一种基于Curvelet变换的指纹图像增强方法

Fingerprint image enhancement method based on the curvelet transform

Curvelet变换具有各向异性和多方向性等良好特性,是表示分段平滑曲线边缘的最优基。为了确保指纹特征提取算法的鲁棒性,需要对原始含噪指纹图像进行预处理以增强纹线的清晰度,增加脊线和谷线的对比度,减少伪信息。本文提出了一种基于Curvelet变换的指纹图像增强方法,该方法对Curvelet变换后的低频系数实现线性灰度拉伸,以增强其对比度,而高频系数采用阈值去噪。仿真实验表明,该方法优于常用的空间域直方图均衡化和小波域图像增强法,具有良好的视觉效果。

Curvelet transform has certain perfect characteristics such as anisotropy and directionality. It＇s the optimal basis for the representation of the edges of smooth curves. In order to ensure the robustness of the minutiae extraction algorithm from the input fingerprint images, it is essential to preprocess the original fingerprint images to enhance the ridge clarity and the contrast of the ridge and valley, and reduce spurious ridge information. The paper gives an algorithm of fingerprint image enhancement based on the Curvelet transform. The coefficients adjustment method is proposed. For the Curvelet that has transformed, the method accomplishes linear gray stretch to enhance its contrast for the low-frequency coefficients, while uses eliminating threshold noises for the high-frequency coefficients. Experimental results show that the proposed approach is superior to both histogram equalization and wavelet based contrast enhancement, and has good visual effect.