Repository logo
Log In(current)
  1. Home
  2. Colleges & Schools
  3. Graduate School
  4. Masters Theses
  5. Forest land use classification from ERTS-1 imagery
Details

Forest land use classification from ERTS-1 imagery

Date Issued
June 1, 1974
Author(s)
Birth, Elvin E.
Advisor(s)
John C. Rennie
Additional Advisor(s)
E. R. Buckner
G. R. Wells
Permanent URI
https://trace.tennessee.edu/handle/20.500.14382/44661
Abstract

The preparation of thematic maps and tabular summaries defining the maps is basic to forest land use decision making. Because objectives differ according to the decision being made, raw input data should be used to prepare the maps and tables. The Earth Resource Technology Satellite (ERTS-1) collects data over large areas and the data are available to any prospective user. Minimum collection of ground truth and handling of both ERTS-1 and ground data with a readily available set of com-puter programs permits classification of land use, forest types and volume classes. Orientation with the ground can be maintained in order to prepare the maps. Frequencies of classifications can be used to prepare tabular summaries. The classification system used consisted of a computer mapping program, a discriminant analysis classification program and chi-square testing of results. The system was complemented by a forest inventory program. In the dis-criminant analysis procedure, an option was used which per-mits user participation by assigning "prior probabilities." A test of the system in Polk County, Tennessee, using ERTS-1 multispectral scanner channels 6 and 7 data acquired on October 15, 1972, showed acceptable results in classifying land use and forest type. Results of classifying volume were less acceptable because volume is a continuous function and discriminant analysis is applicable to discrete functions. The use of aircraft imagery as ground truth for land use classification was acceptable. Aircraft imagery for forest type classification was also acceptable, but forest type classification from ground truth was better when both sets of results were analyzed by the chi-square test of a contingency table. An important conclusion drawn was that ERTS-1 digital tapes should have been used instead of using photographic reproductions in a microdensitometer. Each generation of data results in some degradation. The most important con-clusion drawn was that prospective users of ERTS-1 imagery can construct a system from available computer programs and prepare thematic maps and tables. It is also possible to exert a degree of user control into the system

Degree
Master of Science
Major
Forestry
File(s)
Thumbnail Image
Name

Thesis74B5783.pdf

Size

33.67 MB

Format

Unknown

Checksum (MD5)

499d11718e5ccd4060ba37cfc1893a83


University Libraries

1015 Volunteer Boulevard
Knoxville, TN 37996
865-974-4351

Map & Directions
Donate to the Libraries
  • About
  • John C. Hodges Society
  • Speaking Volumes magazine
  • Outreach
  • Directory
  • Employment
  • Policies
  • Library Intranet
University of Tennessee power T logo

The University of Tennessee, Knoxville
Knoxville, Tennessee 37996
865-974-1000

Events
A-Z
Apply
Privacy
Map
Directory
Give to UT
Accessibility

Built with DSpace-CRIS software - Extension maintained and optimized by 4Science