Enhanced phase measurement profilometry for industrial 3D inspection automation

Abstract

Industrial metrology and inspection systems commonly rely on phase measurement profilometry (PMP) using sinusoidal fringe patterns projecting, yielding dense, and accurate 3D reconstruction regardless of the presence of texture. However, applying PMP method to industrial 3D inspection is still a big challenging problem due to rigorous industrial measurement conditions including large surface reflectivity variation range and vibration. Aiming to solve these problems, an enhanced phase measurement profilometry (EPMP) is proposed. In EPMP, an optimal exposure time (OET) calibration method is proposed to solve large surface reflectivity variation range problem, and it can avoid saturating the camera sensor in areas of specular reflection while keep the signal-to-noise ratio (SNR) of fringe image in areas of weak reflection at most. To resist the influence of vibration, an improved pose calibration method (IPC) is used to allow fast calibration of pose of cameras by acquiring only one image of planar target. Moreover, an automatic online 3D inspection system for evaluating 3D geometric dimension quality of railway truck adapter (RTA) is developed, and according to the experiments, the EPMP indicates a satisfactory result in accuracy and repeatability, which can meet the requirements of the 3D inspection task in industrial measurement conditions.