Structured Light Three-dimensional Measurement System Construction And Error Correction Research

Structured light three-dimensional measurement technology is an active vision measurement technology.By projecting the designed pattern,the space coordinate value of the measured object's outer surface can be calculated from the deformed pattern collected by the camera,so the surface of the solid geometry can be measured and represented.Because of its non-contact,non-destructive,high precision,fast speed and other advantages,it is widely used in industrial measurement,art design,virtual reality and other fields.In the development of products at this stage,the performance of domestic structured light three-dimensional measurement equipment still needs to be improved,and further research is needed.Aiming at improving the performance of structured light three-dimensional measurement system,this paper studies at system construction,phase calculation,system calibration and phase error correction,so that improve the measurement accuracy and speed of structured light three-dimensional measurement system.The structured light three-dimensional measurement system of is mainly composed of camera and projector.This paper studies the mathematical modeling and hardware design of the three-dimensional measurement system of structured light by phase-shift method,including:selection of projectors,cameras and lenses according to the requirements of measurement range and accuracy,realizing the imaging control of DLP optical projector,trigger control of synchronous shooting of DLP optical projector with industrial camera.In terms of software,the measurement flow of the whole experiment is designed.In image processing and calculation,the phase calculation and data processing are carried out by using MATLAB.Finally,a three-dimensional measurement and point cloud data reconstruction system is formed,so that the complete workflow of the three-dimensional reconstruction system is realized.The encoding method adopted by the structured light three-dimensional measurement system determines the measurement accuracy of the corresponding points that the system can achieve.This paper studies the encoding and unwrapping algorithms of phase-shifting structured light measurement,including two most commonly used encoding methods,Gray code combined with phase-shifting method and multi-frequency heterodyne method.It is found that the most widely used multi-frequency heterodyne method is vulnerable to noise interference,which will affect the accuracy of measurement.To solve this problem,a backward phase unwrapping method for solving absolute phase is proposed based on the multi-frequency heterodyne method.This method can get the absolute phase of the first group of frequency gratings from the final heterodyne phase.Experiments show that this method can effectively reduce the noise interference in the measure-ment process.The calibration of structured light three-dimensional measurement system is the basis of measurement,which has great influence on the speed and accuracy of the whole system.The principle and method of phase calibration and camera calibration are studied in this paper.Aiming at the shortcomings of traditional corner extraction methods,such as large computation and slow extraction speed,a fast calibration method based on circular array is proposed.The measurement results show that the speed of this method is 93.1%higher than that of corner extraction method while ensuring the accuracy.Gamma nonlinearity error means that the grating projected by projector does not satisfy sinusoidal distribution,which results in periodic fluctuation error between the calculated object phase and its ideal phase.To solve this problem,an active gamma distortion correction method based on polynomial fitting is proposed to improve the measurement accuracy.It has the advantage of not limited to specific models compared with gamma value precoding method and passive error compensation method.The results show that the average phase error is reduced by 78.07%,which confirms the effectiveness of the gamma distortion correction algorithm designed in this paper.