Recognition of Russian Banknote Nominal Values by Mobile Devices for Blind People

This paper focuses on the system of recognition the nominal values of Russian banknotes by photos for blind people. The system uses the knowledge transfer technique and deep learning methods. In our research we compared and analyzed the performance and accuracy of approaches by using the ResNet-50, VGG-19 and Inception-v3 architectures for the primary feature extraction from photos. After that we developed three prototypes based on these architectures and tested the system on desktop and mobile processors. The system based on the ResNet-50 architecture showed the best recognition accuracy. As for its efficiency, it appeared to be worse than that of the system based on Inception-v3 architecture. However, the Inception-v3 architecture showed very low accuracy of 78 %. Findings of the research show that ResNet-50 architecture could be used in real life conditions due to the accuracy of the solution based on it

References

[1] Pascolini D., Mariotti S.P. New estimates of visual impairment and blindness: 2010. British Journal of Ophthalmology, 2011, vol. 96, no. 5. Available at: http://www.who.int/blindness/estimates2011.pdf (accessed: 15.07.2017).

[2] Bruna A., Farinella G.M., Guarnera G.C., Battiato S. Forgery detection and value identification of euro banknotes. Sensors, 2013, vol. 13, iss. 2, pp. 2515–2529. DOI: 10.3390/s130202515

[3] Park Y.H., Kwon S.Y., Pham T.D., Park K.R., Jeong D.S., Yoon S.S. A high-performance banknote recognition system based on a one-dimensional visible light line sensor. Sensors, 2015, vol. 15, iss. 6, pp. 14093–14115. DOI: 10.3390/s150614093

[4] Semary N.A., Fadl S.M., Essa M.S., Gad A.F. Currency recognition system for visually impaired: Egyptian banknote as a study case. Proc. Int. Conf. on Information and Communication Technology and Accessibility, 2015. DOI: 10.1109/ICTA.2015.7426896

[5] Hasanuzzaman F.M., Yang X., Tian Y. Robust and effective component-based banknote recognition by SURF features. WOCC, 2011. DOI: 10.1109/WOCC.2011.5872294

[6] Singh S., Choudhury S., Vishal K., Jawahar C.V. Currency recognition on mobile phones. Proc. 22nd Int. Conf. on Pattern Recognition, 2014, pp. 2661–2666. Available at: http://web2py.iiit.ac.in/research_centres/publications/download/inproceedings.pdf.9797adb46eb9d9a7.5375726979613230313443757272656e63792e706466.pdf (accessed: 15.07.2017).

[7] Parlouar R., Dramas F., Macé M.J-M, Jouffrais Ch. Assistive device for the blind based on object recognition: An application to identify currency bills. Proc. 11th Int. ACM SIGACCESS Conf. on Computers and Accessibility, 2009, pp. 227–228. Available at: https://www.irit.fr/~Marc.Mace/pdfs/parlouar_r_09_227.pdf (accessed: 15.07.2017).

[8] Bhurke C., Sirdeshmukh M., Kanitkar M.S. Currency recognition using image processing. Int. J. of Innovative Research in Computer and Communication Engineering, 2015, vol. 3, no. 5, pp. 4418–4422.

[9] Gutstein S., Fuentes O., Freudenthal E. Knowledge transfer in deep convolutional neural nets. Proc. of Twentieth Int. Florida Artificial Intelligence Research Society Conf., 2007, pp. 104–109. Available at: http://www.cs.utep.edu/ofuentes/FLAIRS07GutsteinS.pdf (accessed: 15.07.2017).

[10] He K., Zhang X., Ren S., Sun J. Deep residual learning for image recognition. Proc. IEEE Conf. on Computer Vision and Pattern Recognition, 2016. DOI: 10.1109/CVPR.2016.90

[11] Simonyan K., Zisserman A. Very deep convolutional networks for large-scale image recognition. Proc. 3rd Int. Conf. on Learning Representations, 2015. Available at: https://arxiv.org/pdf/1409.1556.pdf (accessed: 15.07.2017).

[12] Szegedy C., Vanhoucke V., Ioffe S., Shlens J., Wojna Z. Rethinking the inception architecture for computer vision. Proc. IEEE Conf. on Computer Vision and Pattern Recognition, 2016, pp. 2818–2826. DOI: 10.1109/CVPR.2016.308

[13] Russakovsky O., Deng J., Su H., Krause J., Sanjeev S., Ma S., Zhiheng H., Karpathy A., Khosla A., Bernstein M., Berg A.C., Li F.-F. ImageNet large scale visual recognition challenge. Int. J. of Computer Vision (IJCV), 2015, vol. 115, iss. 3, pp. 211–252. DOI: 10.1007/s11263-015-0816-y

[14] Kingma D.P., Ba J. Adam: A method for stochastic optimization. Proc. Int. Conf. on Learning Representations ICLR, 2015. Available at: https://arxiv.org/pdf/1412.6980.pdf (accessed: 15.07.2017).

[15] Maaten L., Hinton G. Visualizing data using t-SNE. Journal of Machine Learning Research, 2008, no. 9, pp. 2579–2605. Available at: http://www.jmlr.org/papers/volume9/vandermaaten08a/vandermaaten08a.pdf