基于二维EMD和小波阈值的掌纹图像去噪

为有效抑制掌纹图像中含有的噪声、提高特征提取的精度,提出一种基于二维经验模式分解和小波阈值去噪相结合的掌纹图像去噪新方法。首先,对含有噪声的掌纹图像进行二维EMD分解,得到不同特征尺度的本征模函数子图像;然后对中高频成分的IMF进行小波多阈值去噪;最后将去噪处理后的各IMF与残差图像通过加和进行重构。实验结果表明,该方法与单独的二维EMD滤波及小波阈值去噪相比,去噪效果更明显,提取的主线和细节特征更清晰,因而均方误差最小、峰值信噪比最高。

Palm Print Image De-noising Based on 2-D EMD and Wavelet Thresholding

In order to suppress the palm print image noise effectively and extract the palm print features accurately, a novel de-noising method based on 2-D EMD and wavelet thresholding was proposed. Firstly, the preprocessed palm print image containing noise was decomposed into some IMFs by 2-D EMD, and then the first several IMFs were processed by using wavelet thresholding de-noising method; finally, the image was reconstructed through adding the processed IMFs and the residual component. Experimental results show that the proposed method has a good effect on suppressing noise. Compared to the wavelet thresholding and 2-D EMD de-noising, the method not only has advantages of more sufficiently retaining edge and detail information while de-noising, but also achieves superior PSNR (peak signal-noise-ratio) of the reconstructed image.