Study on predicting method for Haff value of eggs based on grey neural network and image processing

The purpose of this paper is to add grey theory to BP neural network for predicting eggs Haff value and evaluating freshness degree of eggs. Firstly process the egg images with light-transmitting were obtained by the computer vision device including denoising, threshold segmentation, conversing HSI Color model and calculating the averages of hue, saturation, and intensity in the center of the image. Secondly analyze gray model, and then determine the Grey Neural Network topology according to the predicted formula being derived. Thirdly train Grey Neural Network and predicate Haff value by HSI parameter data as the sample. The value of residual errors of Grey Neural Network model are 5.2684, the correct discerning rate of grading table eggs is 92.7%. It proves better than traditional BP neural network in terms of predicted accuracy and robustness. The generalization ability of Grey Neural Network is strengthened.