MULTISECTRAL SENSING OF FOREST TREE SPECIES

COMPUTER RECOGNITION OF FOREST TREE SPECIES AT THE NASA-ANN ARBOR FORESTRY TEST SITE HAS BEEN ACCOMPLISHED USING DATA COLLECTED IN SIX SPECTRAL REGIONS BETWEEN 0.4 AND 1.0 MICROMETERS. THE SIX WAVELENGTH BANDS USED WERE SELECTED ON THE BASIS OF LABORATORY REFLECTANCE DATA PREVIOUSLY COLLECTED BY THE AUTHORS. DATA OBTAINED WITH THE UNIVERSITY OF MICHIGAN C-47 AIRCRAFT AND PROCESSED WITH THE UNIVERSITY OF MICHIGAN SPECTRAL ANALYSIS AND RECOGNITION COMPUTER (SPARC), PROVIDED SUCCESSFUL SEPARATION OF CONIFEROUS AND BROADLEAVE TREES. SPECIFIC RECOGNITION AND SEPARATION OF SUGAR MAPLE, BLACK WALNUT, BLACK LOCUST, RED OAK, AND WHITE OAK WERE ALSO SUCCESSFUL. DISCRIMINATION AMONG CONIFERS WAS NOT SO SUCCESSFUL AS FOR BROADLEAVED SPECIES, BUT SPRUCE WERE CONSISTENTLY SEPARATED FROM PINE.

  • Authors:
    • Rhode, W G
  • Publication Date: 1972-12

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 00202999
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Aug 15 1973 12:00AM