Application of Improved BP Algorithm in Vehicle License Plate Recognition

First, the background, significance and general implementation of vehicle license plate recognition (VLPR) are introduced. Based on analyzing the theory of neural network pattern recognition system, toward the limitation of standard BP, this thesis offers its improved method and a real recognition example. Finally, conclusion on its characteristic is given. The improved BP algorithm had some advantages in reaching high error precision, fast convergence speed, short recognition time and high recognition rate.