Machine-assisted analysis of Landsat data in the study of crop-soils relationships
Links
- Document: Report (pdf)
- Download citation as: RIS | Dublin Core
Abstract
To date, relatively few studies have dealt with crop-soil interactions as they affect the appearance of agricultural areas on Landsat imagery, and hence crop and soil classification or the analysis of agricultural land use.
The Image 100, a computer-based data analysis system which allows an interpreter to interact directly and rapidly with Landsat computer compatible tape data, provided a tool to assist in the evaluation of the extent and significance of these interactions. Used with timely and accurate ground data, the system made possible a determination of the variability in crop spectral appearance, from soil type to soil type, as recorded on Landsat data. Information was provided in the form of spectral distribution histrograms for each crop-soil class on each Landsat band. Several crop categories in a test area in rookings County, South Dakota, were classified using training fields that were selected to be representative of each major crop-soil class. Accuracies in each case, on a total acreage basis, were greater than 90 percent.
Study Area
| Publication type | Report |
|---|---|
| Publication Subtype | USGS Numbered Series |
| Title | Machine-assisted analysis of Landsat data in the study of crop-soils relationships |
| Series title | Open-File Report |
| Series number | 76-603 |
| DOI | 10.3133/ofr76603 |
| Year Published | 1976 |
| Language | English |
| Publisher | U.S. Geological Survey |
| Contributing office(s) | Earth Resources Observation and Science (EROS) Center |
| Description | 19 p. |
| Country | United States |
| State | South Dakota |
| County | Brookings County |
| Online Only (Y/N) | N |
| Additional Online Files (Y/N) | N |
| Google Analytic Metrics | Metrics page |