IEEE Geoscience and Remote Sensing Magazine - December 2020 - 92

of three Japanese katakana letters in a circle representing
the word Toyota. As a result of this competition, the company changed its name from Toyoda to Toyota. Another example is the international competition announced in 1955
by Joseph Cahill, prime minister of New South Wales, for
designing an opera house in Sydney Harbour. Submissions
came in from 233 people from 32 countries. The winner
was the design of Danish architect Jørn Utzon [20]. Architectural competitions for the design of original buildings
continue to be a popular model even today.
In this study, we used the SLR method to investigate the
use of crowdsourcing in remote sensing research. For this
purpose, four research questions were created, and, in the
light of these questions, publications involving the concepts of crowdsourcing and remote sensing were searched.
Afterward, the initial results were examined in detail, and
91 publications using the integration of crowdsourcing and
remote sensing were considered as finalists. These publications are summarized in Table 1.
CROWDSOURCING AND REMOTE SENSING
INTEGRATION EXAMPLES
With crowdsourcing, very-large-scale projects are broken down into small pieces, and the crowds are asked to
do these small, very simple tasks on their own. In this
way, quite large jobs are completed in a very short time.
Since the crowdsourcing approach is used by many different disciplines, various terminologies have emerged
to express the same concept [21]. Some of them focus on
the spatial nature of the data (e.g., volunteered geographic
information [22] or neogeography [23]), while others have
broader applicability (e.g., crowdsourcing [18], citizen science [24], and user-generated content [25]). Even though the
activities related to these terms are different, the terms
themselves are often used interchangeably for the same
basic endeavors related to geographic information science
with the participation of citizens [26].
In remote sensing projects, crowdsourcing is most useful during data generation and interpretation. One of the
first examples of the use of crowdsourcing to solve a problem related to geospatial data acquisition in remote sensing
is the Did You Feel It? project [27] initiated by the USGS in
the early 1990s. This project can be considered the first significant example of crowdsourcing applied to natural disaster management [7]. The purpose of this project is to gather
information from people in different locations about the
site and significance (intensity) of an earthquake and then
to estimate the magnitude of the damage using this information. Figure 2 shows the spatial distribution of the significance of earthquakes as a result of surveys conducted
between 1991 and 2012.
Another example is the Clickworkers project launched
by NASA in 2000 for trial purposes. In this project, which
was proposed to identify and classify craters on Mars, thousands of participants analyzed each image in the database.
Volunteers completed this task with remarkable accuracy in
92

one month. A single professional geologist doing the same
work would have taken two years [19]. The success of the
trial version led NASA to relaunch the Clickworkers project
in 2006 as a real project.
The Global Earthquake Model (GEM) Project, which
uses the concept of crowdsourcing, is an international forum where organizations and individuals come together to
develop, use, and share tools and resources for an unbiased
assessment of earthquake risk. One of the most important
products of the GEM Project is the OpenQuake Module, an
open source software package developed by the GEM community that enables the use of data for different applications. The GEM project uses crowdsourcing in two different
phases:
◗◗ the identification of risk assessment practices
◗◗ the collection of data for these applications.
Crowdsourcing for identifying risk assessment practices
includes the work of experts and professionals from around
the world on the latest global earthquake risk assessment
technology through GEM Nexus. Crowdsourcing related to
the collection of data for these applications involves the Inventory Data Capture Toolkit, which enables participants to
process satellite images and generate data for the system [7].
Another crowdsourcing example is the Geo-Wiki Project,
which has been ongoing since 2009 [28]. There are major
incompatibilities in certain locations on the Moderate Resolution Imaging Stectroradiometer (MODIS), GlobCover,
and GLC-2000 land cover maps. The Geo-Wiki project is
an online volunteer network created to eliminate these conflicts and increase the accuracy of global land cover maps.
This crowdsourcing activity allows users from around the
world to be part of the process of updating land cover maps
on a global scale. Volunteers can display cultivated lands
and forest conflict maps derived from three land cover data
sets: GLC-2000, MODIS, and GlobCover. They also can select and visualize images with the help of Google Earth's
existing high-resolution image archive. In addition, they
had the ability to upload or view geo-tagged photos on
the Panoramio.com website. Regrettably, this service is no
longer available. A similar service is still available on the
Confluence.org website. In this way, volunteers determine
the type of land cover in the region they are examining and
decide which data set is correct, incorrect, or up to date. Results are saved in a spatial database [29]. Figure 3 shows the
agricultural lands created on the Geo-Wiki platform on a
global scale.
Institutions are also interested in crowdsourcing applications. The European Cooperation in Science and Technology (COST) actions, a project of the European Union,
is an example of such an application. Remote sensing provides images uniquely for mapping, but it is difficult to obtain the ground-truth data needed for map generation and
evaluation. COST Action TD1202, Mapping and the Citizen Sensor, was approved by the European Science Foundation. In the 2012-2015 business plan, the International
World Observation Group coordinated efforts to establish
IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE

DECEMBER 2020

http://www.Panoramio.com http://www.Confluence.org

IEEE Geoscience and Remote Sensing Magazine - December 2020

Table of Contents for the Digital Edition of IEEE Geoscience and Remote Sensing Magazine - December 2020