基于维纳滤波与小波相融合的超声医学图像去噪方法

A method of medical ultrasonic image denoising based on Wiener filter and wavelet transform

目的探求一种有效的超声医学图像去噪方法。方法在分析维纳滤波和基于自适应前处理的多尺度小波非线性阈值斑点噪声抑制方法(MSSNT-A)的基础上,提出一种基于维纳滤波与MSSNT-A相融合的超声医学图像去噪方法。利用该方法首先对加噪图像分别进行维纳滤波和MSSNT-A去噪,然后提取维纳滤波处理后的图像边缘,再将其与MSSNT-A去噪后的图像的相应像素点进行融合,得到去噪图像。结果有效地去除了斑点噪声,图像的细节得到保留。结论与维纳滤波和MSSNT-A方法相比,该方法在有效去除斑点噪声的同时,很好地保留了图像边缘和图像细节信息。

Objective To explore an efficient denoising method of medical ultrasonic image. Methods A method of medical ultrasonic image denoising based on wiener filter and adaptive presholding （MSSNT-A） was proposed. First, the noise image was respectively denoised with wiener filter and MSSNT-A; then the edge of image was extracted; at the end, the edge and the image that was denoised with MSSNT-A were fused together and the final denoised image was obtained. Results The speckle noises were removed and the details of image were reserved. Conclusion Compared with wiener filter and MSSNT- A, this method can better reserve the edge and the details of the image, while speckle noises are removed.