Papers by Author: Jun Yi Lin

Abstract: It is a difficult task to get enough numbers of highly accurate control points for projector calibration in line structured light vision system. A new projector calibration method based on binocular stereo vision is proposed in this paper. Two cameras calibration can be done usingtraditional camera calibration method and they composed a binocular stereo vision. In projector calibration procedure, a planar template was located at several different positions in front of the stereo vision system. Two cameras captured the stripe images in each position simultaneously. Every center points of the laser stripe can be used as control points of the projector plane. The 3D coordinate of the stripe center points can be obtained through binocular stereo vision principle easily. So the light plane can be calculated quickly. Experiments were carried out and the result shows that the proposed method is flexible and stable.

Abstract: In this paper, a method combining stereo vision and digital phase-shifting techniques is presented, which can measure the profile of complex freeform surfaces expeditiously and accurately. A chessboard is used to determine the camera parameters for the camera calibration of the system and the epipolar line rectification is executed based on the stereo calibration result. The stereo matching is processed with the combination of four-step phase-shifting and Gray code techniques, and the maximum probability method is proposed to remove the fringe order confusion problem between Gray code and phase shifting measurement. Bilinear interpolation method is also adopted to improve the speed and accuracy of stereo matching. Experiments were carried out with measuring a step gauge block and a shoe last model. Experimental results show that the measurement can be completed within 3 s for a measuring region of 640 mm × 480 mm, and the average measurement error is less than 0.054 mm.

Abstract: The global calibration of multi-sensor measurement system is difficult to accomplish, usually, it needs external high-precision measurement equipment. A novel global calibration method based on common planar template is proposed. Stereo vision combined with laser linear-structured light is adopted to constitute a 3D vision sensor system (3DVSS). Stereo calibration of the 3DVSS can be finished by the planar template calibration method. For the global calibration of multiple vision sensors, it only needs a common planar template which is one of the planar templates can be seen synchronously by every camera in 3DVSS. The homography matrix between the coordinate systems of the 3DVSS and the common planer template can be calculated. The global coordinate is established in the common planar template, so the local coordinate system (LCS) of each 3DVSS can be transformed to the global coordinate system (GCS) through the common planar template. Two experiments have been used to test the proposed method, and the experimental results show that the proposed method is flexible and effective.

Abstract: A New Stereo Matching Method Based Phase Dynamic Programming (PDP) is presented in this paper. This method, based on gray code value and phase value, implements stereo matching through finding an optimal path on a 2D plane composed of cross phase absolute difference in the same gray region. Before matching, mask-image is adopted to reduce matching data point. During matching, the matching strategy is optimized by adopting sequence consistency technique as well as defining the path orientation. After matching, the error data point is eliminated by matching consistency and median filtering based disparity. Finally, experimental results show this algorithm can get a better performance. Matching process of PDP is simple but effective than others. It not only can get continuous and dense parallax but also eliminate effects of noise and the reconstructive object has high reliability, better authenticity and high precision.

Abstract: Phase-based Stereo Matching (PSM) is an efficient matching method in binocular stereo vision. It bases on correspondence points which have the same phase-distribution. Sine raster generating algorithm is a basic and important step in PSM study. This paper advances that using defocusing rectify-binary patterns (DRBP) method substituted the focusing-sinusoidal-patterns (FSP) method. It has following advantages: (1) No gamma correction is required; (2) The object illuminance measurement is not desired; (3) The influence of noise is minified. There are same phases distribution in correspond points of the pair matching lines by experiment analysis. The feasibility of using DRBP to generate sine raster in PSM is demonstrated.