为实现惯性约束聚变实验中核心部件球形靶丸表面形貌的高精度、高效率、无遗漏检测，建立了基于子孔径拼接和衍射干涉技术相结合的靶丸形貌测量装置，给出了测量的基本原理及系统单孔径横向分辨率的计算方法，并对靶丸旋转扫描时子孔径的排布进行了划分。针对点云数据横向间隔过大导致匹配算法迭代收敛速度缓慢，甚至不收敛的问题，提出以虚拟的极小半径代替实际曲率半径，将点云进行横向压缩的方法，在保持形貌特征不变的同时提高算法的匹配速度和准确性。最后，对直径1 mm的镀金球靶进行了实际测量，得到拼接后相对形貌误差峰谷值和均方根值分别为1.332λ和0.479λ的检测结果，可以同时获得表面形貌的整体起伏特征，以及局部的形貌细节。

In order to inspect the surface topography of spherical target accurately, efficiently and exhaustively, which is the core component in inertial confinement fusion experiment, a shape detection system is founded based on the technologies of sub-aperture stitching and diffraction interferometry. The basic testing principle and the calculation method of lateral resolution in once single measurement are provided. The configuration of sub-aperture is planned for the rotation scanning of the target. In order to solve the problem that the matching algorithm shows a slow convergence rate or even misconvergence caused by the overlarge lateral distance between cloud data points, a method of data points transverse compression is proposed, which replaces the actual radius of curvature by a virtual tiny radius. This method can improve the matching rate and the accuracy of the algorithm with the topography feature remaining unchanged. At last, a gold coated target with a diameter of 1mm is measured actually, and the PV value and RMS value of the relative topography errors after stitched are 1.332λ and 0.479λ respectively. The macroscopic fluctuation characteristics and the partial topography detail can be obtained, and the refactored target surface has a good visual quality.