Page Banner

United States Department of Agriculture

Agricultural Research Service

Title: Detection of Pitted Morningglory with Hyperspectral Remote Sensing. Ii. Effects of Vegetation Ground Cover and Reflectance Properties

Authors
item Koger Iii, Clifford
item Shaw, D - MISS STATE UNIV
item Reddy, Krishna
item Bruck, L - MISS STATE UNIV

Submitted to: Weed Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: August 26, 2003
Publication Date: March 1, 2004
Citation: Koger III, C.H., Shaw, D.R., Reddy, K.N., Bruck, L.M. 2003. Detection of pitted morningglory with hyperspectral remote sensing. II. effects of vegetation ground cover and reflectance properties. Weed Science. 52:230-235.

Interpretive Summary: Varying amounts of vegetation ground-cover often exists between weed-free and weed-infested areas of row-crop fields. To determine the effect of varying degree of vegetation ground-cover on weed detection capabilities, field studies were conducted at the Southern Weed Science Research Unit, Stoneville, MS and Plant Science Research Center, Starkville, MS. Plots containing weed-free soybean and soybean infested with pitted morningglory, a troublesome broadleaf weeds species, were established in each study. Hyperspectral data were collected when pitted morningglory plants had 1-, 3-, and 5-leaves. Eight 50-nm-wide spectral bands derived from each hyperspectral measurement were used to detect weed-free soybean from soybean intermixed with pitted morningglory. Detection accuracy was 92 to 100% regardless of pitted morningglory plant size and whether equal or unequal amounts of vegetation ground-cover existed in weed-free soybean and soybean intermixed with pitted morningglory plots. Results indicate that hyperspectral remote sensing has potential to detect weeds even when various amounts of vegetation ground-cover exist between weed-free and weed-infested areas of row-crop fields.

Technical Abstract: Field experiments were conducted to evaluate the potential of hyperspectral reflectance data for discriminating soybean intermixed with pitted morningglory and weed-free soybean when similar and different proportions of vegetation ground cover existed in plots of soybean intermixed with pitted morningglory and weed-free soybean. Hyperspectral data were collected with a hand-held spectroradiometer when pitted morningglory was in the 1-leaf, 3-leaf, and 5-leaf growth stage. Eight 50-nm-wide spectral bands (1 ultraviolet, 2 visible, 4 near-infrared, 1 mid-infrared) derived from each hyperspectral reflectance measurement were used as discrimination variables in discriminating weed-free soybean and soybean intermixed with pitted morningglory. Discriminant accuracy was 92 to 100% regardless of pitted morningglory growth stage and whether equal or unequal proportions of reflectance from vegetation existed in weed-free soybean and soybean intermixed with pitted morningglory. Reflectance in the near-infrared spectrum was higher for weed-free soybean compared to soybean intermixed with pitted morningglory, and this difference influenced the ability to discriminate weed-free soybean from soybean intermixed with pitted morningglory.

Last Modified: 7/30/2014
Footer Content Back to Top of Page