Interpreting Aerial Images: A Knowledge-Level Analysis
Abstract
Many image understanding systems rely heavily on a priori knowledge of their domain of application, drawing parallels with and exploiting techniques developed in the wider field of knowledge-based systems (KBSs). Attempts, typified by the KADS/CommonKADS projects, have recently been made to develop a structured, knowledge engineering approach to KBS development. Those working in image understanding, however, continue to employ 151 generation KBS methods. The current paper presents an analysis of existing image understanding systems; specifically those concerned with aerial image interpretation, from a knowledge engineering perspective. Attention is focused on the relationship between the structure of the systems considered and the existing KADS/CommonKADS models of expertise, sometimes called “generic task models”. Mappings are identified between each system and an appropriate task model, identifying common inference structures and use of knowledge.
Keywords
Task Model Aerial Image Simple Classification Remote Sensing Image Inference StructurePreview
Unable to display preview. Download preview PDF.
References
- 1.W.J. Clancey, “Heuristic Classification,” Artificial Intelligence, vol. 27, pp. 215–251,1985Google Scholar
- 2.S. J. Cosby and R. Thomas, “IRS: A Hierarchical Knowledge Based System for aerial Image Interpretation,” 3rd International Conference on Industrial Engineering Applications of Artificial Intelligence and Expert Systems, Charleston, SC, USA, July 16–19, 1990Google Scholar
- 3.D. Crevier, and R. Lepage, “Knowledge-Based Image Understanding Systems: A Survey,” Computer Vision & Image Understanding, vol. 67, no. 2, pp. 161–185, Aug. 1997CrossRefGoogle Scholar
- 4.R. D. Ferrant Multi-Spectral Image Analysis System Conference on Artificial Intelligence, Denver, Co, USA, 1984Google Scholar
- 5.T. Matsuyama and V. S. Hwang,SIGMA: A Knowledge-Based Aerial Image Understanding System. New York, Plenum Press, 1990Google Scholar
- 6.L. Moller-Jensen, “Knowledge-Based Classification of an Urban Area Using Texture and Context Information in Landsat-EM Imagery,”Photogrammetric Engineering & Remote Sensing, vol. 56, no. 6, June 1990, pp. 889–904Google Scholar
- 7.G. Schreiber, et. aI.,Knowledge Engineering and Management: The CommonKADS Methodology. Cambridge, Mass.: MIT Press, 1999Google Scholar
- 8.H. Murai and S. Omatu, Remote Sensing Image Analysis Using a Neural network & Knowledge-Based Processing International Journal of Remote Sensing, vol. 18, no. 4, May 1997, pp. 811–828CrossRefGoogle Scholar
- 9.G. Schreiber, B. Wielinga and J. Breuker, (ed.),KADS: A Principled Approach to Knowledge-Based System Development. London: Academic Press Ltd., 1993Google Scholar
- 10.D. S. W. Tansley and C. C. Hayball,Knowledge-Based Systems Analysis & Design: A KADS Developer’s Handbook. Hertfordshire, Hemel Hempstead: Prentice Hall International (UK) Ltd., 1993Google Scholar
- 11.J. Ton et. al.: “Knowledge-Based Segmentation of Landsat Images”, IEEE Transactions on Geoscience & Remote Sensing, vol. 29, no. 2, March 1991Google Scholar
- 12.S. W. Wharton, Spectral-Knowledge-Based Approach for Urban Land-Cover Discrimination, IEEE Transactions on Geoscience & Remote Sensing, vol. 25, no. 3, May 1987, pp. 273–282CrossRefGoogle Scholar
- 13.B. Wielinga, A. Th. Schreiber and J. A. Breuker, “KADS: A Modeling Approach to Knowledge Engineering,” Knowledge Acquisition, vol. 5, pp. 5–53, 1992CrossRefGoogle Scholar