Computer analysis of Landsat data can provide several types of useful information in urban and near-urban areas. Thematic classification gives a map of land cover or land use and multi-pass registration can detect areas of change or urban growth. These derived products have roughly the same character as Landsat; they are computer compatible, are in image format, and have a peculiar map projection. Geographic operations on these data require integration with standard digitized (vector) representations of point, line, and area data. One example is generation of maps superimposing image and boundary data. Another example is tabulation of the image data by district polygons. The approach described here uses image processing technology to perform spital or geographic operations and concentrates upon the use of the image raster as a general representation for spatial data. Operations are provided for conversion of graphics (vector) data to image format, and for linking of image data to tabular files.
CITATION STYLE
Zobrist, A. L. (1980). Mapping with census data and Landsat imagery. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 81 LNCS, pp. 49–64). Springer Verlag. https://doi.org/10.1007/3-540-09763-5_3
Mendeley helps you to discover research relevant for your work.