Peer Reviewed Articles
909 Correspondence Analysis for Principal Components Transformation
of Multispectral and Hyperspectral Digital Images
James R. Carr and Korblaah Matanawi
Abstract
Download
Full Article
Correspondence analysis is introduced for principal components
transformation of multispectral and hyperspectral digital
images. This method relies on squared deviations between
pixel values and their expected values (joint probabilities
computed as the product of the sum of all pixels in one
spectral bond and the sum of pixel values across all bands
at a given pixel position). Correspondence analysis is applied
to a multispectral SPOT High Resolution Visible (HRV) in age
of Eleuthera, Bahamas. Correspondence analysis, principal
components analysis, and factor analysis (standardized principal
components) yield similar transformations. Correspondence
analysis, however, compresses more image variance
into fewer principal components. For the particular SPOT HRV
scene chosen, correspondence analysis captures 96 percent
of the original image variance in its first principal component.
Used in a lossy image compression algorithm to reconstruct
the original set of three SPOT HRV images, this first
principal component from correspondence analysis restores
spectral content better than does principal components analysis.
915 Object Recognition Based on Boundary Description
Y. Huang and J.C. Trinder
Abstract
Download
Full Article
Three-dimensional(3D) object recognition is a difficult and
yet important problem in computer vision. It is a necessary
step in many industrial applications, such as the identification
of industrial parts and the automation of the manufacturing
process. and it is essential for intelligent robots
equipped with powerful visual feedback systems. In this paper,
a procedure is described to recognize 3D objects, using
model-based recognition techniques. Objects in the scene are
reconstructed by digital photogrammetry, while models in
the database are generated by a CAD system. They are all described
in a boundary representation. A detailed comparison
between the potential matching graphs of a model and the
object determines the identification of the sensed object, and
its position and orientation.
923 Topographic Effects on the Texture of High-Resolution
Forest-Stand Images Measured by the Semivariogram
Benoît St-Onge
Abstract
Download
Full Article
By studying the statistical relationship between terrain gradient
and the range of the semivariogram of simulated high-
resolution images of forest stands, we assessed the effects of
topography on estimates of tree size and density obtained
through texture measures. Three-dimensional computer models
of hardwood and softwood forest stands were overlaid on
slopes of varying gradient. Using a geometrical-optical approach,
one-meter-resolution images were generated in 120 series
representing different combinations of forest types and
sun-terrain geometry. The range of the semivariogram of these
images was measured in four directions and its relation with
gradient was evaluated through regression. Results show that
topography affects texture mostly in sparse stands, and that
the gradient-induced absolute error in the estimates of tree
size and density is low.
937 A Moisture Index for Surface Characterization over a Semiarid
Area
Lesley-Ann Dupigny-Giroux and John E. Lewis
Abstract
Download
Full Article
A multispectral index, designed to describe surface moisture
characteristics, was derived from the information content at
the blue, near-infrared, and thermal wavelengths of Landsat
Thematic Mapper imagery. The index is given by an openended
triangle within which features of varying moisture
consents are located. Although the index bears some resemblance
to existing soil and vegetation indices as well as to
the Tasseled Cap transformation, it differs in the way in
which moisture, brightness, and vegetation information can
be expressed in one locational space. The index was found
to vary as a function of changes in the season, in vegetation
cover, and in moisture conditions. The index was also found
to be sensitive to the spatial resolution at which it was described.
947 Deforestation in North-Central Yucatan (1985-1995): Mapping
Secondary Succession of Forest and Agricultural Land Use in Sotuta
Using the Cosine of the Angle Concept
Youngsinn Sohn, Emilio Moran, and Francisco Gurri
Abstract
Download
Full Article
A new spectral pattern matching approach that utilizes the
spectral angle (the cosine of the angle) concept was used for
mapping deforestation and successional stages of forest regrowth
in Sotuta in the state of Yucatan, Mexico. By culcu-
lating spectral angles between finely defined spectral clusters
and known reference signatures, and assigning each spectral
cluster to one of the reference classes based on the minimum
spectral angle rule, we were able to map forest regrowth
stages and agricultural land-use classes. Our research shows
that, by adapting a spectral pattern matching approach demonstrated
in this paper, spectral clusters can be assigned into
information classes precisely and objectively, end the inconsistency
involved in visual interpretations can be avoided.
The conceptual difference between the spectral distance and
spectral angle in feature space is also reviewed,
In the study area, the rate of deforestation is high and agricultural
land use is intensifying increasingly.The limited
amount of land granted to ejidos and rapid population growth
seem to be major causes of deforestation in the study area.
| Top | Home |