Abstract: This paper introduced one classification method for fruit grade based on BP neural network. Fruit images were preprocessed by using digital image processing method. The mean color value and variance of fruit surface were selected to express fruit color features. An ellipse that has the same normalized second central moments with the fruit region were adopted to approximately represent fruit shape. It simplified the complex degree of the shape description. The optimum structure parameters of the BP neural network which had 9 hidden layer neurons were determined by RP training algorithm. Results showed that average accuracy for fruit classification can reach 92.5% by using this model, and the executing time of microcomputer for grading of one apple is 10.3 ms. This method has the characteristics of high accuracy and good real-time performance.

Key words: fruit classification, BP neural network, image processing, second central moment