An Intelligent Vision Mobile Measurement Approach for High-speed Inspection of Large-size Flatness

Abstract

In current science and engineering, the demand for large-size surface detection has increased considerably. However, large-size surface detection presents some challenges, such as very large detection target area, discontinuous detection surface, and low detection accuracy. The current detection methods are mainly based on in-position detection and cannot meet the requirements of detection speed and accuracy for large-size surface detection. In this paper, we propose a fast detection method for large-scale flatness based on an intelligent vision mobile platform (IVMP). Specifically, by establishing the path optimization model, beam adjustment model, and largescale flatness calculation model for the IVMP, the binocular vision acquisition of large-scale target information and fast large-scale shape detection are realized. The rigidly fixed position relation of binocular vision is considered, and the parameters of the main camera can be obtained through an error equation, then the parameters of the assistant camera can be acquired quickly to calculate the three-dimensional coordinates of space points. The particle swarm optimization algorithm is integrated into the differential evolution algorithm to improve the detection speed. The IVMP is applied to the flatness detection of a satellite antenna. The experimental results show that the detection precision and efficiency of the IVMP are clearly higher than those of the laser tracking and theodolite systems.