An improved back propagation neural network in objects recognition

The Back Propagation Neural Network(BPNN) has been used widely in objects recognition, but in fact, the BPNN can easily be trapped into a local minimum and has slow convergence. Moreover, the number of neural cells for hidden layer in the BPNN is hard to determine. For this reason, this paper proposes a novel method to improve the performance from the structure and the algorithm. The improved BP algorithm has some advantages in fast convergence speed and short running time. It is applied to objects recognition and has a favorable result. The validity of the improved methods is proved by a series of simulation experiments in the paper.