Peer-Reviewed Articles
169 Automatic Registration and Mosaicking for Airborne
Multispectral Image Sequences
Qian Du, Nareenart Raksuntorn, Adnan Orduyilmaz,
and Lori M. Bruce
Abstract Download
Full Article
Airborne remote sensing has important applications in
agriculture monitoring because of the flexibility of system
deployment. The major obstacle in practical use is its high
cost. To reduce the cost, a multispectral system can be
assembled by using individual cameras onboard a small aerial
platform, such as a miniature unmanned aerial vehicle (mini-UAV). In such a case, the cameras may have shifting and
rotational misalignment, even after careful adjustment.
Contiguous frames are captured as the platform flies. So
multi-band registration within a single frame and frame-toframe
mosaicking are necessary to obtain a co-registered
multispectral image for the entire monitoring area before any
commercial product can be generated to support practical
decision-making. In this paper, we present automatic algorithms
to achieve this goal. These algorithms are particularly
useful to the image scenes where no distinctive features are
available. Both automatic and manual evaluations confirm the
effectiveness of the developed algorithms in multi-sensor data
fusion for overall flat terrain without distinctive features.
183 Pixel Level Fusion of Panchromatic and Multispectral
Images Based on Correspondence Analysis
Halil I. Cakir and Siamak Khorram
Abstract Download
Full Article
A pixel level data fusion approach based on correspondence
analysis (CA) is introduced for high spatial and spectral
resolution satellite data. Principal component analysis (PCA)
is a well-known multivariate data analysis and fusion
technique in the remote sensing community. Related to PCA
but a more recent multivariate technique, correspondence
analysis, is applied to fuse panchromatic data with multispectral
data in order to improve the quality of the final
fused image. In the CA-based fusion approach, fusion takes
place in the last component as opposed to the first component
of the PCA-based approach. This new approach is then
quantitatively compared to the PCA fusion approach using
Landsat ETM+, QuickBird, and two Ikonos (with and without
dynamic range adjustment) test imagery. The new approach
provided an excellent spectral accuracy when synthesizing
images from multispectral and high spatial resolution
panchromatic imagery.
193 Multispectral and Panchromatic Data Fusion Assessment Without ReferenceLuciano Alparone, Bruno Aiazzi, Stefano Baronti, Andrea Garzelli, Filippo Nencini, and Massimo Selva
Abstract Download
Full Article
This paper introduces a novel approach for evaluating the
quality of pansharpened multispectral (MS) imagery without
resorting to reference originals. Hence, evaluations are
feasible at the highest spatial resolution of the panchromatic
(PAN) sensor. Wang and Bovik’s image quality index (QI)
provides a statistical similarity measurement between two
monochrome images. The QI values between any couple of
MS bands are calculated before and after fusion and used to
define a measurement of spectral distortion. Analogously,
QI values between each MS band and the PAN image are
calculated before and after fusion to yield a measurement
of spatial distortion. The rationale is that such QI values
should be unchanged after fusion, i.e., when the spectral
information is translated from the coarse scale of the MS
data to the fine scale of the PAN image. Experimental results,
carried out on very high-resolution Ikonos data and simulated
Pléiades data, demonstrate that the results provided
by the proposed approach are consistent and in trend with
analysis performed on spatially degraded data. However, the
proposed method requires no reference originals and is
therefore usable in all practical cases.
201 Anomaly Detection in Hyperspectral Imagery by Fuzzy
Integral Fusion of Band-subsets
Wei Di, Quan Pan, Lin He, and Yongmei Cheng
Abstract Download
Full Article
Anomaly detection in hyperspectral imagery is gaining
increasing interest. However, most covariance matrix-based
detectors are applied directly to all the hyperspectral data
bands without considering the spectral variation and their
possible different contribution to detection. Besides, the
limited sample number as compared with the high dimensionality
of the data can lead to the imprecise estimation of
the covariance matrix and even the singularity problem. In
this paper, a band-subset fuzzy integral fusion (BS-FI) detection
method is presented to solve these problems. The
complete set of hyperspectral data bands is first partitioned
into several lower dimensional band-subsets, whose detection
results are obtained separately and finally merged by a fuzzy
integral fusion method. We adopt a non-parametric fuzzy
support function which can utilize more statistical information
and avoid the model discrepancy that might be brought
in by a fixed distribution model. In addition, the fuzzy
density is assigned by the ratio between the target signal
and noise, which is the key to the target detection problem
through an adaptive eigenvalue-based approach. In the
experiments on real OMIS-I hyperspectral imagery, the
proposed method outperforms the RX detector both on the
complete set of bands and on each band-subset, and other
band-subset fusion detectors.
215 Fusion of Lidar and Imagery for Reliable Building
Extraction
Dong Hyuk Lee, Kyoung Mu Lee, and Sang Uk Lee
Abstract Download
Full Article
We propose a new building detection and description algorithm
for lidar data and photogrammetric imagery using
directional histograms, splitting and merging segments, and
line segments matching. Our algorithm consists of three steps.
In the first step, we extract initial building regions from lidar
data. Here, we apply a modified local maxima technique
coupled with directional histograms and the entropies of
these histograms. In the second step, given the color segmentation
results from the photogrammetric imagery, we extract
coarse building boundaries based on the lidar results with
region segmentation and merging from aerial imagery. In the
third step, we extract precise building boundaries based on
the coarse building boundaries using line segments matching
and perceptual grouping. Experimental results on multisensor
data demonstrate that the proposed algorithm produces
accurate and reliable results.
227 Extracting Urban Road Networks from High-resolution
True Orthoimage and Lidar
Junhee Youn, James S. Bethel, Edward M. Mikhail, and
Changno Lee
Abstract Download
Full Article
Automated or semi-automated feature extraction from remotely
collected, large scale image data has been a challenging issue
in digital photogrammetry for many years. In the feature
extraction field, fusing different types of data to provide
complementary information about the objects is becoming
increasingly important. In this paper, we present a newly
developed approach for the automatic extraction of urban area
road networks from a true orthoimage and lidar assuming
the road network to be a semi-grid pattern. The proposed
approach starts from the subdivision of a study area into small
regions based on homogeneity of the dominant road directions
from the true orthoimage. Each region’s road candidates are
selected with a proposed free passage measure. This process is
called the “acupuncture” method. Features around the road
candidates are used as key factors for an advanced “acupuncture
method” called the region-based acupuncture method.
Extracted road candidates are edited to avoid collocation with
non-road features such as buildings and grass fields. In order
to produce a building map for the prior step, a first-last return
analysis and morphological filter are used with the lidar point
cloud. A grass area thematic map is generated by supervised
classification techniques from a synthetic image, which
contains the three color bands from the true orthoimage and
the lidar intensity value. Those non-road feature maps are
used as a blocking mask for the roads. The accuracy of the
result is evaluated quantitatively with respect to manually
compiled road vectors, and a completeness of 80 percent and
a correctness of 79 percent are obtained with the proposed
algorithm on an area of 1,081,600 square meters.
239 Multisource Classification Using Support Vector
Machines: An Empirical Comparison with Decision Tree
and Neural Network Classifiers
Pakorn Watanachaturaporn, Manoj K. Arora,
and Pramod K. Varshney
Abstract Download
Full Article
Remote sensing image classification has proven to be attractive
for extracting useful thematic information such as landcover.
However, often for a given application, spectral
information acquired by a remote sensing sensor may not be
sufficient to derive accurate information. Incorporation of
data from other sources such as a digital elevation model
(DEM), and geophysical and geological data may assist in
achieving more accurate land-cover classification from
remote sensing images. Recently, support vector machines
(SVM) have been proposed as an alternative for classification
of remote sensing data, and the results are promising. In
this paper, we employ the SVM algorithm to perform multisource
classification. An IRS–1C LISS III image along with
normalized differenced vegetation index (NDVI) image and
DEM are used to produce a land-cover classification for a
region in the Himalayas. The accuracy of SVM-based multisource
classification is compared with several other nonparametric
algorithms namely a decision tree classifier, and
back propagation and radial basis function neural network
classifiers. The well-known kappa coefficient of agreement
is used to assess classification accuracy. The differences in
the kappa coefficient of classifiers have been statistically
evaluated using a pairwise Z-test. The results show a significant
increase in the accuracy of the SVM based classifier
on incorporation of ancillary data over classification
performed solely on the basis of spectral data from remote
sensing sensors.
247 Mapping Vegetation Communities Using Statistical
Data Fusion in the Ozark National Scenic Riverways,
Missouri, USA
Robert A. Chastain Jr., Matthew A. Struckhoff, Hong He,
and David R. Larsen
Abstract Download
Full Article
A vegetation community map was produced for the Ozark
National Scenic Riverways consistent with the association
level of the National Vegetation Classification System.
Vegetation communities were differentiated using a large
array of variables derived from remote sensing and topographic
data, which were fused into independent mathematical
functions using a discriminant analysis classification
approach. Remote sensing data provided variables that
discriminated vegetation communities based on differences
in color, spectral reflectance, greenness, brightness, and
texture. Topographic data facilitated differentiation of
vegetation communities based on indirect gradients (e.g.,
landform position, slope, aspect), which relate to variations
in resource and disturbance gradients. Variables derived
from these data sources represent both actual and potential
vegetation community patterns on the landscape. A hybrid
combination of statistical and photointerpretation methods
was used to obtain an overall accuracy of 63 percent for a
map with 49 vegetation community and land-cover classes,
and 78 percent for a 33-class map of the study area.