Skin Embryo Classification Model based on BP Neural Network optimized by Genetic Algorithm

For correlation between attribute values in the pickling process, the characteristics of the property value within a certain range, introducing genetic algorithm of BP neural network weights and thresholds for global optimization, the optimized model of pickling process after skin embryo quality classification. Experimental results show that the optimized BP neural network can improve the skin embryo classification calculation accuracy and convergence rate, a 9.16% increase over the mean square error (MSE) is not optimized, the optimized model and the bayesian decision tree algorithm, classification accuracy were higher than in both algorithms. Therefore, this model has certain guiding significance and reference value in the leather production process, and improves the actual production efficiency of the factory.