Diabetic Retinopathy (DR) is an eye ailment that influences eighty to eighty-five percent of the patients having diabetes for more than ten years. The common process for analysis and detection of diabetic retinopathy is through retinal fundus images. The raw retinal fundus images are very hard to process by machine learning algorithms. The pre-processing of raw retinal fundus images is performed through the extraction of the green channel, image enhancement, and resizing techniques. These features are also extracted from pre-processed images for quantitative analysis. The experiments was performed using Kaggle Diabetic Retinopathy dataset, and the results were evaluated by considering the mean value and standard deviation for extracted features. The result yielded an exudate area as the best-ranked feature with a mean difference of 1029.7. The result attributed due to its complete absence in normal diabetic images and its simultaneous presence in the four classes of diabetic retinopathy images namely mild, normal, and severe and Proliferative.