期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2004
卷号：XXXV Part B1
页码：226-231
出版社：Copernicus Publications
摘要：STARIMAGER is a helicopter-borne high resolution Three-Line Scanner (TLS) imaging system developed by STARLABO Corporation, Tokyo, Japan, jointly with the University of Tokyo and the Institute of Geodesy and Photogrammetry (IGP), ETH Zurich for large scale applications, especially for line-shaped ground object recording (such as roads, rivers, railways), and remote sensing applications. As other airborne and space-borne linear array imaging sensors, STARIMAGER is equipped with GPS/IMU to record the exterior orientation parameters of each obtained image line per forward, nadir and backward views during the flight. Since a high-performance stabilizer is used, which absorbs the vibrations of the aircraft and stabilizes the camera's axis within a pixel, the resulting original images have very little waving and blurring, leading to fewer problems for post-processing. A feature of particular interest is that STARIMAGER can currently read out 10 channels simultaneously, so besides providing stereo capabilities it can also be used in multi-spectral modes. The TLS principle brings line perspective images with very little distortion in the flight direction, thus we get less occlusions especially in city and mountainous areas. It can generate seamless and high-quality ortho- images, which have already true ortho-image characteristics in flight direction. This paper focuses on the system hardware configuration and data pre-processing of the STARIMAGER, including the camera system, the stabilizer, GPS/IMU, data controlling and recording units. In addition we will report about the status of the application software development. Given the new sensor geometry a number of new features can be used in the processing algorithms, which are up to now not available on Digital Stations dealing with single frame images. This software includes a modern GUI with image handling and measuring in mono, stereo and multi-image modes. Furthermore we will address the following functions: image orientation and forward intersection, triangulation procedures with three different trajectory models, image rectification using different levels of approximation of object surfaces (plane, DSM), automated DSM generation, ortho-image generation, mono-plotting, 3D city model generation (geometric and textural parts), and support of feature and object extraction in manual and semi-automatic mode. We will also briely address a variety of applications, featuring the TLS principle in already existing and new application fields
