Finger-vein image restoration based on skin optical property

In this paper, a new finger-vein image restoration method is proposed to improve the quality of finger-vein images. In finger-vein imaging, light scattering is the main factor causing finger-vein image degradation, so dealing with scattering issue is helpful for finger-vein restoration. Traditionally, to remove the scattering effect, finger-vein restoration was often implemented using either an estimated point spread function (PSF) or an estimated biological optical model (BOM). However, both PSF and BOM are incapable of representing the light scattering phenomenon in biological tissues. To reliably process the light scattering in finger-vein imaging, the multilayered PSF and BOM are integrated together in this paper considering the real structure of the human skin tissue. Using the proposed method, the finger-vein images can be progressively restored in a layered manner. The experimental results validate that the proposed method has a good performance in implementing finger-vein restoration.