Achievement Of A Printed Digit Recognition System Based On Field Programmable Gate Array

A printed digit recognition system based on FPGA is proposed to increase the processing rate and to enhance its hardware realization. It aims at the present research of OCR. It uses the principle of character segmentation based on the projection feature and character recognition based on statistical feature. Further it uses Nexys-3 board from Xilinx for the hardware platform and it uses the OV7670 camera for image acquisition. The system finishes the image pre-processing, data storage, digit segmentation, feature extraction and matching. And finally it displays the recognition result on seven-segment. The system also uses VGA interface to increase the credibility of image data during processing.In the design, median-filter is applied for the image pre-processing part through comparison of 2 algorithms; the binary images are stored only to reduce demand of the memories in the storage part, additionally, ping-pong operation is used for real-time feasibility. The IP cores, such as internal DCM, Multiplier, Dual-port RAM and FIFO, are applied for making better use of the FPGA resources. Firstly, each sub-module is validated through function and timing stimulation by Modelsim, and then the integrated system is implemented on the Digilent Nexys-3 hardware platform. The experiment shows that this system succeeds to identify the printed digits and outputs the results real-timely using very limited logic resources. And the frame rate of the camera is 30 fps.