Font Size: a A A

Research On Structured Light Vision Measurement Based On Phase Shift Coded

Posted on:2015-09-23Degree:MasterType:Thesis
Country:ChinaCandidate:Z K NingFull Text:PDF
GTID:2348330518972004Subject:Engineering
Abstract/Summary:PDF Full Text Request
Due to advantages of solving the problem of matching binocular vision measurement quickly and effectively, structured light vision measurement system have been widely studied and applied by a large number of scholars. Compared with binocular vision measurement system composed by two different cameras, structured light vision measurement system has a camera and a projector composition. An encoded pattern image was projected by the projector,and then image modulated by the scene was captured by the camera. Compared the projection decoding coded image with the image that camera captured, the matching points can be identified on the two images. Finally, the object was reconstructed using the principle of triangulation with those matching points.With the development of structured light measurement technology and a variety of structured light measurement methods were proposed. Structured light measurement technology has become one of the very stable and reliable 3D measurement technology. But,,there are many shortcomings and deficiencies of structured light measurement technology,such as for the object of the movement it can not be a good measurement, for scenes captured through modulation of images it can not be both fast and stable decoding. Therefore, the structure light measurement technology has much room for improvement.This paper uses a coding method based on sinusoidal fringe grating for structured light measurement experiment. The projector projected many different standard sinusoidal gratings stripe on the object then captured the modulated striped raster image after object modification by the camera, then obtained wrapped phase diagram of each wavelength of images that camera captured by phase shift method. Then made phase unwrapping for the obtained wrapped phase diagram, and then got the absolute phase diagram of the measured object.Finally based on uniqueness of the absolute final phase and the corresponding matching points obtained by epipolar constraint, and then completed the trebling reconstruction.Therefore, the key of a three-dimensional reconstruction of the structure light measurement is how to expend the wrapped phase diagram and get the corresponding matching point fast and accurately. However, due to the complexity of depth information of the object surface and the noise of the intensity of the background light, errors and noise of phase unwrapping often occur, it was difficult to conduct phase unwrapping. therefore, this paper studies the multi-wavelength phase unwrapping algorithm and the epipolar constraint matching algorithm, the main contents are:First, the paper introduces the basic principles and methods of measurement of three-dimensional objects, focusing on structured light measurement based on sinusoidal gratings. This paper also summarized phase unwrapping algorithms of the phase measurement,and then proposed a new multi-wavelength phase unwrapping algorithms.Secondly, the paper studies the multi-wavelength phase unwrapping algorithms. Cameras captured raster image distortion stripes with different wavelengths, and then using three-step phase shift method to analysis the three images of each wavelength to obtain the wrapped phase map of each wavelength and with improved multi-wavelength phase unwrapping algorithm proposed obtained the object phase value, thus the measurement accuracy and speed had been greatly improved.Again, the article studied the principle and method based on epipolar constraint to obtain the matching points. The matrix parameters of the camera and projector were got by calibrating the camera and projector. Then according to the correspondence between the image coordinate system of the camera and image coordinate system of the projector as well as the uniqueness of absolute phase of projection vertical stripes grating finally matching pixels from projection of standard raster image and captured deformation raster images were completed. The way of seeking match point using epipolar constraint reduced the number of projection grating stripe of the projectors, also dramatically reducing the time phase unwrapping.Finally,the paper studied the principles and methods of three-dimensional reconstruction.According to the principle of three-dimensional reconstruction, a number of complex objects reconstruction experiment were conducted,and got a lot of three-dimensional point cloud images and three-dimensional reconstruction diagram of the object. And those verified the effectiveness of the proposed algorithm.
Keywords/Search Tags:multi-wavelength phase unwrapping, epipolar constraint, structured light, coding, decoding
PDF Full Text Request
Related items