Page 36 - Digital Analysis of Remotely Sensed Imagery

Overview 9

can be regarded as its acceptable substitute, commonly known as the
ground reference. Evaluation of the accuracy of a classification may
be undertaken for each of the categories identified and its confusion
with other covers, as well as for all the categories. The outcome of
accuracy assessment is usually presented in a table that reveals
accuracy for each cover category and for all categories as a whole.

1.3.5 Change Detection
Change detection takes remote sensing to the next stage, during
which results from respective analysis of remotely sensed data are
compared with each other, either spatially or nonspatially. This is
commonly known as multitemporal remote sensing that attempts to
identify what has changed on the ground. Change may be detected
from multitemporal remotely sensed data using different methods,
all of which are covered in this book. A number of issues relating to
change detection (e.g., operating environment, accuracy, and ease of
operation) and their impact on the accuracy of detected results are
examined in depth, as well.

1.3.6 Integrated Analysis
In addition to satellite imagery data, non-remote sensing data have
been increasingly incorporated into digital image analysis to over-
come one of the limitations identified above, namely, to make use of
more image elements in the decision making so that classification
results can be more accurate. Many kinds of ancillary data, such as
topographic, cadastral, and environmental, have found use in image
analysis. Different methods have been developed to integrate them
with remotely sensed data for a wide range of purposes, such as
development of more accurate databases and more efficient means of
data acquisition. This book explores the various methods by which
different sources of data may be integrated to fulfill specific image
analysis objectives.

1.4 Preliminary Knowledge

1.4.1 Pixel
Formed from the combination of picture and element, pixel is the
fundamental building block of a digital image. An image is composed
of a regularly spaced array of pixels (Fig. 1.3). All pixels have a
common shape of square, even though triangle and hexagon are also
possible. When a pixel is stored in a computer, it is represented as an
integer. In this sense, a pixel does not have any size. Nevertheless, a
pixel still has a physical size. Also known as cell size, it refers to the
ground area from which the reflected or emitted electromagnetic
radiation is integrated and recorded as a single value in the image