Automated Analysis of Spatial Grids
Environmental data are often spatial in nature. In this chapter, we will examine image processing techniques which play a key role in artificial applications operating on spatial data. These AI applications often seek to extract information from the spatial data and use that information to aid decision makers.
Consider for example, land cover data. Since different locations have different types and amounts of forestry, land cover information has to be explicitly tied to geographic location. Such spatial data may be collected either through in-situ (in place) measurements or by remote sensing over large areas. An insitu measurement of land cover, for example, would involve visiting, observing and cataloging the type of land cover at a particular location. A remotely sensed measurement of land cover might be carried out from a satellite. The remotely-sensed measurement would cover a much larger area, but would be indirect (i.e., the land coverage would have to be inferred from the satellite channels) and would be gridded (i.e., one would get only one land cover value for one pixel of the satellite image). Users of land-cover data often wish to use the data to recover higher-level information such as determining what fraction of a particular country is wooded — AI applications help provide such an answer, building on well understood image processing methods.
KeywordsLand Cover Tropical Cyclone Spatial Data Automate Analysis Radar Data
Unable to display preview. Download preview PDF.
- Dixon, M., & Wiener, G. (1993). TITAN: Thunderstorm identification, tracking, analysis and nowcasting: A radar-based methodology. Journal of Atmospheric and Oceanic Technology 106Google Scholar
- Lakshmanan, V., Rabin, R., & DeBrunner, V. (2003). Multiscale storm identification and forecast. Journal of Atmospheric Research367–380Google Scholar
- Tuttle, J., & Gall, R. (1999). A single-radar technique for estimating the winds in tropical cyclones. Bulletin of the American Meteorological Society80, 683–685Google Scholar
- Vincent, L., & Soille, P. (1991). Watersheds in digital spaces: An efficient algorithm based on immersion simulations. PAMI (13)6, 583–598Google Scholar
- Wolfson, M., Forman, B. E., Hallowell, R. G., & Moore, M. P. (1999). The growth and decay storm tracker. Eighth Conference on Aviation (pp. 58–62). Dallas, TX: American Meteorological SocietyGoogle Scholar