基于稀疏梯度域自适应字典学习方法的CT重建-电子与通信工程专业论文.docxVIP

摘 摘 要 万方数据 万方数据 摘 要 CT 成像技术是一种能准确、直观地显示物体内部信息的技术。自第一台 CT 机问世起，CT 技术得到了广泛的关注，并被应用到了各个领域。如何加快 CT 的扫描速度和减少对被检体的射线剂量一直是 CT 的研究热点，一个有效途径就 是减少重建图像所需的投影数量，即采用稀疏投影。但传统的 CT 图像重建方法 因其固有的局限性在稀疏投影下重建图像会产生严重的图像失真、伪影及低信 噪比。2006 年提出的压缩感知理论证明了稀疏信号可以以远低于奈奎斯特采样 定理所要求的采样率实现信号的精确重建，为稀疏投影下 CT 的图像重建提供了 可能。 基于压缩感知理论，本文采用了一种基于稀疏梯度域的自适应字典学习算 法的 CT 重建(简称 GradDL-CT)方法。该方法将梯度域字典学习方法应用于 X-CT 图像中，采用自适应字典学习方法有效减少由 TV 模型带来的块效应，选取的训 练样本不是以往的图像像素域的像素，而是基于更加稀疏的图像梯度域，相比 于图像本身，其对应梯度域要稀疏的多。因此对梯度域图像进行字典学习比对 图像本身的字典学习有更稀疏的表示。训练样本的稀疏性的提高在一定程度上 可以提高学习字典的精度和鲁棒性，从而得到更好的图像重建效果。 GradDL-CT 算法的思路为：首先分别从梯度图像的水平和竖直两个方向训 练字典，然后从稀疏表示的梯度图投影数据中重建所需的图像，利用梯度图像 的稀疏性达到减少 X 射线剂量的目的。实验结果表明，相比 TV-CT 算法和 DL-CT 算法，GradDL-CT 算法在稀疏投影条件下具有更好的图像重建效果。 关键词：CT；压缩感知理论；不完全投影；梯度；自适应字典学习 I AB ABSTRACT ABSTRACT Owing to the advantage that CT can accurately, intuitively display the the interior information of object, CT technology has received extensive attention and been applied to various fields since the first CT was invented. How to improve the scanning speed and reduce the radiation dose to specimen has been a focus of research in CT domain. An effective way is to reduce the number of projections for reconstruction or uses sparse projection. However, the traditional methods of CT image reconstruction will cause serious image distortion, artifacts and low signal to noise ratio when the projections are sparse because of their inherent limitations. Fortunately, the theory of compressed sensing proposed in 2006 proved that we can use far fewer sampling data than the Nyquist sampling theory to get an accurate reconstruction, which provides the possibility to reconstruct image when the projections are sparse . Based on compressed sensing theory, this paper proposes a novel gradient based dictionary learning method for CT reconstruction(GradDL-CT). It applies the gradient based dictionary learning method for CT reconstruction, uses adaptive dictionary learning method to effectivly reduse the drawbacks of the populartotal variation (TV) regul