3-D Object Measurement And System Calibration Of Raster Projection

3-D object measurement is an important method for 3-D object shape feature extraction. It plays an important role in many industrial fields, such as vehicle manufacture, aviation, spacecraft, electronic household products, industrial pattern, computer spare parts, automatic manufacturing, machine robot vision, CAD, medicine, industrial inspection and so on. Raster projection is one of the most prospective methods for 3-D object measurement. Recent progress of raster projection is quickly developing abroad, but the conesponding instruments is quite expensive. At the same time, equipments made in domestic are not so good. Therefore the research of the thesis and technologies is more significant and important. In this thesis the theories, approaches and its realization of raster projection 3-D imaging and 3-D object reconstruction are presented. It is proved that the system we developed has the advantages of simple architecture, high precision, fast speed, wide application environment and high degree of automation.A new method, non-separated focal, is proposed for the intrinsic and extrinsic parameters calculation. On the basis of projective theory, a mathematic modeling of camera imaging is built up for 3-D measurement of raster projection system. The relationship between objects and their images is derived. Lastly, a convenience method is realized to exchange data with other CAD and Reverse Engineering softwares.A new efficient fast circle detection method is proposed to detect feature point, the center of feature circle object. The images used to calibrate the cameras, obtained...