Recognition of Paper Currency Research Based on AGA-BP Neural Network

Abstract:

The genetic algorithm is a randomized search method for a class of reference biological evolution of the law evolved, with global implicit parallelism inherent and better optimization. This paper presents an adaptive genetic algorithm to optimize the use of BP neural network method, namely the structure of weights and thresholds to optimize BP neural network to achieve the recognition of banknotes oriented. Experimental results show that after using genetic algorithms to optimize BP neural network controller can accurately and quickly achieved recognition effect on banknote recognition accuracy compared to traditional BP neural network has been greatly improved, improved network adaptive capacity and generalization ability.

• Periodical: Advanced Materials Research (Volumes 989-994)
• Pages: 3968-3972
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.989-994.3968
• Online since: July 2014
• Authors: Xue Xiao*, Qing Hong Wu, Ying Zhang
• Keywords: BP Neural Network (BPNN), Genetic Algorithm (GA), Oriented Identification

References

[1] Kong Fanhui. Research-oriented high-speed sorter RMB Par recognition technology [D]. Harbin Institute of Technology master's degree in theory Man, 2004: 30-35.

[2] Cui Mingxing. Research banknote recognition algorithm [D]. Harbin Industrial University, a master's degree thesis, 2002: 22-27.

[3] Rafael C. Gonzalez, Richard E. Woods, Steven L. Eddins. Digital image processing [M]. Electronic Industry Press . 2009: 117-121.

[4] Wei Zhengyi. Study notes image recognition algorithm [D]. Harbin Polytechnic University master's degree thesis, (2009).

[5] Cao Buqing. Based Currency Recognition GA evolution BP neural network [D]. Zhongnan University graduate dissertations pack, (2007).