6.4.3 Space-Borne-Based Data Collection
Space-borne-based data collection is also a remote sensing-based data collection
approach. Unlike airborne-based data collection, which is the mounting of a camera
on the aircraft, the remote sensing instrument of space-borne-based data collection is
onboard the satellite platform. According to the uses of satellites, the collected data
could be divided into meteorological data, oceanographic data and Earth resources
data. According to the sensor on the satellite, these data could be grouped into veryhigh spatial resolution data (e.g., IKONOS, QuickBird, OrbView-3, Cartosat,
WourdView, GeoEye-1), moderate-to-high spatial resolution data (e.g.,
Landsat, Sentinel-2), Moderate Resolution Imaging Spectroradiometer (MODIS)
data, hyperspectral data (e.g., AVIRIS, CASI), and radar data (e.g., JERS, ERS,
Radarsat, EnviSat) (Gao 2008). As the major approach to collect Earth observation
data, space-borne-based remote sensing plays an unreplaceable role in agricultural
image data collection.
6.4.4 Big Data Challenge in Agricultural Image Data
Collection
With the rapid development of remote sensing technology, more and more satellites
characterized by high spatial, temporal, and radiometric resolution are available and
launched. Therefore, how to manage and analyze the volumes of high-resolution
agricultural image data captured by the different types of satellites is becoming a
challenge in the big data era. In 2001, Laney (2001) from META Group described
big data in three Vs: volume as the scale of data, variety as the different forms of
data, and velocity as the analysis of streaming data. These three dimensions also
work for describing agricultural image data.
Volume: Terabytes of high-resolution Earth observation data were generated every
day by satellites which characterized by high spatial, temporal, and radiometric
resolution. As one of the major Earth science data manage systems, NASA’s
Earth Observing System Data and Information System (EOSDIS), a key core
capability in NASA’s Earth Science Data Systems Program, is managing NASA’s
Earth science data from various sources, including satellites, aircraft, and field
measurements (https://earthdata.nasa.gov). All Earth observation data can be
accessed via EOSDIS Distributed Active Archive Centers (DAACs) online
(https://earthdata.nasa.gov/about/daacs). The most recent EOSDIS key science
system metrics (NASA 2017) show that the total archive volume is 17.5
petabytes, and approximately 12.1 terabytes volume of data is generated every
day (average daily archive growth between Oct 1, 2015, and Sept 30, 2016),
which means the Earth observation data is growing so fast and has been doubled
with 7.5 petabytes in 2012 (Ramapriyan et al. 2013).
6 Image Processing Methods in Agricultural Observation Systems
89
Space-borne-based data collection is also a remote sensing-based data collection
approach. Unlike airborne-based data collection, which is the mounting of a camera
on the aircraft, the remote sensing instrument of space-borne-based data collection is
onboard the satellite platform. According to the uses of satellites, the collected data
could be divided into meteorological data, oceanographic data and Earth resources
data. According to the sensor on the satellite, these data could be grouped into veryhigh spatial resolution data (e.g., IKONOS, QuickBird, OrbView-3, Cartosat,
WourdView, GeoEye-1), moderate-to-high spatial resolution data (e.g.,
Landsat, Sentinel-2), Moderate Resolution Imaging Spectroradiometer (MODIS)
data, hyperspectral data (e.g., AVIRIS, CASI), and radar data (e.g., JERS, ERS,
Radarsat, EnviSat) (Gao 2008). As the major approach to collect Earth observation
data, space-borne-based remote sensing plays an unreplaceable role in agricultural
image data collection.
6.4.4 Big Data Challenge in Agricultural Image Data
Collection
With the rapid development of remote sensing technology, more and more satellites
characterized by high spatial, temporal, and radiometric resolution are available and
launched. Therefore, how to manage and analyze the volumes of high-resolution
agricultural image data captured by the different types of satellites is becoming a
challenge in the big data era. In 2001, Laney (2001) from META Group described
big data in three Vs: volume as the scale of data, variety as the different forms of
data, and velocity as the analysis of streaming data. These three dimensions also
work for describing agricultural image data.
Volume: Terabytes of high-resolution Earth observation data were generated every
day by satellites which characterized by high spatial, temporal, and radiometric
resolution. As one of the major Earth science data manage systems, NASA’s
Earth Observing System Data and Information System (EOSDIS), a key core
capability in NASA’s Earth Science Data Systems Program, is managing NASA’s
Earth science data from various sources, including satellites, aircraft, and field
measurements (https://earthdata.nasa.gov). All Earth observation data can be
accessed via EOSDIS Distributed Active Archive Centers (DAACs) online
(https://earthdata.nasa.gov/about/daacs). The most recent EOSDIS key science
system metrics (NASA 2017) show that the total archive volume is 17.5
petabytes, and approximately 12.1 terabytes volume of data is generated every
day (average daily archive growth between Oct 1, 2015, and Sept 30, 2016),
which means the Earth observation data is growing so fast and has been doubled
with 7.5 petabytes in 2012 (Ramapriyan et al. 2013).
6 Image Processing Methods in Agricultural Observation Systems
89