中文摘要：

压缩感（CompressedSensing,cs）在理论上可以为快速磁共振成像（MagncResonanceImaging．MRJ）提供一种系统的框架，即通过少量的非相干采样数据便可实现精确的图像重建。然而，在高度欠采样的情况下，重建伪影依然严重，图像细节也几近丢失，并伴有明显的图像噪声。本文提出一种基于参考制导的CS-M砌重建方法，即通过约束目标图像梯度方向的切向量与参考图像中对应位置的法向量相垂直，以使目标图像结构边缘和参考图像保持一致，提高快速MRI图像重建质量。本文运用多对比度扫描的实验数据，通过与传统的CS-MRI成像方法相比较，验证了该方法能够实现快速且高质量的MRI图像，尤其是在数据高倍率欠采样的情况下，重建的图像依然能有较好的保边效果。

英文摘要：

Theoretically, the compressed sensing （CS） provides a systematic framework for fast magnetic resonance imaging （MRI） reconstruction from under-sampled k-space data. However, with severely under-sampled pattern, aliasing artifacts are still severe; the details of image are almost lost; the image noises are obvious. A reference-guided CS-MRI reconstruction method was proposed in the paper. The tangent vector on the gradient direction of the target image was regularized to be perpendicular to the corresponding normal vector of the reference image to make the edge of target image and that of reference image consistent, improving fast MRI reconstruction quality. The multi-scan experimental data was applied to compare the proposed method with the traditional CS-MRI method. The comparative results showed the proposed method provided faster MRI of higher quality, and that the reconstructed image had better edge-persevering effect in the case of severely under-sampling.