| High-resolution, real-time 3D shape measurement for dynamically deformable objects has a huge potential for applications in many areas, including entertainment, security, design and manufacturing, etc. However, due to the challenging nature of the problem, no system with such capability has ever been developed. The focus of this dissertation research is to develop such a system and to demonstrate its practical value for applications in many fields.; The system we develop is based on a digital fringe projection and phase-shifting technique. It utilizes a single-chip Digital-Light-Processing (DLP) projector to project computer generated fringe patterns onto the object and a high-speed Charge-Coupled-Device (CCD) camera synchronized with the projector to acquire the fringe images at a frame rate of 120 frames per second. Based on a three-step phase-shifting technique, each frame of the 3D shape is reconstructed using three consecutive fringe images. Therefore the 3D data acquisition speed of the system is 40 frames per second. Together with fast 3D reconstruction algorithms and parallel processing software we developed, high-resolution, real-time 3D shape measurement is realized at a frame rate of up to 40 frames per second and a resolution of 532 x 500 points per frame.; Real-time 3D reconstruction is difficult if the traditional sinusoidal three-step phase shifting algorithm is used with an ordinary personal computer. Therefore, we developed a novel phase shifting algorithm, namely, trapezoidal phase-shifting algorithm, for real-time 3D shape measurement. This new algorithm replaces the calculation of a computationally more time-consuming arctangent function with a simple intensity ratio calculation, thus boosting the processing speed by at least 4.5 times when compared to the traditional sinusoidal algorithm. With this algorithm, 3D reconstruction in real time was shown to be feasible.; One shortcoming of the trapezoidal phase-shifting algorithm is that the measurement accuracy is affected by image defocus, which limits the dynamic range of measurement. Even though the error caused by image defocus is rather small, especially when compared with other intensity ratio based methods, this error has to be eliminated if high accuracy measurement is desired. In this research, we found that we could use the trapezoidal algorithm to process sinusoidal fringe images with a small error and then use a LUT method to eliminate the error. The result is a new algorithm, namely, fast phase-wrapping algorithm, which is 3.4 times faster than and just as accurate as the traditional algorithm. Essentially this new algorithm combines the speed advantage of the trapezoidal algorithm and the accuracy advantage of the traditional algorithm. By implementing this algorithm in our system, we were able to achieve real-time 3D reconstruction with high accuracy. (Abstract shortened by UMI.)... |
