基于脊波变换的手指静脉图像增强研究

摘要

针对人体手指静脉图像的结构和特点,提出一种基于脊波变换的图像增强算法.该算法对手指静脉图像小波域各高频子带系数进行脊波变换,利用非线性新脊域系数确定法对脊域系数进行处理,然后对各高频子带进行脊波逆变换和小波图像重构.该算法对手指静脉二维曲线奇异处理、边缘增强等具有较好的效果,克服了小波变换在高维曲线奇异和方向选择上的不足.通过与传统的二维离散小波变换边缘增强法和自适应调整系数的脊波变换方法作比较,实验结果表明本算法具有更好的手指静脉图像效果.%According to features of the human finger vein images, the algorithm of image enhancement based on the ridgelet transformation is presented in this paper. Firstly the high-frequency sub-band coefficients are transformed by ridgelet after the finger vein image's wavelet transformation. Then the ridgelet coefficients are processed by using the non-linear new coefficient-determined method. At last do the inverse ridgelct transformation for each high-frequency sub-band and the wavelet reconstruction. The algorithm is effective in processing the finger vein's two-dimensional curve singularity and the image edge enhancement. It overcomes the shortages of wavelet transformation in high-dimensional curve singularity and direction choices. Comparing with the traditional two-dimensional discrete wavelet transformation and the ridgelet transformation algorithm through adaptively adjusting the ridge]et transformed coeffients, the experimental results show that the algorithm presented in this paper has a better effect in finger vein's image enhancement.