Research On Structured Light 3D Measurement Algorithm And System Based On Grating Projection

In recent years,with the continuous progress and development of science and technology,the exploration of optical 3D measurement technology has become more and more in-depth,and 3D measurement technology based on structured light projection has emerged,which has the advantages of non-contact,high speed,high accuracy and encodability,and has been widely used in the fields of reverse engineering,industrial inspection,restoration of cultural relics,medical inspection,unmanned vehicles and virtual reality.At this stage,scientists’ research focuses on fast 3D measurement of dynamic objects,and there is a lack of systematic and in-depth research on system calibration,phase calculation and unfolding.In the process of developing optical 3D measurement technology toward high accuracy,high speed and complex environment measurement,etc.,the influence of external environment leads to measurement errors,while the system’s phase-solving algorithm also faces many problems and challenges.Based on this,this paper researches and analyzes the key issues such as system parameter calibration,phase calculation,phase height mapping,imaging influencing factors and imaging system development,and makes a comparative demonstration for different algorithms,and analyzes the quality of structured light 3D measurement.The research content of this paper can be summarized as follows:1.System parameter calibration:The experiment uses a projection system to generate structured light stripes and a camera to capture deformation stripes.The position and pose of the projection system and the camera determine the height of the object,and the two need to establish a correspondence to calculate the real height information of the object.The camera and the projection system realize the conversion between 3D world information and 2D pixel information,the conversion involves the conversion process of world coordinate system,camera coordinate system,image coordinate system and pixel coordinate system respectively.The experiments use Zhang Zhengyou’s tessellation calibration method to calculate the pose of the camera,and by analyzing the Gamma effect of the projector and the inverse camera principle,the data obtained from the calculation is used to correct the pattern and improve the accuracy of the optical 3D measurement.2.Phase calculation of stripes:After the structured light stripes are generated by the projection system,they are deformed by modulation on the object surface,and the object surface information is stored in the deformed stripes.In this paper,the phase information of the deformed stripes is calculated using the phase shift method and the Fourier transform method,respectively.The wrapped phase is solved to obtain the wrapped phase,and then the wrapped phase is expanded using the phase expansion algorithm to compare the accuracy of the phase calculation of different methods.Among them,the Fourier transform contour technique can realize the calculation of the phase information of the object using only one image,but its spectrum is easy to be overlapped which it’s difficult to calculate the complex deformed objects;the phase shift method uses the phase shift technique to project multiple phase-shifted stripes for the calculation of the phase information.Its algorithm is simple and insensitive to external noise,but it reduces the measurement efficiency because it needs to project multiple stripes patterns.The above two methods use the inverse trigonometric function arctan when calculating the stripe phase information,and the phase information is wrapped in(-π/2,π/2),which needs to be calculated by phase expansion.In this paper,we use the spatial phase expansion algorithm and the temporal phase expansion algorithm for phase expansion calculation,and compare the performance of different phase expansion algorithms,among which the triple frequency external difference method has excellent characteristics such as high accuracy and insensitivity to noise,but it is used with the phase shift method The 2D-SRNCP phase unfolding algorithm,which is based on reliability ranking and follows a discontinuous path,shows excellent performance in dealing with the noise that destroys the true wrapped phase image.3.Phase height mapping algorithm:The acquisition of the true height of an object requires not only the calculation of the phase of the deformation stripe,but also the establishment of the correspondence between the phase and the height of the object.In this paper,three imaging methods are introduced,and the relationship between the measured object height and spatial frequency and object-image distance is calculated using the imaging method of intersecting the projected optical axis with the camera optical axis within the error allowed.The experiments in this paper use the triangulation method for measurement,and the correspondence between the object height and phase information is calculated by the calibration data.4.Analysis of the factors affecting the optical three-dimensional measurement:the phase calculation is accurate and error-free when the three-dimensional measurement of the object obtained from the computer simulation,but the acquisition of information will be affected by various noises when the actual measurement is carried out,such as environmental random noise,equipment out of focus,etc.Meanwhile,the surface height of the object will also affect the acquisition of phase information,and when the surface is too complex or too smooth,it will produce shadows,occlusions.When the surface is too complex or too smooth,it will produce shadows,occlusions,reflections and other phenomena,resulting in truncation of phase information.The above problems make the phase information distorted,and the wrong phase information will be transmitted in the phase calculation.This paper also analyzes the sensitivity of the algorithm noise and compares the advantages and disadvantages of different phase calculation algorithms and phase expansion algorithms.5.Development and application of structured light three-dimensional measurement system:This paper designs a measurement system based on the principle of structured light three-dimensional measurement,uses laser as the light source,calculates and deduces the information accuracy of sinusoidal structured light stripes,and proves the accuracy and stability of the system by experimentally shooting and processing the target object.