NASA NTRS · 19800063574
Unsupervised classification of MSS Landsat data for mapping spatially complex vegetation
Abstract
The various stages in carrying out a monocluster block unsupervised classification using Landsat MSS data are described. Procedures for carrying out these various stages were found to be far from well-established for the type of terrain being investigated, which is rugged and contains many small land cover units. Two particular difficulties were encountered: first, that of precise ground location of pixels; and, secondly, that of objectively evaluating the results. Ways in which these can be surmounted are suggested.
Keep this discovery
Explore connections, maps & timelines
Townshend, J. R. G., Justice, C. O.. 1980-06-01. Unsupervised classification of MSS Landsat data for mapping spatially complex vegetation. https://ntrs.nasa.gov/citations/19800063574
Cite the original work for its findings. Save a collection to share your selection of sources.