The 15 papers that constitute the present issue aim to show the importance of recent developments in remote sensing and image processing in Geographical Information Science (GISci). In the past 10 years, new sensors have appeared, based on lasers or radar, or using interference principles, and the resolution of the more conventional optical devices has dropped from 10 m to 0.70 m, while the number of accessible bands, in the visible and the infrared range, has multiplied by 10. The availability of spatial data, for natural, anthropogenic and socioeconomic studies, from such a wide range of sources and a variety of formats opens new horizons to the GISci community. For example, urban areas, which the previous satellites used to resolve rather poorly, have become richer from year to year in significant details in shapes and contours. However, such new complexity leads to new problems. In relation to spatial information, schematically there are four aspects, which are also four challenges that GISci scientists face. They have to retrieve this information, which is assumed to segment the space in homogeneous zones according to some criteria. This often implies filtering steps. They must analyse the selected regions, and associate with them certain numbers and numerical functions, such as size distributions. They have to apply the above geometrical descriptors in some specific context, such as ‘what is the best place to locate a hospital, or to trace a road?’ And, sometimes, they have to conceive random or deterministic models for synthesizing the results of the analysis phase, in order to make forecasts. The authors provide short introductions to the techniques they use, but more extensive presentations may be helpful. Indeed, image processing, as well as GISci, has evolved considerably in the past two decades, and some difficult segmentations require up-to-date versions of wavelets or watersheds, for instance. The studies in this issue borrow their methodology from various sources, including wavelets, random sets (Matheron 1975), geostatistics, Radon transformation, and fuzzy geometry (Zadeh 1965). This list could have contained fractal geometry (Mandelbrot 1982) or rough set theory (Pawlak 1982) as well. Among the methods, mathematical morphology (Matheron 1975, Serra 1982) deserves a special mention because it stems from set descriptors, which have been extended to functions and partitions. Its origin makes it particularly convenient for handling high-resolution images, and the method is involved in the majority of the papers in this issue. In fact, many algebraic operations on maps (Tomlin 1983) involved in GISci-related analyses can be performed through mathematical morphology (e.g. Pullar 2001, Stell 2007). The reader will find a clear presentation of this theory in the book written by Soille (1999), a GISci scientist. We also recommend a recent text by Najman and Talbot (2010) that covers more topics, including connections, connective segmentation, random models and simulations. In March 2009, the annual conference on Spatial Information Retrieval, Analysis, Reasoning and Modelling (SIRARM) was organized at the Bangalore Centre of the Indian Statistical Institute, focusing on the themes that this Special Issue covers. The aim of SIRARM is to bring together remote sensing specialists, GISci experts, and
[1]
K. Parvathi,et al.
Feature extraction from satellite images of hilly terrains using wavelets and watersheds
,
2010
.
[2]
Nicolas Passat,et al.
Multi-resolution region-based clustering for urban analysis
,
2010
.
[3]
Erika Upegui,et al.
Retrieving urban areas on Google Earth images: application to towns of West Africa
,
2010
.
[4]
Pierre Soille,et al.
Constrained connectivity for the processing of very-high-resolution satellite images
,
2010
.
[5]
C. Jeganathan,et al.
Mapping the phenology of natural vegetation in India using a remote sensing-derived chlorophyll index
,
2010
.
[6]
Laurent Najman,et al.
A complete processing chain for ship detection using optical satellite imagery
,
2010
.
[7]
Marina L. Gavrilova,et al.
A divide-and-conquer approach to contour extraction and invariant feature analysis
,
2010
.
[8]
M. Mahmud,et al.
Prediction and simulation of Malaysian forest fires by random spread
,
2010
.
[9]
Pierre Soille,et al.
Morphological Image Analysis
,
1999
.
[10]
Paolo Gamba,et al.
Geographic information system (GIS)-aided per-segment scene analysis of multitemporal spaceborne synthetic aperture radar (SAR) series with application to urban areas
,
2010
.
[11]
J. Brickmann.
B. Mandelbrot: The Fractal Geometry of Nature, Freeman and Co., San Francisco 1982. 460 Seiten, Preis: £ 22,75.
,
1985
.
[12]
John G. Stell,et al.
Relations in Mathematical Morphology with Applications to Graphs and Rough Sets
,
2007,
COSIT.
[13]
Jean Serra,et al.
Image Analysis and Mathematical Morphology
,
1983
.
[14]
Anna E. Klene,et al.
Map Algebra
,
2009,
Encyclopedia of Database Systems.
[15]
Alan Wilson.
Remote sensing as the ‘X-ray crystallography’ for urban ‘DNA’
,
2010
.
[16]
Lorenzo Bruzzone,et al.
Extended profiles with morphological attribute filters for the analysis of hyperspectral data
,
2010
.
[17]
G. Matheron.
Random Sets and Integral Geometry
,
1976
.
[18]
Sanghamitra Bandyopadhyay,et al.
Use of different forms of symmetry and multi-objective optimization for automatic pixel classification in remote-sensing satellite imagery
,
2010
.
[19]
Lotfi A. Zadeh,et al.
Fuzzy Sets
,
1996,
Inf. Control..
[20]
Stéphane Couturier,et al.
A fuzzy-based method for the regional validation of global maps: the case of MODIS-derived phenological classes in a mega-diverse zone
,
2010
.
[21]
Pierre Soille,et al.
Morphological Image Analysis: Principles and Applications
,
2003
.
[22]
J. Chanussot,et al.
On the influence of feature reduction for the classification of hyperspectral images based on the extended morphological profile
,
2010
.
[23]
Benoit B. Mandelbrot,et al.
Fractal Geometry of Nature
,
1984
.