Premium
Determining Density of Maize Canopy from Digitized Photographic Data 1
Author(s) -
Stoner E. R.,
Baumgardner M. F.,
Swain P. H.
Publication year - 1976
Publication title -
agronomy journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.752
H-Index - 131
eISSN - 1435-0645
pISSN - 0002-1962
DOI - 10.2134/agronj1976.00021962006800010015x
Subject(s) - remote sensing , canopy , environmental science , vegetation (pathology) , biomass (ecology) , mathematics , soil science , agronomy , geography , medicine , archaeology , pathology , biology
Plant and soil scientists have devised many methods with varying degrees of success to determine the density of vegetative cover. Canopy density is important in estimating yield of biomass, characterizing vegetative stress, predicting soil losses from erosion, and planning resource management programs. The objective of this investigation was to examine quantitatively the relationship between the densities of maize ( Zea mays L.) canopies and the energy reflected by them. Spectral and spatial data were obtained from color and color infrared (IR) photographs from 10 m above field plots of four growth stages of maize on dark and light‐colored soils. Leaf area index (LAI) measurements were made in the field. Photographs were used for determining percent ground cover 1) by a point grid technique and 2) by microdensity scanning of the three separate emulsion layers of the color photographs and digital analysis of the resulting data. A statistical pattern recognition computer program was used to determine the spectral separability of the digitized photographic data and to classify the data into two categories—green vegetation and soil. Of the emulsion layers of the color IR photographs, classification of the digitized infrared wavelength band best represented the actual percent ground cover. A plot having an LAI of 4.44 had a ground cover of 96.5% as determined by the point grid method and 93.2% by digital analysis of IR data. Classification of data from the visible red (0.59 to 0.71 μ m) overestimated percent ground cover and from the green (0.47 to 0.61 μ m) underestimated percent ground cover.