Evaluation of Traditional and 'Green Measure' Remote Sensing Techniques for Shrub Crop Assessment in Sri Lanka,

Abstract

Remote sensing techniques for agricultural inventories have generally been applied to monocatyledenous field vegetation in flat terrain. This paper evaluates the ability to monitor shrub-type vegetation in hilly terrain with Landsat. As part of the Tea Rehabilitation Project in Sri Lanka, the extent and condition of Tea plantations was mapped from February 1980 Landsat data. The Government of Sri Lanka was interested in Tea Crop condition, which was defined as a function of maturity or percent vegetative cover. The results were presented to them in April 1982. Soil albedo and slope and aspect were quite variable in the Sri Lanka study area. These factors inhibited the utility of standard maximum likelihood classifiers, particularly for sparse vegetation. Alternative techniques that are less affected by variable conditions were evaluated. These include ratios and linear transformations of MSS bands, that offer a single continuous measure of vegetative cover.

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 1982
Accession Number
ADP002060

Entities

People

  • J. S. Ott

Organizations

  • Environmental Research Institute of Michigan

Tags

DTIC Thesaurus Topics

  • Argentina
  • Environment
  • Governments
  • Inventory
  • Landforms
  • Machine Learning
  • Rehabilitation
  • Remote Sensing
  • Sri Lanka
  • Standards
  • Terrain
  • Test And Evaluation
  • Vegetation

Readers

  • Archaeological Resource Survey
  • Computer Vision.
  • Virology (or Medical Virology).