness. DR is one of the primary reasons for blindness, especially among working-age populations in developing countries. Treatment can solely assist in preserving vision due to the irreversible nature of the disease. Timely identification and intervention substantially diminish the likelihood of blindness. Manual diabetic retinopathy diagnosis with retinal images is costly, time-consuming, and labor-intensive. The automatic detection of diabetic retinopathy disease can be possible using deep leaning methods specially CNN. In this paper, several diabetic retinopathy detection methods based on deep learning are evaluated. The accuracy of the suggested method is 94.04%.