Fruit diseases cause devastating problem in economic losses and production in agricultural industry worldwide. In this paper, an image processing approach was proposed for identifying fruit diseases based on Convolutional Neural Network. According to the CNN algorithm, fruit image details were taken by the existing packages from the front end used. The algorithm was used for detecting the disease of the fruit. So, this proposed approach can be used to identify fruit diseases quickly and automatically with high accuracy. This proposed approach is composed of the following main steps that getting input image, image preprocessing, identifying affected places, highlighting those affected places, verifying training set and showing result. Images were provided for training, such as bitter rot images, sooty blotch images and powdery mildew images. Before the image processing, images were converted to HSV. Local Binary Pattern method was used for feature extraction. This approach was tested according to fruit disease type and its stages, such as fresh and affected.