We proposed a subaperture stitching method for concave–convex microstructured surfaces. The method is mainly based on wavelet transform (WT) and three-dimensional (3D) data matching. WT is applied to separate the high-frequency information from the low-frequency information, then the speeded-up robust feature-random sample consensus algorithm is used to match 3D surface data and calculate lateral displacement between adjacent subapertures. Finally, the pose is adjusted to obtain the stitching result. Simulations and experiments show that the proposed method can effectively stitch concave–convex microstructured surfaces.