Abstract:

According to the characteristics of finger vein image, a new denoising model based on partial differential equations was presented. This model uses the new diffusion function based on the traditional PM model, and combines with fourthorder partial differential equations model to transform the original model structure. The performance of the new model is verified by both synthetic and real finger vein images. It shows that the new model could increase the signaltonoise ratio up 5 dB and maintain the features of finger vein image better compared with the original model.

Key words: partial differential equations (PDE), image denoising, PM model, signaltonoise ratio (SNR)