Intelligent Discrimination of Fruit Variety Quality Based on Bp Neural Network

Abstract

In recent years, thanks to the development of computer vision technology and digital image technology, traditional agriculture and information technology have been deeply integrated. This study uses image recognition and bp neural network technology to identify the types of fruits and selects apples as objects for quality evaluation. First, 150 images of apples, bananas, peaches, avocados, and cherries were selected from the fruit database Fruit360 after grayscale processing as a data set. After the images were grayscaled, image denoising, edge detection, and other feature processing in MATLAB, The classification of five kinds of fruits has been successfully realized, and the correct rate of evaluation reached 94.8%. Then divide the data sets for apples of different varieties and qualities, and input them into the bp neural network for training. After testing, the 6-layer bp network has the highest accuracy rate, and can effectively classify and score apple images.