A Method of Image Classification with Optimized BP Neural Network by Genetic Algorithm

BP neural network is an important and efficient method in machine learning. But there are some drawbacks lying in its local minimum and slow convergence speed. To solve these problems and enhance the performance of BP network, an optimized BP neural network by genetic algorithm is proposed in this paper. Firstly, we design a fitness function based on genetic algorithm for the view of obtaining the best individual, and then optimize the weights and thresholds of the connections between the layers of BP network. The experiments and analysis indicate that the optimized method not only improves the classification percentage, but also obtain satisfactory mean square error.