Peer-Reviewed Articles
581 True Orthophoto Generation of Built-Up Areas Using Multi-View
Images
Jiann-Yeou Rau, Nai-Yu Chen, and Liang-Chien Chen
Abstract Download
Full Article
Hidden areas and shadow effects are major causes of information loss in large-scale
aerial photos of built-up areas. Both types of loss severely degrade the
interpretability of orthophotos. An abrupt change of surface height is
the primary cause of
these defects. Thus, surface discontinuities, the orientation parameters
of the sensors, and solar orientation are all key factors in determining
the extent
of defects. We thus propose an ortho-rectification scheme, which will compensate
for hidden areas and shadow effects in built-up areas, by using multi-view
images. The proposed scheme utilizes projection geometry to detect hidden
and shadowed areas. For hidden areas, lost information is recovered from
corresponding images. A seamless mosaic technique, utilizing
gray-value balance, is suggested to reduce gray value discontinuity. For
shadowed areas, dimmed features are enhanced using the local
histogram matching method to improve image interpretability. Experimental
results indicate that the proposed scheme can significantly reduce hidden
and shadow
defects. Both radiometric and geometric aspects of the proposed product and
process are investigated.
589 Large-Area Land-Cover Mapping through Scene-Based Classification
Compositing
B. Guindon and C.M. Edmonds
Abstract Download
Full Article
Over the past decade, a number of initiatives have been undertaken to create
definitive national and global data sets consisting of precision corrected
Landsat Multispectral Scanner (MSS) and Thematic Mapper (TM) scenes. One
important application of these data is the derivation of large area landcover
products
spanning multiple satellite scenes. A popular approach to land-cover mapping
on this scale involves merging constituent scenes into image mosaics prior
to image clustering and cluster labeling, thereby eliminating redundant geographic
coverage arising from overlapping imaging swaths
of adjacent orbital tracks. In this paper, arguments are presented to support
the view that areas of overlapping coverage contain important information
that can be used to assess and improve classification performance. A methodology
is presented for the creation of large area land-cover products
through the compositing of independently classified scenes. Statistical analyses
of classification consistency between scenes in overlapping regions are employed
both to identify mislabeled clusters and to provide a measure of classification
confidence for each scene at the cluster level. During classification compositing,
confidence measures are used to rationalize conflicting classifications in
overlap regions and to create a relative confidence layer, sampled at the
pixel level, which characterizes the spatial variation in classification
quality
over the final product. The procedure is illustrated with results from a
synoptic mapping project of the Great Lakes watershed that involved the
classification
and compositing of 46 Landsat MSS scenes.
597 Textural and Contextual Land-Cover Classification Using
Single and Multiple Classifier Systems
Olivier Debeir, Isabelle Van den Steen, Patrice Latinne, Philippe Van
Ham, and Eléonore Wolff
Abstract Download
Full Article
The objective of this study was to improve the quality of the digital land-cover
and land-use classification when using high-resolution (10 to 30 m) remote
sensing data. Three classification techniques were compared, which can be
divided into two groups: single classifiers (a five-nearest neighbor and
the C4.5 decision
tree classifier) and multiple classifier systems (BAGFS). Textural and contextual
features (roads, hydrology, relief, etc.) were introduced during the classification
process. Eleven land-cover categories, in a Belgian varied landscape, were
analyzed and classified using Landsat Thematic Mapper data. The accuracy
assessment increased with the introduction of textural features and contextual
data, between
0.60 and 0.82 for the Kappa
coefficient. The best kappa value was achieved using numerous textural and
contextual features with the multiple classifier system (BAGFS).
607 Evaluation of Narrowband and Broadband Vegetation Indices
for Determining Optimal Hyperspectral Wavebands for Agricultural Crop Characterization
Prasad S. Thenkabail, Ronald B. Smith, and Eddy De Pauw
Abstract Download
Full Article
The main goal of the study was to determine optimal waveband centers and
widths required to best estimate agricultural crop characteristics. The
hyperspectral
narrowband data was acquired over 395 to 1010 nanometers using a 1.43-nanometer-wide,
430 bands, hand-held spectroradiometer. Broadband data were derived using
a Landsat-5 Thematic Mapper image acquired to correspond with field spectroradiometer
and ground-truth measurements. Spectral and biophysical data were obtained
from 196 sample locations, including farms and rangelands. Six representative
crops grown during the main cropping season were selected: barley, wheat,
lentil,
cumin, chickpea, and vetch. Biophysical variables consisted of leaf area
index, wet biomass, dry biomass, plant height, plant nitrogen, and canopy
cover.
Narrowband and broadband vegetation indices were computed and their relationship with quantitative crop characteristics were established and compared. The simple narrow-band two-band vegetation indices (TBVI) and the optimum multiple-band vegetation indices (OMBVI) models provided the best results. The narrowband TBVI and OMBVI models are compared with six other categories of narrow and broadband indices. Compared to the best broadband TM indices, TBVI explained up to 24 percent greater variability and OMBVI explained up to 27 percent greater variability in estimating different crop variables. A Predominant proportion of crop characteristics are best estimated using data from four narrowbands, in order of importance, centered around 675 nanometers (red absorption maxima), 905 nm (near-infrared reflection peak), 720 nm (mid portion of the red-edge), and 550 nm (green reflectance maxima). The study determined 12 spectral bands and their bandwidths (Table 5) that provide optimal agricultural crop characteristics in the visible and near-infrared portion of the spectrum.
623 Predicting Mammal Species Richness and Abundance Using
Multi-Temporal NDVI
Boniface O. Oindo
Abstract Download
Full Article
There is need to map indicators of biodiversity such as species richness
and abundance of individuals in order to predict where species loss is
occurring.
Species richness and abundance have been hypothesized to increase with ecosystem
productivity. Moreover, productivity of ecosystems varies in space and time,
and this heterogeneity is also hypothesized to influence species richness
and abundance of individuals. Ecosystem productivity may be estimated using
remotely
sensed data, and researchers have specifically proposed the Advanced Very
High Resolution Radiometer-Normalized Difference Vegetation Index (AVHRR-NDVI).
Interannual average NDVI and its variability (standard deviation) were correlated
with large mammal species richness and abundance of individuals at a landscape
scale in Kenya. The biodiversity indicators associated negatively with interannual
average NDVI
and positively with variability of NDVI. Understanding these relationships
can help in estimating changes in mammalian species richness and abundance
in response to global climate change.
| Top | Home |