A 3D scanning approach is normally used for inspecting the geometrical tolerances of manufactured products, especially when there is a limit of deformations allowed for the acceptance of the product. Allowed deformation is usually determined by the allowable area, depth, and location of deformation on the product. There are usually different demands set for different parts of the product (face, edge, side, etc.).
There are two basic principles of 3D scanning methods. First, a laser profiler can be used, where the laser line is projected onto the product and captured by one area scan camera. This way with each image, the shape of the laser line on the object is analysed by calculating the distances of each point on the projected line. The laser profiler method can be used in continuous applications, as it does not have to conform to a fixed field of view.
The second principle is a so-called ‘stereo vision and structured light projection,’ which is basically an advanced version of a laser profiler. Instead of one area scan camera, a pair of them is used, and instead of a laser line, a 2D matrix is projected onto the surface. Each line is captured from two different angles, which results in a 3D point cloud of the scanned object.
The results of both methods are 3D point clouds, which are then further analysed. Deformities or other points of interest are filtered out, measured, and compared with the tolerances as per the customer’s needs.