Agriculture Reference
In-Depth Information
the vegetation and the interaction is exploited and analyzed, such as in LIDAR (light
detection and ranging).
Remote sensing for detecting and estimating severity of plant diseases is used at
three altitudes or levels above the crop canopy. At the lowest altitude, within 1.5-2.0
m above crop height, hand-held multispectral radiometers or multiple waveband
video cameras are used; at 75-1500 m, aerial photography is used, whereas at the
highest altitude, satellite imagery is employed utilizing satellites orbiting at 650-850
km above the earth's surface. In addition, video image analysis systems, such as that
described by Lindow and Webb (1983), (Fig. 2.9), which uses a video camera
interfaced through a digitizer to a microcomputer and display monitor, can be used
under laboratory conditions for measuring diseased or damaged tissue at close
quarters; systems such as the Delta-T Devices WinDIAS true-colour Windows-
based system are able to differentiate the primary colours of diseased and healthy
tissue (brown, yellow and green) in order to analyze percentage diseased leaf area
automatically. In 2002, image analysis software called ASSESS was made available
by The American Phytopathological Society for plant disease quantification. The
software was optimized for the measurement of leaf area, percent area infected,
lesion/pustule count, root length and ground cover. ASSESS relies on the Hue-
Saturation-Intensity colour model enabling the user to effectively extract the leaf
from the background and then the lesions from the leaf. Bannon and Cooke (1998)
used NIH Image 1.60/fat (National Institute of Mental Health, USA), a public
domain image analysis programme for the Apple Macintosh, to assess splashed
areas on white card in simulated studies on the dispersal of
Septoria tritici
pycnidiospores; the NIH programme can be used for a wide variety of image
analyses including the measurement of diseased areas on plant organs.
Whereas hand-held multispectral radiometers or multiple waveband video
cameras are most appropriate for disease measurements on plants or pots within
fields, aerial infrared photography is most useful at field level, and satellite imagery
has been used since 1972 for large areas or regions of the earth's surface devoted to
agriculture and forestry. Images are transmitted to earth stations by satellites such as
the American
N
ational
O
ceanic and
A
tmospheric
A
dministration (NOAA) and
LANDSAT series (1, 4 and 5), and the French SPOT satellite (which uses 10 metre
resolution imagery), that feature advanced very high resolution radiometer
(AVHRR) optical and thermal sensors; these have been joined by IRS, Ikonos and
EROS satellites. However, the importance of ground truth, that is actual visits to the
target crop to verify remote sensing data, is an important part of the process. The
persistence of cloud cover in countries such as the UK and Brazil has been a serious
impediment to the progress of this technology; however synthetic aperture radar
(SAR) high-resolution technology can overcome this problem and was used in 1991
on board the European Remote Sensing Satellite ERS-1. An excellent overview of
digital imaging was published by Graham in 1998.
The application of remote sensing to plant pathology lies mainly in the detection
of crop stress. A plant or plant population becomes stressed when a biotic or abiotic
factor adversely affects growth and development (Nilsson, 1995). Stress or disease can
be expressed in various ways, such as imbalance in water supply leading to stom-
atal closure, decreased photosynthesis with associated changes in leaf fluorescence
