No abstract

No abstract

Consumer-grade digital cameras have become commonplace and with calibration there is significant potential for their use in photogrammetric applications. This paper describes a new method for automatic camera calibration, which is suited to colour digital cameras. Two innovations are central to the developed process. The first is exploitation of RGB colour attributes in the design of automatically detected and identified coded targets. The second is a new approach to initial value determination for relative orientation, which then supports fully automatic initial network exterior orientation and subsequent bundle adjustment with self-calibration. Accuracy aspects are also touched upon and the impact of chromatic aberration is discussed. Finally, an example of the automatic calibration process, which is implemented in the iWitness system, is presented for the case of multiple zoom settings of a consumer-grade digital camera.

An understanding of the evolution of cracks in concrete structures due to long term natural deformation is important to civil engineers, but quantitative measurements can be difficult to make. However, digital imaging offers a potential solution. This short paper illustrates the operational application of automated image processing techniques for accurate, multi-temporal crack measurements. The first part of this paper provides an overview of automatic feature extraction, essential for automatic crack detection. The latter part describes the methods developed for detecting and measuring cracks. Due to the long term nature of the application, operational results have yet to be finalised, although sample results are presented.K: automation, crack monitoring, feature extraction I F  has been the cornerstone of many applications in photogrammetry and remote sensing for several years (Fö rstner, 1993). The inevitable trend towards automation in all things digital has meant that research into automatic feature extraction has received considerable attention for some time now. Although significant progress has been made in many areas of automatic feature extraction (for example, Firestone et al., 1996; Sowmya and Trinder, 2000), transfer of those algorithms from the research community to the commercial domain has been slow.In the photogrammetry and remote sensing research communities, automatic feature extraction is being used for a considerable number of different applications.

An investigation of the accuracy potential of Ikonos 1 m satellite imagery is reported. Three sensor orientation/triangulation models are applied to stereo-and three-image configurations of ''Geo'' imagery with the aim of achieving 3D geopositioning to sub-metre accuracy. The models considered comprise rational functions with bias compensation, affine projection and the direct linear transformation. Test results from the Melbourne Ikonos Testfield are reported and these show that with modest provision of good quality ground control, Ikonos ''Geo'' imagery can yield 3D object-point determination to an accuracy of 0·5 m in planimetry and 0·7 m in height. The accuracy achieved is not only consistent with expectations for rigorous sensor orientation models, but is also readily attainable in practice with only a small number of ground control points being required.K: 3D geopositioning accuracy, high-resolution satellites, Ikonos imagery, rational functions, sensor orientation I I    in this journal, Hanley and Fraser (2001) discussed the accuracy potential of Ikonos satellite imagery, as indicated by two-dimensional (2D) imageto object-space transformations within a precision testfield. It was shown that planimetric positioning to better than 0·5 m (half-pixel ) accuracy was achievable with Ikonos ''Geo'' imagery, using relatively few ground control points (GCPs) and allowing for height effects. Moreover, the high precision achieved indicated that the imaging system was free of significant non-linearities, which augured well for subsequent application of linear sensor orientation models for three-dimensional (3D) ground point determination from stereo and multi-image Ikonos coverage. The present article constitutes a sequel to the earlier paper, in that it extends the investigation to 3D geopositioning. Specifically, the application of three distinct methods for triangulating ground feature points from two-and three-image Ikonos ''Geo'' data sets is investigated. The three methods employ rational functions and the linear affine and direct linear transformation (DLT ) models. The aim of the