Enhancing the Dorsal Side of Fingers Using An Image Enhancement Technique with FPGA Output Comparison

Abstract

Most of the image enhancement techniques are implemented on CPU and GPU, but there is limited implementation on the FPGA platform. This paper presents a work on an enhancement technique called Histogram Equalization, HE for finger-knuckle images. This project is divided into three phases, which are image acquisition, image enhancement, and evaluation. For image acquisition, a USB webcam is set up as the acquisition device to acquire the image of fingers. For image enhancement, the Histogram Equalization (HE) method is chosen due to the less complex algorithm, especially when evaluating the performance on the FPGA platform. Two processing platforms are considered to complete the study, which are desktop computers that use MATLAB programming and the FPGA DE1-SoC platform. A comparison of the results is carried out between these two processing platforms, where it is found that the results for both platforms have shown identical output in terms of PSNR, which achieved a value of 13.43 dB and MSE with 0.0454.

Keywords

• Finger-Knuckle-Print (FKP), FPGA, Histogram Equalization (HE), Image Enhancement, Verilog HDL