Using Photogrammetry to Register Laser Scanner Data
Digital imagery is now being routinely used as a complement to terrestrial laser scanners (TLS) to enable the TLS generated point clouds to be textured to produce photorealistic 3D object and scene models. As a consequence, TLS manufacturers generally offer the option of having an SLR-type digital camera mounted on the scanner. The registration of overlapping point clouds acquired from different TLS stations forms the first task of 3D model building. The provision of overlapping images from the camera mounted on the laser scanner offers an alternative, photogrammetric means to achieve point cloud registration between adjacent scans. The digital camera is first calibrated via a self-calibration process, after which it is mounted on the scanner. The camera’s exterior orientation with respect to the TLS coordinate system is then determined, again photogrammetrically. Subsequently, images from the digital camera are used to first relatively orient the network of images, and then to transfer this orientation to the TLS stations to provide exterior orientation and thus scan-to-scan registration. The proposed approach, called Image-Based Registration (IBR), offers a one-step registration of the point clouds. In the case of more than two laser scans, exterior orientation is simultaneously determined for all TLS stations by bundle adjustment of the images. Moreover, the orientation solution can be strengthened by adding further images to the network, these being recorded with the camera removed from the TLS. Reported test results obtained with the IBR show that the process is efficient, practical and accurate, and offers a viable alternative to approaches such as the well-known iterative closest point (ICP) algorithm. Unlike the ICP algorithm, the IBR method does not require any overlap between the separate TLS point clouds.