IMAGE PROCESSING, PATTERN RECOGNITION
FPGA-based device for handwritten digit recognition in images
I. V. Zoev, A. P. Beresnev, N. G. Markov, A. N. Malchukov
National Research Tomsk Polytechnic University, Tomsk, Russia
We describe the design and manufacture of a mobile and energy efficient device that allows one to recognize handwritten digits in images using convolutional neural networks. The device is implemented on a field-programmable gate array (FPGA), which is included in the system-on-a-chip Cyclone V SX. Functional diagrams of the computational blocks implementing the convolution and pooling procedures are developed. Functional diagrams of the convolution neural network for the proposed architecture are also described. Results of testing the developed FPGA-based device for its efficiency in terms of handwritten digit recognition accuracy, recognition rate, and power consumption are presented. Results of a performance comparison between a hardware and software implementation of convolutional neural networks are presented.
handwritten digit recognition in images, convolutional neural networks, FPGA-based device.
PDF file (1514 kB)
I. V. Zoev, A. P. Beresnev, N. G. Markov, A. N. Malchukov, “FPGA-based device for handwritten digit recognition in images”, Computer Optics, 41:6 (2017), 938–949
Citation in format AMSBIB
\by I.~V.~Zoev, A.~P.~Beresnev, N.~G.~Markov, A.~N.~Malchukov
\paper FPGA-based device for handwritten digit recognition in images
\jour Computer Optics
Citing articles on Google Scholar:
Related articles on Google Scholar:
|Number of views:|