An Introduction to Remotely Sensed Data Analysis - Sampling Spatial Units for Agricultural Surveys - page 64

each pixel corresponds to a numeric value that represents a gray level and describes
the spectral luminosity of a specific area of the scene.
RS techniques are widely used in agriculture and agronomy (Dorigo et al. 2007 ).
In fact, remotely sensed images provide a spatial coverage of a field, and can be
used as a proxy for measuring crop and soil attributes (Fitzgerald et al. 2006 ). In
many developing countries and over most of the oceans, satellite data are the only
source of quantitative information of the atmosphere and the Earth ' s surface. Thus,

it is an invaluable source of real-time severe weather information.
RS is needed because agriculture monitoring has specific challenges that are not
found in other economic sectors (The World Bank 2011 ). First, agricultural pro-
duction heavily depends on seasonal patterns related to the life cycle of crops.
Secondly, production varies according to the physical landscape (i.e., soil type),
climatic conditions, and agricultural management practices. Finally, agricultural
variables vary substantially over space and time. For these reasons, agricultural
monitoring systems must be timely. RS has many advantages in that it can signif-
icantly help to address these needs, it is appropriate for collecting information over
large areas, and can have a high revisit frequency.
RS has been progressively used for standardized, faster, and possibly cheaper
methods for agricultural statistics. Many countries have RS programs that support
their official agricultural developing countries in Africa, Southeast Asia, and Latin
America.
Today, agricultural intelligence is needed to address various social requirements
such as national and international agricultural policies. Additionally, global agri-
cultural organizations that deal with food security issues greatly depend on reliable
and timely crop production information (Becker-Reshef et al. 2010 ).
In this chapter, we consider remotely sensed images from aircraft or satellite. In
particular, we discuss some statistical techniques for improving and interpreting
these images. The layout of this chapter is as follows. Section 4.2 contains a brief
review of the basic concepts of RS. In Sect. 4.3 we describe image restoration with
particular reference to geometric and radiometric correction. Section 4.4 discusses
image enhancement. Section 4.5 contains a description of the problem of multi-
spectral transformations. In Sect. 4.6 we outline the thematic extraction of infor-
mation. Section 4.7 contains a brief discussion of the possible applications of
GRASS to image analysis. Finally, last section concludes the chapter and contains
some examples of RS applications to agricultural spatial sampling.

4.2 Basic Concepts

Researchers collect information about the Earth s characteristics to formulate

models and validate hypotheses. This information can be recorded by analysts
who are on site , i.e., who are directly observing the phenomenon under investiga-
tion, either by special sensors or RS methods.
In the past, aerial photography was the principal RS tool. It was based on analog
devices. However, technology has progressed, and now this method has been

'