基于半耦合字典的CT图像超分辨率重建方法研究

Research on CT Image Super-resolution Reconstruction Based on Semi Coupled dictionary

在CT图像超分辨率重建中,传统方法存在峰值信噪比和结构相似性低、重建效果差的问题,为获得更好的图像,提出基于半耦合字典的CT图像超分辨率重建方法。该方法利用初始图像与神经元函数之间的卷积形式,曝光处理CT图像。基于图像分辨率的离散化特点,构建了图像降质退化模型。设置匹配点和计算机断层扫描的基本参数,获取CT图像的超分辨率属性特征。经由傅里叶算法,准确匹配图像超分辨率特征点,通过训练半耦合算法,设计了图像超分辨率重建算法,实现CT图像的超分辨率重建。结果表明,该方法不仅可以提高重建的峰值信噪比和结构相似性,还提高了重建图像边缘的圆滑程度,从而实现提高重建质量,获取更好的图像。

Taking into account the problems of low peak signal-to-noise ratio and structural similarity in CT image super-resolution reconstruction in traditional methods,and poor reconstruction effect,in order to get a better image,a CT image super-resolution reconstruction method based on semi-coupled dictionary is proposed.The CT image is exposed and processed using the convolution form between the initial image and the neuron function.Based on the discretization characteristics of image resolution,an image degradation model is constructed.By setting the matching points and the basic parameters of computed tomography,the super-resolution attribute characteristics of CT images are obtained.The Fourier algorithm is used to accurately match the image super-resolution feature points,and the semi-coupling algorithm is trained to design the image super-resolution reconstruction algorithm to realize the super-resolution reconstruction of the CT image.Experimental results show that this method can not only improve the peak signal-to-noise ratio and structural similarity of the reconstruction,but also improve the smoothness of the edges of the reconstructed image to get better reconstruction effect.