%0 Journal Article
%T 基于重加权相关全变分模型的医学图像去噪方法<br>Medical Image Denoising Method Based on Reweighted Correlation Total Variation Model
%A 张芝浩
%A 李涛
%J Artificial Intelligence and Robotics Research
%P 203-212
%@ 2326-3423
%D 2024
%I Hans Publishing
%R 10.12677/airr.2024.132022
%X 图像去噪一直是一个受到大量研究者关注的问题，并成功应用到医学等领域。典型的图像去噪方法是利用图像中存在的先验信息，例如低秩先验，局部光滑先验等。但是，目前的图像去噪方法并没有充分利用到图像的这些先验信息，导致去噪效果不是很理想。针对上述问题，本文提出了基于重加权相关全变分正则项的图像去噪模型。该模型利用重加权核范数的方式对相关全变分正则项进行约束，来保证更加充分地利用图像中的低秩先验和局部光滑先验，以此来提升图像恢复效果。我们应用该方法到医学图像中，并和常见的几种图像去噪方法进行比较，实验结果显示，该方法所得到的图像质量得到了明显的提升。<br />
Image denoising has always been a problem that attracts the attention of a large number of researchers, and has been successfully applied to medical fields. Typical image denoising methods use the prior information existing in the image, such as low-rank prior, local smooth prior, etc. However, the current image denoising methods do not make full use of the prior information of the image, resulting that the denoising effect is not very ideal. To solve the above problems, this paper proposed an image denoising model based on reweighted correlation total variation regularization term. In this model, the reweighted nuclear norm method is used to constrain the correlated total variation regularization term to ensure that the low-rank prior and local smooth prior in the image are fully utilized, so as to improve the image restoration effect. We applied this method to medical images and compared it with several common image denoising methods. The experimental results show that the image quality obtained by this method has been significantly improved.
%K 图像去噪，先验信息，重加权核范数，相关全变分正则项<br>Image Denoising
%K Prior Information
%K Reweighted Nuclear Norm
%K Correlated Total Variation Regularization Term
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=86456