Measurement of contact patterns is a widely employed method for evaluating gear transmission and meshing performance in aviation spiral bevel gear manufacturing enterprises. Nevertheless, directly segmenting and extracting these contact patterns from point clouds or images is not feasible, as each tooth of the spiral bevel gear possesses a complex three-dimensional(3D) spiral surface and intricate texture details. To meet this measurement requirement, it is necessary to utilize both the gear's color texture information and point cloud coordinate information simultaneously. In other words, multidimensional information of the gear tooth surfaces needs to be fusion and calibrated. This paper introduced a high-precision multidimensional calibration methodology for spiral bevel gears. This methodology leverages the fusion of image and point-cloud data, encompassing two crucial phases. Firstly, dense point clouds from the 3D calibration block are transformed into a planar representation via the bird's-eye view (BEV) encoding approach to facilitate corner detection. Secondly, a two-stage corner detection algorithm, which exploits the calibration block's intrinsic geometric features, is employed for rapid and precise identification of corners. The corners that have been determined are used to convert coordinates from the pixel-based system to that of the linear laser system. The corners thus identified facilitate the computation of a transformation matrix, which translates from the pixel coordinate system to the line laser coordinate system. Utilizing this transformation matrix, image textures are accurately projected and merged with the 3D geometry. Also, point clouds can be precisely projected onto two-dimensional (2D) images. The proposed methodology was verified through its application to the multidimensional information fusion of tooth surfaces on 12 different types of aviation spiral bevel gears. The experimental outcomes indicated that the reprojection error associated with this method remains below 0.1 mm.