Three-dimensional Reconstruction Of Key Technologies Based On Binocular Stereo Vision

Binocular stereo vision is used to restoration of three-dimensional structure of a scene based on geometry principle requiring two cameras to observe the identical scenery shown at a few different orientations. Along with the development of computer technology, stereo vision is becoming a research hotspot of the computer vision domain, and will also be widely applied in intelligent robot, the production automation, the medicine domains, etc. Therefore the research of this subject has the important theory significance and the practical value.This paper based on Marr vision theory has focused on the research of three components of binocular stereo vision: camera calibration, stereo match and 3D reconstruction. Considering experimental instruments and environment, Tsai's method is adopted to get the exterior and inner parameter in camera calibration; Due to fixed window area-based stereo algorithms have given rise to relatively large matching errors around depth discontinuities, a adaptive window for stereo matching algorithm using integral image was proposed. In this algorithm, a new similarity function and Four-direction line mask were employed not only to reduce noise effects but also to improve matching accuracy while the integral image technology was used to make computing matching cost constant time. Meanwhile, a GA-based stereo matching algorithm using rectangle matching window was improved to gain more accurately disparity map. Two algorithms mentioned upon could be used in more accurately situation or faster situation respectively. After camera calibration and stereo match, a simplified formula was used to calculate point's 3D coordinates in order to reconstruct the structure of object, and display the image drawn through OpenGL at last.For the purpose of reconstructing the structure of 3D scene and proving practicability of these algorithms proposed upon, a hardware and software platform which including image processing, camera calibration, stereo match, 3D reconstruction was designed and put up. All of the work lay a good foundation of further research and development of binocular stereo vision system.