Study On The Key Technologies Of 3D Shape And Deformation Measurement Based On Industrial Photogrammetry And Computer Vision

In recent years, 3D shape and deformation measurement has become more and more important in many fields, such as quality control, reverse engineering, material testing, biomedicine, cultural relics protection and so on. As we all know, the traditional measuring content and measuring method can not meet the needs of practice. Acquisition of three dimensional geometry, surface shape and deformation data has already been the hot topics in the industrial measurement fields. The key technologies of 3D shape and deformation measurement are deeply and systematically researched in this thesis by means of combining theoretical analysis and experimental study, which are based on mechanical engineering, optical electronics, computer vision, digital image processing, photogrammetry and other disciplines. The main contents and achievements are listed as follows:(1) The technical principle, advantages and disadvantages, the present research and development trends of the 3D shape and deformation measurement are reviewed and discussed. The fundamental theory of computer vision and close range photogrammetry including coordinate transformation, photography geometric constraints, space intersection and three-dimensional reconstruction are explored in detail, which provide a good theoretical basis for our research.(2) Camera calibration in the computer vision system is one of the key technologies in 3D measurements. The calibration precision directly determines the accuracy of the 3D reconstruction. A flexible calibration method is proposed based on photogrammetry technique in order to improve the precision of the computer vision system. A cross-shaped calibration pattern which is portable and easy to be manufactured is designed. The cross target can be used to calibrate stereo vision systems with large measurement volume and obtain higher measurement precision conveniently. The mathematical model of the stereo vision system with 10 distortion parameters for each camera is presented. The calibration initial values are computed using the relative orientation method and the Direct Linear Transform (DLT) method of photogrammetry. The bundle adjustment algorithm is used to optimize the calibration parameters as well as the 3D coordinates of the feature points. Experiment results show that the RMS error of the reprojection in our method is less than 0.05 pixels and the distance measurement error is 0.031 mm with a high precision scale bar which length is 221.001±0.003mm.(3) A high accuracy method for 3D reconstruction is deeply researched which can be used to single-camera measuring, dual-camera measuring and multi-camera measuring. The 3D reconstruction equations in different situations are derived respectively. A stereo vision system is proposed based on digital fringe projection. A heterodyne multiple frequency phase-shifting algorithm is employed for overcoming the unwrapping problem of phase functions and for a reliable unwrapping procedure. According to the phase unwrapping algorithm of the heterodyne multiple frequency phase-shift technology, the phase unwrapping process is mainly dependent on phase principal value of the different frequency sinusoidal grating images. It is not sensitive to object surface color and needs no surface treatment. The redundant phase information is used to increase the accuracy of the 3D reconstruction.(4) According to large-scale structure deformation, a method of 3D static deformation measurement is presented. A correspondence points matching algorithm is researched in detail. The coordinate registration algorithm is improved to increase the efficiency and accuracy of the coordinate registration. A matching method based on correlation principle is proposed, which can realize coordinate registration and deformation points matching with good efficiency. Moreover, the deformation points matching of the static deformation measurement is implemented, which can be used to displacement field measurement.(5) According to 3D displacement and strain measurement, the principles and key technologies of the digital image correlation method (DICM) are systematically researched including correlation coefficient, correlation search, correlation area model and displacement field smoothing. A method of uneven illumination correction is proposed. It can be used to improve the accuracy of dynamic deformation measurement. The speckle matching and detection methods are implemented with high efficiency based on the improved DICM.According to above theories and technologies, the systems we developed including binocular surface scanning system, static deformation measurement system, the dynamic deformation measurement system have been used in 3D shape measurement for large airplane, automobile panel inspection, real-sized transmission tower loading test, welding deformation measurement of sheet metal, tensile test, and the results are encouraging.