Font Size: a A A

Research On Structured Light 3D Measuring Technology And System Based On Digital Fringe Projection

Posted on:2010-07-01Degree:DoctorType:Dissertation
Country:ChinaCandidate:Z W LiFull Text:PDF
GTID:1118360275486682Subject:Materials Processing Engineering
Abstract/Summary:PDF Full Text Request
Phase measuring profilometry is one of the most widly used structured light 3D measuring techniques, it has huge potential for applications in many areas, including industrial inspection, product quality control, reverse engineering, medicine, virtual reality, culture heritage and human body measurement etc. Most commercial phase measuring systems employ expensive physical grating to project fringe images. The high production cost limits their application in numerous small and medium-sized enterprises. Therefore, we systematic research the key techniques of the structure lighted system based on digital fringe projection, including phase calculation, phase error compensation, system parameters calibration, phase-heigt mapping and color texture acquisition, and dedicate to develop low cost, high accurate and high stability structured light 3D measuring system, which is suitable for small and medium-sized enterprises, and extent its application ranges.Phase calculation, including relative phase calculation and phase unwrapping, is the basic of the phase measurement technique. We describe phase-shifting method and phase unwrapping algorithm, and choose four-step phase-shifting method and multi-frequency heterodyne principle as the calculation method. Experimental results demonstrate that the method can accomplish phase calculation of complex and discontinuous objects with high precision.During the actual phase calculation process, the non-sinusoidal attribute of the captured fringe image, caused by the gamma nonlinear of the DLP projector and the nonlinear response of the CCD camera, is the mainly error source. The non-sinusoidal process is described, and the phase error attributes is analyzed through theoretical analysis, simulation experiment and practical measurement. A general phase error compensation algorithm based on phase error look-up-table is proposed, our experimental results demonstrate that by using the proposed method the phase error is extremely reduced.System parameters calibration is one of the key issues in the structured light system. We propose a simple and high accurate system calibration method. This method endows the projector capturing capabilities by establishing the correspondence between camera pixels and projector pixels. It treats the projector as a camera to unify the calibration procedures of a structured light system and a classic stereo vision system. This method is simple and effective, the calibration precision can achieve 0.26 pixels, the absolute measurement accuracy is 0.05mm, and relative accuracy is 1:5000.Though this system calibration method is quite simple and fast, it is difficult to find the best camera model for the system, especially for the projector, thus the calibration precision of this kind of method is still limited. Therefore, a neural network, which has powerful ability of function approximation, is employed to solve the complex nonlinear phase-height mapping issue. The structure of the neural network and the training principle is introduced. By comparing the convergence rate and training precision of the neural network before and after phase error compensation, we indicates that phase error influences the precision of the trained network, so phase error compensation is necessary. On the other hand, as the phase error can be compensated during sample data set collection, this phase-height mapping method is not sensitive to the non-sinusoidal attribute of the fringe image. The 3D measurement experiments demonstrate that the neural network method can achieve higher measurement precision.In the fields of culture heritage, medical imaging, and virtual reality etc, color textures have important information, so the 3-D measurement systems are required to acquire the 3-D shape and color texture, simultaneously. Inspired by color-space definition, we use three color images to combine the color texture of the target, and a simple white balance algorithm is proposed to optimize the color texture.Based on above key techniques, we have developed two types of structured light system, and they have been used in industrial inspection, human body measurement and culture heritage etc. The practical applications indicate the developed structured 3D measurement system is characterized with predominant performance, obvious price advantages and great marketing potential.
Keywords/Search Tags:Structured light 3D measurement, Phase measurement, Phase calculation, Phase error compensation, System parameters calibration, Phase-height mapping, Color texture acquisition
PDF Full Text Request
Related items