Diagnosis of some apple fruit diseases by using image processing and artificial neural network

Abstract

Farmers typically lack the knowledge of diagnosis and control of different apple diseases. However, some apple diseases have visual symptoms and can be diagnosed by eyes, but their diagnosis by eyes is time-consuming and costly for farmers. A solution is to design an automatic disease diagnosis system using image processing techniques. This paper presents a low-cost method of apple disease diagnosis using a neural network and fruit classification into four classes of scab, bitter rot, black rot, and healthy fruits. This method uses color and texture features. The research used a multi-layer perceptron neural network whose input was the features extracted from the images and its output was the defined classes. After the network was trained by 60% of the images and the remaining images were reserved for its testing, the accuracy of the proposed method was assessed with different structures of single-layer and two-layer neural network structures. Based on the results, the application of a two-layer structure with eight neurons in the first layer and eight neurons in the second layer resulted in an optimal accuracy of 73.7%. • Developing a novel automatically method for diagnosing apple fruit diseases. • Using image processing and artificial neural network for diagnosing apple fruit diseases. • Providing a new dataset of apple images in field of diagnosing fruit diseases.