Ga Evolution Of The Currency Of The Bp Neural Network To Identify Research

With the rapid development of world economy, people increasingly pursue better efficiency and automation in commercial activities. The currency automation identification has become a new research hotspot, especially in retail and banking sectors. Therefore, it is a major task to rapidly improve the efficiency of millions of systems worldwide at low costs.After analyzing the current situation from home and abroad, the paper makes preliminary investigations and discussions on the artificial neural networks and genetic algorithm, and applies artificial neural networks based on the genetic algorithm to currency identification. Firstly, the paper analyses the background of the currency identification application and the situation of the current study, introduces the concept and application of neural networks and genetic algorithm. Secondly, it gives an recount of the basic theory of the artificial neural networks and genetic algorithm, and analyzes the detailed feasibility study on combining neural networks and genetic algorithm. Thirdly, the model of the currency identification making use of BP neural networks is proposed. At the same time, the author uses a concrete scheme by using genetic algorithm to improve BP neural networks structure and linking values to build the currency identification model. Finally, MATLAB simulation is made to compare and analyze results of the two kinds of models.The experiment indicates: GA-BP model neural networks has shortened training time and gained higher speed and better outcomes; thereby, it shows that BP neural networks based on genetic algorithm have advantage in currency identification.