Research On Portable 3D Measurement Technology Based On Hilbert-huang

Fourier transform profilometry(FTP)can realize three-dimensional(3-D)reconstruction through a fringe image,because fast speed and non-contact become one of the research hotspots in the field of optical measurement.The portable and economic type of 3D measurement system is a very meaningful research direction and has wide application prospects.To this end,we developed a portable 3D reconstruction system using a smart phone and a micro projector.The fringe pattern was projected onto the surface of the object to be measured by a micro projector,and then the distortion fringe image was captured by the smart phone to complete the 3D imaging measurement.However,the recorded image has distortion and noise problems,which seriously reduces the accuracy of 3D reconstruction.In addition,the Fourier transform loses part of the high-frequency information,and it is impossible to accurately measure object which sharp changes in shape.How to carry out high-precision 3D reconstruction becomes the core problem of the paper.The title of the paper is " Research on Portable 3D Measurement Technology Based on Hilbert-Huang",using a smart phone,micro projector to build a FTP system,and using Hilbert-Huang theory to complete 3D Image reconstruction.it can be improves the accuracy of the system while suppressing zero-order terms and background noise.The specific contents of this paper are as follows:Firstly,the 3D measurement theory of objects 3D measure and the process of 3D reconstruction are analyzed based on the basic optical knowledge.The characteristics of Fourier Transform Profilometry and Phase Measurement Profilometry are analyzed.FTP is selected as the theoretical basis of this paper in order to realize 3D reconstruction after capture an image.Secondly,Further improvement of the fast empirical mode decomposition algorithm and the modified Hilbert-Huang algorithm combined with FTP for use in portable 3D measurement.The modified Hilbert-Huang algorithm improves the accuracy of Fourier transform profilometry by removing direct-current components and suppressing background noise.In this paper,the modified Hilbert-Huang algorithm is used to reconstruct the fringe image,which overcomes the pattern mode mixing phenomenon of the traditional Hilbert Huang algorithm and shortens the running time of the traditional algorithm.The amplitude distribution map of the Hilbert spiral transform is used to mark the noise.The simulation results show that the modified Hilbert-Huang algorithm is better than the traditional Hilbert-Huang algorithm,and the accuracy of the 3D reconstruction of Hilbert-Huang profilometry is higher than that of FTP.Finally,the portable 3D measurement light path is used to collect the fringe image,and the influence of the intelligent equipment and experimental environment on the collected fringe image quality is analyzed in detail,and an improved scheme is proposed.By comparing the traditional calibration methods,the Zhang Zhengyou calibration method and the Brown optimization method are selected to calibrate the smart device,and the verification calibration method and mobile smart device can meet the experimental requirements.The maximum error of the portable 3D measurement based on Hilbert-Huang profilometry is 0.89%,which proves that the 3D reconstruction of the system achieves high precision.