数字体图像相关方法通过分析两幅以体素为单位的三维体图像获得物体内部的三维变形信息。在利用数字体图像相关方法测量物体内部变形的过程中，整体素的位移场可以通过简单的空域搜索获得，因此如何计算亚体素位移是提高该方法位移测量精度的关键。介绍了一种基于空间灰度梯度的亚体素位移测量算法，并用计算机模拟生成的可精确控制位移的三维散斑体图像和真实的激光扫描共焦显微镜三维体图像测试了该算法的计算精度和计算效率。结果表明，基于梯度的亚体素位移测量算法原理简单，计算效率高，可用于实际物体内部变形的高精度测量。

Digital volume correlation (DVC) measures the internal deformation of an object by analyzing two volumetric images in unit of voxel. In the practical implementation of DVC, integer-voxel displacement can be readily acquired by use of a simple searching scheme performed in spatial domain. How to achieve sub-voxel accuracy has been considered as the key to improve the accuracy of DVC measurements. A spatial-gradient based sub-voxel algorithm is proposed, and the principle of the algorithm is described in detail. The sub-voxel algorithm is investigated in terms of accuracy and computational efficiency using computer generated volume data and actual volume images recorded using laser scanning confocal microscope. Results show that the proposed algorithm can be used as a simple yet effective technique for accurate internal deformation measurement.