Registration of Terrestrial Laser Scanner Data Using Imagery

Building 3D models using terrestrial laser scanner (TLS) data is currently an active area of research, especially in the fields of heritage recording and site documentation. Multiple TLS scans are often required to generate an occlusion-free 3D model in situations where the object to be recorded has a complex geometry. The first task associated with building 3D models from laser scanner data in such cases is to transform the data from the scanner’s local coordinate system into a uniform Cartesian reference datum, which requires sufficient overlap between the scans. Many TLS systems are now supplied with an SLR-type digital camera, such that the scene to be scanned can also be photographed. The provision of overlapping imagery offers an alternative, photogrammetric means to achieve point cloud registration between adjacent scans. The images from the digital camera mounted on top of the laser scanner are used to first relatively orient the network of images, and then to transfer this orientation to the TLS stations to provide exterior orientation. The proposed approach, called the IBR method for Image-Based Registration, offers a one-step registration of the point clouds from each scanner position. In the case of multiple scans, exterior orientation is simultaneously determined for all TLS stations by bundle adjustment. This paper outlines the IBR method and discusses test results obtained with the approach. It will be shown that the photogrammetric orientation process for TLS point cloud registration is efficient and accurate, and offers a viable alternative to other approaches, such as the well-known iterative closest point algorithm. INTRODUCTION The first requirement of 3D modeling an object or scene via terrestrial laser scanning (TLS) is to transform the overlapping point clouds from adjacent scans into a uniform Cartesian reference coordinate system. One of the popular approaches adopted for this registration process is the Iterative Closest Point (ICP) algorithm for surface matching developed by Besl & MacKay (1992). Several variations of the ICP algorithm have been formulated (Chen & Medioni, 1992; Zhang, 1994; Masuda & Yokoya, 1995; Bergevin et al., 1996), including the use of sensor acquisition geometry to search for point correspondences, as proposed by Park & Subbarao (2003). Also, an initial solution for a subsequent application of the ICP method using the normal distribution transform (NDT) has been reported by Ripperda & Brenner (2005), while Gruen & Akca (2005) have suggested the Least Squares 3D Surface Matching (LS3D) method. For a review of current 3D surface registration algorithms the reader is referred to Campbell & Flynn, (2001) and Gruen & Akca (2005). Terrestrial laser scanner manufacturers generally supply an SLR-type digital camera with the laser scanner to facilitate generation of photo-realistic 3D virtual models. In this paper, a point cloud registration method is described, which makes use of the imagery accompanying the TLS point cloud to determine scanner exterior orientation and thus scan-to-scan registration. The digital camera is first photogrammetrically calibrated via a self-calibration process, after which it is mounted on the scanner. The camera’s exterior orientation with respect to the TLS reference coordinate system is then determined, again photogrammetrically. Subsequently, the photogrammetric orientation of imagery from a digital camera mounted on the TLS is performed via relative orientation or bundle adjustment in order to transform all the scanned data into a common coordinate system.