Camera Calibration for 3D Computer Vision
| Autor: | Ting-Hsuan Liu, 劉定宣 |
|---|---|
| Rok vydání: | 2003 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 91 The technology of computer vision is broadly applied in many aspects such as industrial inspection, medical image processing, remote sensing and nanotechnology. By analyzing the images provided by two difference cameras, the desired target position corresponding to the world coordinate can be obtained. Due to the image distortion, abd quantization error of the processing algorithm, the required information may not be obtained straightforward. Therefore, the camera calibration is important to 3D information reconstruction. In the thesis, Zhang’s calibration applied to calibration the camera system parameters. This method only requires the camera to observe a planar pattern from a few different orientations. Either the camera or the planar pattern attitude can be altered. The corresponding motion need not be available. The camera intrinsic parameters, extrinsic parameters, and the parameters of radial lens distortion can be obtained under a systematic procedure. The experimental error is constrained to sub-pixel scale. Associating with camera calibration, the constraints of Epipolar Geometry is applied to obtain the mapping of two different images, and to reconstruct the world coordinate. From the experimental results, it reveals that the increase of baseline can improve the error of depth effectively. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|