Fruit disease detection and classification using convolution neural network

Abstract

The fruit industry is India's largest industry. Due to carelessness and incorrect human inspection, the fruit disease results in significant losses in production, quality, and quantity. The manual examination is a laborious and time-consuming process. An image processing method is provided for the detection and classification of apple fruit diseases using various colour, texture, and shape feature combinations. The essential steps of the recommended approach include Fruit illness detection utilizing a typical neural network (CNN), where fruits are sorted, picture segmentation, feature extraction (color, texture, and shape), feature merging, and feature extraction into ill or healthy classifications. In the lab, our suggested technique was evaluated and confirmed. Using the suggested technique, 97% accuracy was attained.