IEEE Geoscience and Remote Sensing Magazine - December 2020 - 89

Crowdsourcing
in Remote
Sensing
A review of
applications and
future directions
EKREM SARALIOGLU AND OGUZ GUNGOR

©SHUTTERSTOCK/LIGHTSPRING

he data obtained with remote sensing sensors are processed to fuel studies related to the Earth's resources
and environment. Sensor technologies being developed are
increasing the diversity and capabilities of remote sensing
data. However, these advancements are also multiplying
the volume of information enormously, straining resources
for processing the data. Meanwhile, the analysis and interpretation of most satellite data have not yet been fully automated, and human intervention is still necessary at certain
stages. In addition, recent deep-learning methods that offer
solutions to many remote sensing problems require a large
amount of labeled data.
Recent research shows that crowdsourcing can contribute to solving such problems. The use of crowdsourcing
allows easy solutions to difficult-to-solve geospatial problems and also reduces the time, cost, and effort involved in
various applications. For this study, we conducted a systematic literature review (SLR) and examined the use of crowdsourcing in remote sensing science. In addition, we present
a comprehensive online table summarizing crowdsourcing
studies in remote sensing for the benefit of readers and future studies. The study shows that the use of crowdsourcing
in remote sensing is relatively new. The study also found
that remote sensing-related crowdsourcing articles began to

Digital Object Identifier 10.1109/MGRS.2020.2975132
Date of current version: 11 March 2020

DECEMBER 2020

IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE

be published in 2016 and the number of such articles
is increasing.
OVERVIEW
The modern discipline of remote sensing arose with the
invention of the first aircraft by the Wright brothers. In
the period between World Wars I and II, aerial photography was intensively carried out for military purposes, and
the images were later used in other civilian applications. A
good example of this is Fairchild's mosaic of Manhattan Island, created in 1921, with a series of 100 overlapping aerial
photographs taken at an altitude of 10,000 ft [1].
On 23 July 1972, the launch of the first Earth-observing satellite, the Earth Resources Technology Satellite (later
renamed Landsat), marked the beginning of a new era in
the remote sensing of Earth from space. For the first time,
satellites were able to capture four-band (a green, red, and
two infrared bands) multispectral images (MSIs) with a
spatial resolution of about 80 m from space using the multispectral scanners sensor. The Landsat series continues with
the latest launched Landsat-8 and contributes to the nearly
47-year Landsat image archive by capturing MSIs at 30-m
resolution. In 2010, the U.S. Geological Survey (USGS) created
the Landsat Global Archive Consolidation to make Landsat
program data a universally accessible global archive. As of
1 January 2015, there were 5,532,454 images in the USGS
archive and about 2.3 million more images in the process
of being added [2].
0274-6638/20©2020IEEE

IEEE Geoscience and Remote Sensing Magazine - December 2020

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