Canadian Forest Service Publications

SEIDAM for Forestry: Intelligent fusion and analysis of multi-temporal imaging spectrometer data. 1996. Bhogal, A.S. (Pal); Goodenough, D.G.; Charlebois, D.; Matwin, S.; Portigal, F.; Barclay, H.J.; Thomson, A.J.; Niemann, O. Pages 59-62 in Proceedings for the 26th International Symposium on Remote Sensing of Environment, March 25, 1996. -, Vancouver, BC.

Year: 1996

Issued by: Pacific Forestry Centre

Catalog ID: 21315

Language: English

CFS Availability: Not available through the CFS (click for more information).

Mark record

Abstract

SEIDAM (System of Experts for Intelligent Data Management) intelligently fuses remotely sensed, GIS, and field data for answering queries about the forests and the environment of British Columbia. SEIDAM experts draw on image analysis, GIS and database systems in order to address user queries using case-based reasoning. SEIDAM relies on software agents to carry out sophisticated processing tasks on remotely sensed data. Software agents assist the users in this processing by providing the necessary information to successfully complete the tasks. Theses tasks include calibration, atmospheric and topographic corrections and geocoding of high spectral-resolution imaging spectrometer data. The resulting geocoded imagery is subjected to image processing algorithms (which include band-moment analysis, spectral endmember analysis and sophisticated classifiers) and fused with GIS data in order to extract information on the forest objects. In this paper, we discuss the applications of SEIDAM expert systems to forestry using multi-temporal AVIRUS data acquired over SEIDAM test sites in 1993 and 1994.