PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
June 2015
449
Foreward for the GEOBIA Special Issue
Ioannis Gitas and Giorgos Mallinis, Guest Editors
During the last decades the remote sensing community has wit-
nessed major advances in almost every aspect related to this
science, including, sensor capabilities, algorithms and methods,
information and communication technology, as well as policies
and regulations related to the generation and distribution of re-
mote sensing products. Undoubtedly, a major step forward was
the launch of the Ikonos satellite, providing unprecedented ob-
ject resolving capability from an altitude of 681 kilometers.
This major event opened a new era in image analysis. The
improved sensor characteristics fostered the introduction of
spatial and geographical concepts into the classification pro-
cess in order to formulate meaningful image objects (Blaschke
and Strobl, 2001). The dominant-till-then pixel-centric ap-
proach was also criticized by Hay
et al.
(2001) who suggested a
multiscale framework for analysis of high spatial resolution re-
mote sensing data by incorporating an object-specific approach
in order to overcome the ‘H-Resolution’ and MAUP problems.
From this point onward, a significant part of the remote
sensing community gradually focused on image-objects instead
of pixels for extracting and delivering information related to:
land use and land cover (Myint
et al
., 2011; Salehi
et al.
, 2013),
forests (Chubey
et al.
, 2006; Mallinis
et al.
, 2008), rangelands
(Laliberte
et al.
, 2007), forest fires (Gitas
et al.
, 2004; Mitri and
Gitas, 2004), natural disaster management (Gitas
et al.
, 2008;
Mallinis
et al.
, 2013; Stumpf and Kerle, 2011; Tiede
et al.
,
2011), invasive species detection (Xie
et al
., 2008), and resi-
dential waste heat monitoring (Hay
et al
., 2011) among others.
This shift in image analysis, termed ‘Object-Based Image
Analysis (OBIA)’ or as proposed later by Hay and Castilla (2008)
‘Geographic Object-Based Image Analysis (GEOBIA)’ success-
fully managed to bridge the spatial concepts applied in multi-
scale landscape analysis, Geographic Information Science and
the synergy between image-objects and their radiometric char-
acteristics and analyses in EO data (Blaschke, 2010). The rate
of the shift was affected from technology developments such
as hardware and software evolution, facilitating the resource
demanding GEOBIA processes.
Today, after years of sharp evolution, Geographic Ob-
ject-Based Image Analysis is considered as a new and evolv-
ing paradigm (Blaschke
et al
., 2014). Yet, the most important
achievement is that within the last two decades and through
various conceptual developments, GEOBIA has gained wide
popularity and attracted the interest of both the scientific
and professional communities for its efficiency to provide en-
hanced and reliable geospatial intelligence.
Given its growing success and the needs of the remote sens-
ing community GEOBIA faces new challenges in order to be-
come a dominant paradigm in image analysis. GEOBIA will
have to serve the trend of processing ever more accurate mea-
surements of the Earth’s surface, in space and time. The spatial,
spectral, radiometric, and temporal characteristics of the current
and forthcoming sensors have further raised the expectations
and demands of users for high level, accurate and timely infor-
mation related to the state and the processes of the environment.
The relentless advance of computer and processing technology
that provides an ever increasing ability to acquire, store, access,
and process large image datasets, will also impose the need to
process massive archives with incredible information potential,
available in various cloud-based data repositories.
The increased availability and the wide range of Earth Ob-
servation and geospatial data, as well as the establishment of
global initiatives such as the Global Earth Observation System
of Systems (GEOSS), is expected to further stimulate evolve-
ment and spread of the GEOBIA paradigm not only among the
Remote Sensing and GIScience communities but to a broad ar-
ray of earth-related disciplines.
Among the drivers of evolution and establishment of the
GEOBIA paradigm, the bi-annual international conferences de-
voted exclusively to Geographic Object-Based Image Analysis,
have played and continue to play an outstanding role. The lat-
est forum of this kind, GEOBIA 2014, was held in Thessaloniki,
Greece from 21 to 24 May 2014. The conference that was hosted
by the Laboratory of Forest Management and Remote Sensing,
Aristotle University of Thessaloniki (AUTH), was organized in
collaboration with the Interbalkan Environment Center (i-BEC)
and the Democritus University of Thrace (DUTH).
GEOBIA 2014 was the fifth conference in the series fol-
lowing the successful OBIA International conference held in
Salzburg, Austria (2006), GEOBIA 2008: Pixels, Objects, Intel-
ligence: Geographic Object-based Image Analysis for the 21
st
Century, hosted in Calgary, Canada (2008), GEOBIA 2010, held
in Ghent, Belgium (2010), and GEOBIA 2012, hosted in Rio de
Janeiro, Brazil (2012) which have all provided the impetus for
much progress in Geographic Object-Based Image Analysis.
GEOBIA 2014 eventually gathered over 230 contributors
from 41 countries, five keynote addresses, two industry led
workshops, one special session on ontologies (co-organized
with IRD-ESPACE-DEV), and a roundtable with renowned ex-
perts representing international associations, academic, indus-
try, public sector, and governmental organizations. Both oral
and poster contributions presented the most recent evolutions
on conceptual and methodological aspects of geographic ob-
ject-based image analysis as well as innovative applications
over a broad spectrum of domains. GEOBIA 2014 verified the
progress and the growing interest in geographic object-based
image analysis, as it is also demonstrated by the increasing
number of high-quality peer-review manuscripts appearing in
the literature, and the recent development of GEOBIA-related
(open source/commercial) software packages.
It should be noted that extended abstracts of the works pre-
sented at GEOBIA 2014 were included in a special issue of
the South-Eastern European Journal of Earth Observation and
Geomatics
while the short versions of the abstracts were included in the
conference abstract book.
As a continuum to GEOBIA 2014, all conference partic-
ipants were invited to submit a paper to this special issue.
As a result, twenty-seven very interesting manuscripts from
authors located in all five continents were submitted. Seven-
ty-six colleagues, to whom we are indebted, assisted with the
double-blind review process which resulted in the selection of
the seven articles to be included in this publication. These are
briefly discussed below:
In the field of quantitative segmentation evaluation, Costa
et
Photogrammetric Engineering & Remote Sensing
Vol. 81, No. 6, June 2015, pp. 449–450.
0099-1112/15/449–450
© 2015 American Society for Photogrammetry
and Remote Sensing
doi: 10.14358/PERS.81.6.449
