摘要: 三维图像的处理和操作需要将一般的断层序列插值成为具有各坐标轴一致的分辨率的体数据，而目前最常用的线性插值方法在层间距较大时会导致图像边缘模糊和出现伪影。Penney根据现有的非刚体匹配方法，提出了利用图像形变场数据的插值算法，明显提高了层间插值的质量。本研究对Penney提出的算法进行了改进，用多分辨率B样条拟合插值方法替代Penney使用的最邻近直线插值方法，并将新算法的实验结果与原算法、线性插值进行了对比。新算法既保持了高质量插值又提高了计算速度，并且使生成的图像达到了C 2 平滑。此算法可用于精度要求比较高的体数据插值重建过程。

    关键词: 图像插值;非刚体配准;体数据重建;多分辨率B样条

    Nonrigid Registration Based on Multi-level B-spline Interpolation Approach

    ZHU Yang-xing，BAO Xu-dong

    (Laboratory of Image Science and Technology，Southeast University，Nanjing Jiangsu210096，China)

    Abstract:For common medical slices(e.g.CT or MR)，the space between neighboring pixels in a slice is often smaller than the center-to-center slice separation，while in most3D image processing，analysis and visualization tasks it requires voxels to be isotropic.Many in-terpolation techniques have been proposed for processing such data.The commonly used grey level interpolation can produce artifacts or blurred contours when two given slices shift considerably.More reasonable interpolation results is achieved by shape-based methods，especially the nonrigid registration based methods.Rueckert′s registration method is took to produce deformed data fields ......