Agriculture Reference
In-Depth Information
remote sensing methods. For instance, knowledge of leaf chlorophyll content can
be obtained by analyzing its spectral absorption at various wavelengths from typi-
cal broadband radiometers, current satellite platforms, or hyperspectral sensors that
measure reflectance in narrow bands.
In the past few decades, remote sensing has shown its great potential in certain
areas of crop management such as nutrient status assessment and weed density map-
ping. Modern-day precision agriculture requires efficient and reliable determination
of the nutrition status of crops, which is mostly achieved by using remote sensing
tools. There are many advantages of acquiring data using remote sensing, such as
temporal resolution, a synoptic view, and digital formatting that allows initially ana-
lyzing large amounts of data. Crop status estimation over a wide area can be provided
by spectral information because of its high correlation to vegetation parameters, and
scientists have reported interesting developments in the estimation of crop yields on
the basis of remote sensing (Foody et al., 2003; Franklin and Hiernaux, 1991; Lu,
2005, 2006; Nelson et al., 1988; Sader et al., 1989; Santos et al., 2003; Steininger,
2000; Zheng et al., 2004). To meet society's growing need for food and materials
from crops, it is required to rapidly and accurately monitor crop yield and to achieve
site-specific management in order to reduce costs and help the environment.
Biophysical properties of crops can be estimated using space-borne, airborne,
or ground-based multispectral remote sensing methods. The hyperspectral remote
sensing technique records reflectance or emission spectra from an object of region
of interest (ROI) in hundreds of bands of the electromagnetic spectrum, whereas the
multispectral remote sensing technique records data in multiple wavelength bands.
The remote sensing system first receives electromagnetic energy that comes from
the phenomena of interests and passes through the atmosphere. The detected energy
is recorded as analog signal and is converted to digital value through an analog-to-
digital conversion. If a spacecraft platform is used, the digital data are telemetered to
a receiving station on Earth via tracking and data relay satellites, whereas when an
aircraft platform is used, the digital data are simply returned to Earth after the data
acquisition mission is completed. Some geometric and/or radiometric preprocessing
of the digital remotely sensed data is usually required to improve its interpretability.
Visual or digital image processing is used to extract biophysical and/or land cover
information, which is distributed and used to aid decision-making (Jensen, 2007).
Currently, numerous digital multispectral and hyperspectral remote sensing systems
are designed. They can be organized according to the type of remote sensing tech-
nology used for vegetation and earth resource mapping. The main criteria for selec-
tion of the most appropriate remote sensing method for site-specific management
are spatial, temporal, and spectral resolutions. The various sensing methods used to
estimate the biophysical properties of crops are highlighted in the following sections.
Satellite remote sensing has the capability of acquiring a single image covering
large areas and monitoring changes on the basis of regularly updated information
(Ahamed et al., 2011). This technique becomes more cost-effective when only low
resolution is required. Additional information related to crop status can be provided
by satellite remote sensing when nonvisible spectrum wavebands are applied. There
are several typical remote sensors used on satellites. The Multispectral Scanner
(MSS) is an optical sensor installed on Landsat 1 to Landsat 5. There are four
