An airborne multispectral video system was used to collect soil spectral data over a four-square mile region in northeastern Arizona. Six multispectral video images were digitized. Using the red and blue bands of each image, an unsupervised classification was performed. Each was referenced to a digitized U.S. Soil Conservation Service map resulting in classification precisions ranging from 0-92.4 percent. Ground radiometric measurements were made to ascertain spectral separability of the soil samples. Soil color was determined to try to relate Munsell value to classification precision. Misclassification of soil map units was unrelated to soil brightness or areal extent of each soil. Rather, features such as slope, boundary complexity, and surface condition was responsible for misclassifications seen in this study. Best classification results occurred when soil mapping units were relatively homogeneous, possessed slight changes in slope, and had a regular surface with smooth and distinct boundaries.
| Identifer | oai:union.ndltd.org:arizona.edu/oai:arizona.openrepository.com:10150/276549 |
| Date | January 1987 |
| Creators | Nolin, Anne Walden, 1958- |
| Publisher | The University of Arizona. |
| Source Sets | University of Arizona |
| Language | en_US |
| Detected Language | English |
| Type | text, Thesis-Reproduction (electronic) |
| Rights | Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author. |
Page generated in 0.0022 seconds