In this thesis, a binocular stereo vision system is constructed for 3D image reconstruction. We propose a new method called “support value” to solve correspondence problems. Geometric constraints are added to reduce the ambiguities in stereo matching. Left/right consistency check is used to increase the confidence for matching results. Post-processing helps remove the outliers and make original dense disparity map smooth and reliable. The three-dimensional characteristics can be reconstructed by perspective projection.  

        The 3D image reconstruction algorithm is successfully applied to a micro manipulation system for cell image reconstruction. Cell images need to be segmented first to do stereo matching tasks. An automatic threshold selection and boundary extraction algorithm are used to detect positions and contours of cells. By combining image segmentation techniques and stereo matching procedure, 3D reconstruction of microscopic cell images can be performed. The 3D fine structures of cells can also be recovered. The 3D-display monitoring system will provide multi-functionalities visual effect to facilitate the manipulation process.