TYPED DIGITS RECOGNITION USING SEQUENTIAL PROBABILITY RATIO TEST
Keywords:Bayesian classifier, digits recognition, neural network, sequential test
Purpose: the represented research results are aimed to better understanding of computer vision methods and their capabilities. Both the statistical classifier and an artificial neural network allows processing of typical objects with simple descriptors. Methods: considered methods are grounded at probabilistic theory, optimization theory, kernel density estimation and computer-based simulation as a verification tool. Results: the considered artificial neural network architecture for digits recognition has advantage in comparison with statistical method due to its better classification ability. Presented results of experimental verification prove that advantage in both single observation and sequential observation scenarios. Discussion: the approach can be implemented in a variety of computer vision systems that observe typed text in difficult noisy conditions.
Vasudeva N., Parashar H. J., and Vijendra S. (2012) Offline Character Recognition System Using Artificial Neural Network. International Journal of Machine Learning and Computing, No. 4(2), pp. 449–452.
Samadiani N., and Hassanpour H. (2015) A neural network-based approach for recognizing multi-font printed English characters. Journal of Electrical Systems and Information Technology, No. 2, pp. 207–215.
Dong Xiao Ni Seidenberg, “Application of Neural Networks to Character Recognition”, CSIS, Pace University, School of CSIS, Pace University, White Plains, NY, 2007
Coates A., Carpenter B., Case C., Satheesh S., Suresh B., Wang T., Wu D.J., Ng A.Y. (2011) Text Detection and Character Recognition in Scene Images with Unsupervised Feature Learning. Proc. 2011 Int. Conf. on Document Analysis and Recognition (ICDAR). Beijing, China, doi: 10.1109/ICDAR.2011.95
LeCun Y., Jackel L., Bottou L., Brunot A., Cortes C., Denker J., Drucker H., Guyon I., Müller U., Säckinger E., Simard P., Vapnik V. (1995) Comparison of learning algorithms for handwritten digits recognition. Proc. 1995 Int. Conf. on Artificial Neural Networks. Paris, France, pp. 53-60.
Pham C.H., Yaguchi Y., and Naruse, K. (2016) Feature Descriptors: a Review of Multiple Cues Approaches. Proc. 2016 IEEE Int. Conf. on Computer and Information Technology. Fiji, pp. 310-315. doi: 10.1109/CIT.2016.61
Wasserman L. (2006). All of Nonparametric Statistics. Springer Texts in Statistics. New York, NY, Springer, 268 p.
Kharchenko V.P., Kukush A.G., Kuzmenko N.S., and Ostroumov I.V. (2017) Probabilistic Approach to Object Detection and Recognition for Videostream Processing. Proceedings of the NAU, No. 2(71), pp. 8–14. doi: 10.18372/2306-1472.71.11741
Kharchenko V., Pawęska M., Bugayko D., Antonova A., Grigorak M. (2017) Theoretical Approaches for Safety Levels Measurements–Sequential Probability Ratio Test (SPRT). Logistics and Transport, no. 2(34), p.p. 25-32.