The severity of grape leaf disease damage to a grape may be estimated by analysing the condition of the grape leaves, which are a reliable indication of the grape plant's overall health. In order to speed up the process of grape leaf disease detection in grape leaf samples, this proposed research work suggests using deep learning based classification and image processing techniques. To achieve this goal, this proposed research work use a technique for extracting relevant features from images. We present an Enhanced CNN Model for Grape Leaf Disease Detection. We found that E-CNN provided the most consistent results with an accuracy of 98.81%.