基于各向异性逆扩散方程的指纹图像锐化去噪方法

Fingerprint Image Sharpening and Denoising Method Based on Nonlinear Anisotropic Reverse Diffusion Equation

基于逆扩散过程的启发,提出各向异性逆扩散算法用于指纹图像的锐化去噪方法,克服了退化扩散方程对大曲率边缘点的模糊效应,兼顾了去噪和保持边界这一矛盾的两个方面,尤其适合于纹理密集的指纹图像锐化。实验结果表明,本文算法对于带噪指纹图像的锐化效果明显优于以往非线性扩散处理算法。

Considering the feature of the fingerprint, a nonlinear anisotropic reverse diffusion algorithm is proposed from the reverse diffusion phenomena in physics. The method is effective for both noise removing and edge preserving, mainly because it prevents the blurring effect on the degenerate diffusion equation for the points with big gray level variations. Compared with nonlinear diffusion methods, the algorithm has better performances for preprocessing noisy fingerprint images.