Fast and Accurate Self-calibration Using Vanishing Point Detection in Manmade Environments
- PDF / 2,955,633 Bytes
- 12 Pages / 594.77 x 793.026 pts Page_size
- 49 Downloads / 185 Views
ISSN:1598-6446 eISSN:2005-4092 http://www.springer.com/12555
Fast and Accurate Self-calibration Using Vanishing Point Detection in Manmade Environments Sang Jun Lee and Sung Soo Hwang* Abstract: Interests of auto-calibration have been increased in several camera systems. This paper presents a novel self-calibration method using fast and accurate vanishing point detection algorithm that works in manmade environments. The proposed algorithm estimates focal length assuming that the principal point is the center of an image to satisfy the orthogonality of three vanishing points. By using proposed vanishing point detection algorithm and minimization of the proposed objective function, the proposed system detects accurate vanishing points with focal length outperforming other methods. The proposed vanishing point detection algorithm detects vanishing points by using J-linkage based method that is more delicate by fragmentation and re-merging strategies. The proposed objective function finally detects vanishing points that meets orthogonality among estimated hypotheses for vanishing points by checking several geometric relationships. We believe that the proposed method can be used for automatic camera calibration, localization of a camera in an autonomous navigation system, and three-dimensional reconstruction of a single-view image. Keywords: Auto camera calibration, line clustering, single view geometry, vanishing point detection.
1.
INTRODUCTION
Camera calibration is the essential issue for ubiquitous applications in robotics and control such as visual servoing [1] and autonomous navigation system [2]. For examples, position-based visual servoing or hybrid visual servoing takes visual information from 3D geometry for feedback control of the pose and motion of robot. Thus, it highly depends on the quality of accurate intrinsic parameters, and it is sensitive to calibration errors [1]. Moreover, visual SLAM or visual odometry system [2] needs accurate calibration parameters since this system is basically operated on a map including 3D points generated by camera poses with calibration matrix. Thus, estimating accurate calibration matrix is important problem for theses applications that need data association between 3D-2D points. While various methods are proposed for camera calibration using different approaches such as Morphological pattern [3], laser [4] calibration matrix by using vanishing points [8–10]. Vanishing points are similar to the principal of image projection since an ideal point in 3D space is projected in finite 2D point by camera calibration matrix. Therefore, given orthogonal three vanishing points according to three axes x, y, and z, a relevant camera cali-
bration matrix can be estimated as discussed in [5]. Conventional approaches for vanishing point detection are as follows: Vanishing points are detected as intersection points from each cluster of line segments in the image space that are parallel in the world coordinate system. To estimate line clusters easily, most approaches are operated on M
Data Loading...
