ABSTRACT
Recognising geographic objects in imagery offers much potential for automating land use change detection and for constructing and visualising three-dimensional scenes (Georgopoulos et al., 1997), but it is a hugely complex task. The use of height information extracted from stereo-images is seen to offer a valuable extra clue as part of a wider recognition strategy. The aim of this chapter is to consider the reliability of such height information using a range of surveying methods to provide ground truth.
