In order to achieve 3D digital measurement of large-scale equipment, based on OpenMVG (Open Multiple View Geometry), a technical framework, which is suitable for industrial photogrammetry, is generated. Firstly, the coded points in the set of images were identified and positioned, the matching relationships between multiple images were built by ID number, and the data interface for OpenMVG was established. Then, by means of OpenMVG solver, the 3D coordinates of the coded points at the normalized scale were solved. The actual scale factor was restored by using multiple scale bars arranged in the scene. Finally, the coordinates of the code points, the transformation matrix between the images and the camera parameters were obtained, which constitute the core parameters of the industrial photogrammetry framework and can meet many applications of industrial photogrammetry. Taking an airplane as a test object, which is in the space of 10m × 3m × 4m, the number of images by photogrammetry capturing is more than 500. The finally measured error with our method is less than 0.085mm. Compared with the current commercial photogrammetry system, the proposed method can achieve the same accuracy level with greater efficiency and expansibility.