The Research Of Super-resolution Reconstruction Algorithm For MRI Image Based On Compressive Sensing

Magnetic resonance imaging is one of the most important tools in clinical digital image diagnosis and treatment.It is widely used in disease detection and diagnosis.But in the actual process,the resolution of MRI image will be limited by some factors such as signal to noise ratio,equipment performance,scanning time.Moreover,the reconstruction algorithm also will affect its image resolution,and high-resolution means spenting more software cost.This paper uses image super-resolution reconstruction technique to obtain higher resolution MRI image.By combining image super-resolution reconstruction technique with compressive sensing theory,a super-resolution reconstruction model of MRI image under the framework of compressive sensing is established,and the MRI image super-resolution reconstruction algorithm based on compressive sensing is proposed.Meanwhile,considering the characteristics of MRI image and the limitations of joint dictionary learning,this paper also proposes an improved algorithm for MRI image super-resolution reconstruction of compressive sensing based on local information and the simulation experiment is carried out.The experimental results show that the improved algorithm proposed in this paper fully considers the characteristics of MRI image,it has less calculation,and better reconstruction performance than traditional reconstruction methods.Compared with the previous super-resolution reconstruction algorithm for MRI image based on compressive sensing,the PSNR and SSIM of the image obtained increases by 0.3747 and 0.0078 respectively.It shows that the reconstructed MRI image by the improved algorithm has higher resolution and better visual effect,and it has certain clinical application value.