remote sensing

Article
Mass Processing of Sentinel-1 Images for
Maritime Surveillance
Carlos Santamaria 1, *, Marlene Alvarez 1 , Harm Greidanus 1 , Vasileios Syrris 2 , Pierre Soille 2
and Pietro Argentieri 1
1 European Commission, Joint Research Centre (JRC), Directorate for Space, Security & Migration,
Via E. Fermi 2749, I-21027 Ispra (VA), Italy; marlene.alvarez@ec.europa.eu (M.A.);
harm.greidanus@ec.europa.eu (H.G.); pietro.argentieri@ext.ec.europa.eu (P.A.)
2 European Commission, Joint Research Centre (JRC), Directorate for Competences, Via E. Fermi 2749,
I-21027 Ispra (VA), Italy; vasileios.syrris@ec.europa.eu (V.S.); pierre.soille@ec.europa.eu (P.S.)
* Correspondence: carlos.santamaria@ec.europa.eu; Tel.: +39-0332-785985

Received: 24 May 2017; Accepted: 27 June 2017; Published: 2 July 2017

Abstract: The free, full and open data policy of the EU’s Copernicus programme has vastly increased
the amount of remotely sensed data available to both operational and research activities. However,
this huge amount of data calls for new ways of accessing and processing such “big data”. This paper
focuses on the use of Copernicus’s Sentinel-1 radar satellite for maritime surveillance. It presents
a study in which ship positions have been automatically extracted from more than 11,500 Sentinel-1A
images collected over the Mediterranean Sea, and compared with ship position reports from
the Automatic Identification System (AIS). These images account for almost all the Sentinel-1A
acquisitions taken over the area during the two-year period from the start of the operational phase
in October 2014 until September 2016. A number of tools and platforms developed at the European
Commission’s Joint Research Centre (JRC) that have been used in the study are described in the paper.
They are: (1) Search for Unidentified Maritime Objects (SUMO), a tool for ship detection in Synthetic
Aperture Radar (SAR) images; (2) the JRC Earth Observation Data and Processing Platform (JEODPP),
a platform for efficient storage and processing of large amounts of satellite images; and (3) Blue Hub,
a maritime surveillance GIS and data fusion platform. The paper presents the methodology and
results of the study, giving insights into the new maritime surveillance knowledge that can be gained
by analysing such a large dataset, and the lessons learnt in terms of handling and processing the
big dataset.

Keywords: maritime surveillance; ship detection; Sentinel-1; big data; Synthetic Aperture Radar

1. Introduction
Maritime surveillance can be defined as the monitoring of human activities at sea. The surveillance
is intended to support efforts related with security (e.g., irregular sea border crossing, and smuggling of
illegal goods or substances), safety (e.g., Search and Rescue, and shipping traffic), and environmental
and sustainability (e.g., fishing control, and pollution) aspects. This paper focuses on shipping
activities, although some views on non-shipping activities (e.g., fixed oil and gas platforms, aquaculture
equipment, or offshore wind farms) are also gained as a by-product.
Shipping monitoring systems can broadly be classified in two groups: cooperative and
non-cooperative. Cooperative systems rely on the ships reporting information about themselves
(e.g., identification, position, and speed). Commonly used self-reporting systems are the Automatic
Identification System (AIS [1]), Long Range Identification and Tracking (LRIT [2]) and Vessel
Monitoring System (VMS [3,4]). Some of the most widely used non-cooperative systems employ
radar sensors (coastal, shipborne, airborne, and spaceborne) to detect the ships from the background

Remote Sens. 2017, 9, 678; doi:10.3390/rs9070678 www.mdpi.com/journal/remotesensing

Remote Sens. 2017, 9, 678 2 of 20

sea clutter without relying on the ships’ cooperation. This study concerns the use of spaceborne SAR
sensors for ship detection and the fusion of this information with AIS reports.
Ship detection in satellite SAR images is operationally used in various regions, for instance in
Canada [5] and in Europe [6]. Maritime surveillance conducted with such images does not require
the vessels’ cooperation, provides wide-area monitoring even of remote regions, and can be operated
day and night and regardless of cloud cover conditions (although weather phenomena such as strong
wind and strong precipitation will hamper the performance). Most SAR ship detection techniques
rely on the idea that the ships reflect the radar signal differently to how the sea surface does it, thus
affecting some property (intensity and polarimetry) of the signal that is received back by the sensor,
and therefore allowing the discrimination of the ships against the background sea clutter. See [7,8] for
more details on ship detection approaches. Validation of these techniques is not an easy task since it
involves a precise knowledge of the shipping activity in an area, which can be uncontrollable especially
in wide areas, and because the performance of the techniques depends on many different variables
(sea state, wind, ship’s size, material and geometry, radar resolution and incidence angle, and aspect
angle). See [9–15] for some validation efforts.
Sentinel-1 is the SAR satellite constellation of the European Union’s Copernicus programme for
Earth Observation, operated by the European Space Agency (ESA). It consists of two units, Sentinel-1A
launched on 3 April 2014 and Sentinel-1B launched on 25 April 2016, which entered into operation
on 3 October 2014 and 26 September 2016, respectively. It uses a C-band instrument (frequency
5.405 GHz), and can operate in four acquisition modes: Extra Wide (EW), Interferometric Wide (IW),
Stripmap (SM) and Wave (WV). The data can be processed to various types of image product, and the
Ground Range Detected High resolution (GRDH) product [16,17] is the one used in this study. The
radar can transmit in vertical (V) or horizontal (H) polarisation, and can receive in H, V or in both of
them simultaneously. A Sentinel-1 product can therefore have one of the following combinations of
polarisations: single channel HH, VV, HV or VH, or dual HH + HV or VV + VH, the first letter being the
transmitted polarisation and the second one the received polarisation [16]. Following the Copernicus’s
“free, full and open” data access policy, all the images (“full”) are made available to all users (“open”)
at no cost (“free”) through the Copernicus Open Access Hub (formerly known as Sentinels Scientific
Data Hub), while dedicated data access is provided to international partners (International Hub),
collaborative ground segments (Collaborative Hub) and Copernicus Services (Copernicus Services
Hub) [18]. Sentinel-1 operates in a predefined observation plan, which dictates the areas that will be
monitored and the modes and polarisations that will be used [19]. This plan responds to a number
of user requirements and technical constraints [20]. According to the current operation, Sentinel-1
monitors the entire Mediterranean (“Med”) Sea in IW mode and VV + VH polarisation. Each satellite
has an orbit cycle of 12 days; this means that the satellite can monitor the same area on the ground
from the same orbital position every 12 days. Due to overlaps between image swaths in ascending and
descending orbits and between swaths in adjacent orbits, the revisit time of a point on the ground can
be much shorter than 12 days: at the Med latitude, the revisit time can be as short as three days using
the acquisitions of one of the satellites only, and 1.5 days using the acquisitions of both satellites. The
Sentinel-1 satellites fly in a sun-synchronous near-polar orbit. At low and medium latitudes (which
includes the Med Sea) the overhead passes are confined to narrow time windows around 6:00 a.m. and
6:00 p.m. local time.
The objective of the study was to test automatic ship detection processing capabilities in large
volume maritime monitoring campaigns that leave no room for manual verification of the detection
results to assess Sentinel-1’s contribution in a multi-source AIS + SAR monitoring campaign (e.g.,
number of non-reporting ships detected in the Sentinel-1 images), and to quantify how the use of repeat
acquisitions can improve the monitoring results (by analysing the recurrence of targets). It builds on
the experience gained in earlier analyses of Sentinel-1 images [21,22].
Remote Sens. 2017, 9, 678 3 of 20

Remote Sens. 2017, 9, 678 3 of 20

2. Study Definition and Datasets
2. Study Definition and Datasets
2.1. Study Definition
2.1. Study Definition
The area of interest (AoI) of the study is the Mediterranean Sea (including the Sea of Marmara).
The studyTheperiod
area ofisinterest
almost(AoI) of the
two full study
years, is the
from Mediterranean
3 October Sea (including
2014 (start the Sea
of Sentinel-1A of Marmara).
operational phase)
to The study period
30 September 2016. is Figure
almost 1two
mapsfullthe
years,
AoI. from 3 October 2014 (start of Sentinel-1A operational
phase) to 30 September 2016. Figure 1 maps the AoI.

Figure1.1. Study
Figure Study area (Mediterranean
(MediterraneanSea).
Sea).

2.2.2.2. Sentinel-1
Sentinel-1 Data
Data
Duringthe
During thecampaign
campaign period,
period, Sentinel-1
Sentinel-1primarily
primarilyused used thethe
IWIW acquisition
acquisitionmode in the
mode inAoI,
the inAoI,
accordance with its observation and planning strategy [20]. The EW
in accordance with its observation and planning strategy [20]. The EW and SM modes were occasionally and SM modes were
occasionally
used in the firstused in the
months firstthe
after months after
start of thethe start of the
operational operational
phase, but they phase, butlonger
are no they areusedno routinely
longer
used routinely in the AoI. Consequently, and to facilitate the interpretation of the results, only IW
in the AoI. Consequently, and to facilitate the interpretation of the results, only IW acquisitions have
acquisitions have been used in this study. Furthermore, only images that contain at least 10% of their
been used in this study. Furthermore, only images that contain at least 10% of their extent inside the
extent inside the AoI have been analysed. This is to prevent images that will only marginally
AoI have been analysed. This is to prevent images that will only marginally contribute to the results of
contribute to the results of the study from putting an unnecessary burden on the analysis platform.
the study from putting an unnecessary burden on the analysis platform. On the other hand, images
On the other hand, images that stretch outside the Med (into the Atlantic Ocean beyond the Strait of
that stretch outside
Gibraltar, the Med
into the Black (into thethe
Sea beyond Atlantic
Bosphorus Oceanandbeyond
into the the
RedStrait of Gibraltar,
Sea beyond the Suez into the Black
Canal) are
Seakept
beyond the Bosphorus and into the Red Sea
in the analysis and contribute to the final statistics. beyond the Suez Canal) are kept in the analysis and
contribute to the final statistics.
A total of 97 Sentinel-1B images were acquired over the AoI between 26 September 2016 (start of
A total ofoperations)
Sentinel-1B 97 Sentinel-1B and images were acquired
30 September 2016 (end over the study).
of this AoI between
Although26 September
the A and 2016 (start of
B satellites
Sentinel-1B operations)
are identical and all theandcomponents
30 September of2016 (end of thischain
the processing study). Although
have the Abeen
successfully and Btested
satellites
withare
identical and all
Sentinel-1B the components
products, for better of guaranteed
the processing chain have
uniformity successfully
it has been tested
been decided with Sentinel-1B
to exclude those 97
products, for better
images from guaranteed
the study uniformity
since those it has
images will notbeen
have decided
a sizabletoimpact
exclude onthose 97 images
the results frominthe
reported
study
this since
paper.those images will not have a sizable impact on the results reported in this paper.
Table
Table 1 presents
1 presents the the
main main details
details of Sentinel-1
of the the Sentinel-1
images images
usedused instudy.
in this this study. The overall
The overall number
number of Sentinel-1 IW images used in the study is 11,647. All of them are
of Sentinel-1 IW images used in the study is 11,647. All of them are GRDH products, with a resolution GRDH products, with a
resolution of around 20 m. As can be seen in the table, the vast majority (11,421)
of around 20 m. As can be seen in the table, the vast majority (11,421) of them are dual-pol VV + VH. of them are dual-pol
VV +2VH.
Figure plots Figure 2 plots the
the number number of observations
of observations for each geographic
for each geographic location inlocation
the area in (the
the area
maximum(the
maximum number of observations of a single location is 225). The diamond
number of observations of a single location is 225). The diamond patterns are due to the overlap patterns are due to the of
overlap of images acquired in ascending and descending orbits
images acquired in ascending and descending orbits and in neighbouring orbit tracks. and in neighbouring orbit tracks.
A number of these 11,647 images are affected by radiofrequency interference (RFI). In most
A number of these 11,647 images are affected by radiofrequency interference (RFI). In most cases
cases this interference originates from ground sources, and more rarely from the Radarsat-2 SAR
this interference originates from ground sources, and more rarely from the Radarsat-2 SAR satellite,
satellite, which operates in the same frequency band as Sentinel-1 [23,24]. Figure 3 shows typical
which operates in the same frequency band as Sentinel-1 [23,24]. Figure 3 shows typical examples of
examples of how this interference is manifested in Sentinel-1 images. The visible bright stripe
how this interference is manifested in Sentinel-1 images. The visible bright stripe artefacts can lead
artefacts can lead to a large number of false ship detections in very narrow sea areas, which can be
to a large number of false ship detections in very narrow sea areas, which can be misinterpreted as
misinterpreted as unusual shipping activity. To mitigate this problem, some of these images have
unusual shipping activity.
been identified and labelled To mitigate this problem,
as RFI-affected. some of these
No automated images have
identification been identified
algorithm has been and
Remote Sens. 2017, 9, 678 4 of 20

labelled as RFI-affected. No automated identification algorithm has been implemented and, instead,
theRemote
process
Sens.relies
2017, 9,on
678a manual intervention explained next. Due to the very large number of 4images of 20
Remote Sens. 2017, 9, 678 4 of 20
in this study, the identification cannot be done by visual inspection of each image. Instead, once the
shipimplemented and, instead,
detection analysis the applied
has been process relies on
theaaimages,
to all on manual intervention explained next. Due to the
implemented and, instead, the process relies manual the detections
intervention have been
explained next.aggregated
Due to the and
very large number of images in this study, the identification cannot be done by visual inspection of
displayed on a map. Visual inspection of this detection map has shown several
very large number of images in this study, the identification cannot be done by visual inspection of conspicuous long and
each image. Instead, once the ship detection analysis has been applied to all the images, the
narrow areas with
each image. high once
Instead, density theofship
detections.
detection The imageshas
analysis these
beendetections
applied originate
to all thefrom havethe
images, then
detections have been aggregated and displayed on a map. Visual inspection of this detection map
been visuallyhave
detections examined and, if the presence
been aggregated of RFI on
and displayed is confirmed,
a map. Visual the images
inspectionare of
labelled as RFI-affected.
this detection map
has shown several conspicuous long and narrow areas with high density of detections. The images
The has shown
ship severalanalyses
detection conspicuous longimages
of these and narrow areas
are not with high
included density
in the of detections.
results presented The images
in this paper.
these detections originate from have then been visually examined and, if the presence of RFI is
these
It is clear detections
that the originate
images from as
labelled have then beencontain
RFI-affected visuallylarge
examined
RFI-free and, if the
areas with presence
genuine ofships,
RFI is and
confirmed, the images are labelled as RFI-affected. The ship detection analyses of these images are
thatconfirmed,
these the images
genuine are labelled
detections will beasmissed
RFI-affected.
in the The ship detection
results. However, analyses
in this of these
paper it images
is are
preferable
not included in the results presented in this paper. It is clear that the images labelled as RFI-affected
to not
miss included
contain those
large
in the results
targets
RFI-freethan topresented
areas include in thisamounts
large
with genuine
paper. It is clear
ships, and of false
that
that
these
the images
alarms
genuine
labelled as RFI-affected
in spurious
detectionslinear
will befeatures
missed in that
contain
could lead large
to RFI-free areas
misinterpretation with ofgenuine
the ships,
detection and
maps,that these
with genuine
the detections
additional will be that
constraint missedhumanin
the results. However, in this paper it is preferable to miss those targets than to include large amounts
the results.
assistance However,
in the detectionin this paper ittoisbe
process preferable atominimum.
miss thoseA targets than
74 to include large amounts
of false alarms in spurious linear is
featureskeptthat at
could total of
lead to misinterpretation images
of thehave been
detection labelled
maps,
of false alarms in spurious linear features that could lead to misinterpretation of the detection maps,
as with
RFI-affected. These constraint
the additional images arethat included
humaninassistance
the imageindensity map inprocess
the detection Figure is 2, to
butbeare excluded
kept at a
with the additional constraint that human assistance in the detection process is to be kept at a
from the overall
minimum. detection
A total and correlation
of 74 images have been statistical
labelled results (Section 4.3)
as RFI-affected. Theseand from the
images are detection
included in density
the
minimum. A total of 74 images have been labelled as RFI-affected. These images are included in the
mapsimage(Section
density 4.5).
map in Figure 2, but are excluded from the overall detection and correlation statistical
image density map in Figure 2, but are excluded from the overall detection and correlation statistical
results (Section 4.3) and from the detection density maps (Section 4.5).
results (Section 4.3) and from the detection density maps (Section 4.5).

Figure 2. Number of observationsfor

Number for eachgeographic
geographic location used in the study (11,647 images, only
Figure
Figure2. 2. Numberofofobservations
observations for each locationused
each geographic location usedininthe
thestudy
study (11,647
(11,647 images,
images, only
only
Sentinel-1A).
Sentinel-1A). Calculated
Calculated with
with a a grid
grid size
size of
of 0.1°.
0.1 ◦ .
Sentinel-1A). Calculated with a grid size of 0.1°.

(a) (b)
(a) (b)
Figure 3. Example of Sentinel-1 images affected by RFI (radiofrequency interference): (a) from
Figure 3. 3. Example
Example of of Sentinel-1
Sentinel-1 images affected by (radiofrequency
RFI (radiofrequency interference): (a) from
Figure
ground source (image name: images affected by RFI interference):
S1A_IW_GRDH_1SDV_20160712T164623_20160712T164648_012118_ (a) from ground
ground source (image name: S1A_IW_GRDH_1SDV_20160712T164623_20160712T164648_012118_
source (image
012C3E_5B16); name: S1A_IW_GRDH_1SDV_20160712T164623_20160712T164648_012118_
and (b) from Radarsat-2 (S1A_IW_GRDH_1SDV_20150909T165710_ 012C3E_5B16);
012C3E_5B16); and (b) from Radarsat-2 (S1A_IW_GRDH_1SDV_20150909T165710_
and20150909T165735_007641_00A97F_B8EB).
(b) from Radarsat-2 (S1A_IW_GRDH_1SDV_20150909T165710_
The VH polarimetric 20150909T165735_007641_00A97F_B8EB).
channel is shown in both cases.
20150909T165735_007641_00A97F_B8EB). The VH polarimetric channel is shown in both cases.
TheAzimuth
VH polarimetric channel
is from top is shown
to bottom; rangein both
fromcases.
left toAzimuth
right. Theis from top to bottom;
RFI (indicated with range
arrows)from left to right.
is visible as
Azimuth is from top to bottom; range from left to right. The RFI (indicated with arrows) is visible as
Thehorizontal
RFI (indicated
rowswith arrows)
or short is visible
bright asstripe
vertical horizontal rowsSentinel-1
artefacts. or short bright vertical
images stripe artefacts.
© Copernicus 2015, Sentinel-1
2016.
horizontal rows or short bright vertical stripe artefacts. Sentinel-1 images © Copernicus 2015, 2016.
images © Copernicus 2015, 2016.
The area off Israel in the eastern Mediterranean is severely affected by RFI. Almost every image
The area off Israel in the eastern Mediterranean is severely affected by RFI. Almost every image
there is affected by strong RFI. Since identifying and labelling the ones affected by RFI would have
Theisarea
there off Israel
affected in theRFI.
by strong eastern
SinceMediterranean
identifying andislabelling
severelythe
affected by RFI.by
ones affected Almost everyhave
RFI would image
imposed a heavy burden on the amount of manual assistance provided, it was decided for that area
there is affected
imposed by burden
a heavy strong RFI. Since
on the identifying
amount of manualand labellingprovided,
assistance the onesitaffected by RFI
was decided forwould have
that area
not to attempt to label the images but instead delineate the area most affected by the RFI-related
not to attempt to label the images but instead delineate the area most affected by the RFI-related
detections and exclude that area from the overall ship detection and correlation statistics (presented
detections and exclude that area from the overall ship detection and correlation statistics (presented
Remote Sens. 2017, 9, 678 5 of 20

imposed a heavy burden on the amount of manual assistance provided, it was decided for that area
not to attempt to label the images but instead delineate the area most affected by the RFI-related
detections and exclude that area from the overall ship detection and correlation statistics (presented in
Section 4.3). The area is small and the impact of this exclusion on the overall statistics is minor: only
3% of the detections, 0.5% of the interpolated positions, and 1% of the correlations are inside the area
excluded. The area is indicated in Section 4.5.

Table 1. Main details of the satellite SAR images used in this study.

Area of Interest Mediterranean Sea

Study period From 3 October 2014 to 30 September 2016
Sensor Sentinel-1A
Number of images 11,647
Excluding RFI 11,573
Image mode IW
Product type GRDH
Polarisation 11,421 VV + VH
25 HH + HV
199 VV
2 VH
Image swath ~250 km
Resolution ~20 × 22 m [range × azimuth]

While Sentinel 1B data were not considered for ship detection, a comparison between the revisit
times obtained with Sentinel-1A only and with the full Sentinel-1A + Sentinel-1B constellation in the
period from 29 November 2016 to 9 February 2017 during which they were both in full operational
capability was carried out. For this, only the metadata of the Sentinel-1 images were used.

2.3. Ship Position Reports

Terrestrial AIS from the Maritime Safety and Security Information System (MSSIS) network and
from the Italian Coast Guard, and satellite AIS from up to three satellites from the Norwegian Coastal
Administration/Norwegian Defence Research Establishment (FFI) has been received during the study.
Overall, areas around the European countries are well covered by the terrestrial AIS data, while the
coverage around African countries is poor. The satellite AIS data cover the entire Mediterranean Sea;
however, on account of radio interference from land and message collisions due to the high intensity
of the ship traffic, the satellite AIS detection rate is not high over the Mediterranean, especially over
the northern part [25]. In addition, the satellite AIS data were only available for the first 13 months of
the study. Long range AIS messages, designed to better cope with message collisions, were not used.

3. Methods
This section describes a number of tools and platforms that have been developed in JRC through
the years and that have been used in this study. SUMO (ship detection software) and the image
metadata extraction process were run on JEODPP (platform for storage and processing of Earth
Observation data), Blue Hub (maritime surveillance platform) and the recurrent target analysis were
run outside JEODPP.

3.1. Ship Detection: SUMO

SUMO is the JRC’s tool for ship detection in satellite SAR images [26]. It was recently extensively
described in [8], so here only a brief overview is given. SUMO accepts as input an image from any
of several satellite SAR sensors—current (Sentinel-1, Radarsat-2, TerraSAR-X, Cosmo-SkyMed, and
ALOS-2 PALSAR-2) or past (ERS-1, ERS-2, Radarsat-1, and ENVISAT ASAR)—carries out a Constant
False Alarm Rate (CFAR) detection, and outputs the list of detected ships in an XML file. For the
Remote Sens. 2017, 9, 678 6 of 20

detection, SUMO assumes that the sea clutter conforms to a K distribution and locally estimates
the parameters of this distribution in small non-overlapping tiles. Any pixel whose value is above
a threshold derived from the local distribution is then detected. Groups of neighbouring detected
pixels are agglomerated in detected targets or ships. SUMO carries out the pixel-based CFAR detection
on each polarimetric channel independently and then merges the results in a single set of targets. Based
on a target’s attributes, a reliability level is assigned. Detections deemed to be azimuth ambiguities
(based on the deterministic distance between the ambiguity and its originating object and on the
object-ambiguity intensity ratio [27]) are automatically flagged and assigned the lowest reliability
by SUMO. There is no attempt to flag range ambiguities. This is because the identification of range
ambiguities is often a hard task: typical distance between a range ambiguity and its originating target
is 100 km or more, while it is only around 5 km for azimuth ambiguities. This means that the source
of the range ambiguity is, in many cases, outside the image extent, which prevents the flagging of
the ambiguity by simple deterministic distance as in the case of azimuth ambiguities. SUMO can be
operated in semi- or fully automated ways, depending on whether a human operator supervises and
corrects the detections or not, respectively.
In this study, the OpenStreetMap coastline [28] buffered by 250 m has been used to mask the land
areas of the images. SUMO’s operation has been fully automatic, i.e., the detections have not been
verified by a human. Relatively high detection thresholds have been chosen, especially in the co-pol
channels (HH and VV), since lower thresholds would produce a significant number of false alarms that
are not eliminated by a human operator. Specifically, a nominal false alarm rate of 10−7 and detection
threshold adjustments of 10.0 and 2.0 for the co-pol and cross-pol channels, respectively, have been used.
These adjustments regulate the CFAR detection threshold that is finally applied (see [8] for more details).

3.2. Platform for Earth Observation Data Storage and Processing: the JEODPP
With its fleet of Sentinel satellites operated by the European Space Agency, the Copernicus
programme of the European Union is making Earth Observation truly enter the big data era. Indeed,
the Sentinel satellite series in full operational capacity will produce a continuous and voluminous
stream of image data (up to 10 terabytes per day) with a variety of sensors at different spectral, spatial,
and temporal resolutions. The resulting vast amount of free and open data generated by the Sentinel
and other Earth Observation satellites calls for new approaches in data management and processing to
enable the timely extraction of relevant information from these data in combination with data from
other sources. Indeed, a fragmented approach whereby individual projects are managing their data
and processing infrastructure is not sustainable anymore. At the Joint Research Centre, this motivated
the development of a common platform to serve the needs of projects relying on the processing
and analysis of Earth Observation data in support of policy needs. This platform is called the JRC
Earth Observation Data and Processing Platform (JEODPP) [29,30]. It is versatile in the sense that
it accommodates existing scientific workflows such as those based on the SUMO software without
the need for rewriting the underlying code to match specific requirements. The main components of
the JEODPP are briefly detailed hereafter with emphasis on those used for the massive processing of
Sentinel-1 imagery for maritime surveillance.
The JEODPP relies on commodity hardware for both storage and processing components as well
as a series of software layers enabling different access levels to users. It is scalable, but at the time
of this study, the storage component consisted of 1.4 petabyte organised in 10 storage servers each
equipped with 24 disks of 6 terabytes. This type of storage architecture is referred to as Just a Bunch
of Disks (JBODs). The whole disk space is seen as a unique logical volume thanks to the distributed
file system developed by CERN and called EOS [31,32]. To secure high availability and the correction
of errors, all data are stored with a redundancy level equal to 2 leading to a net storage capacity of
0.7 petabyte. The processing component consists of 472 processors originating from 23 servers (16 with
12 cores and 7 with 40 cores).
Remote Sens. 2017, 9, 678 7 of 20

The data required for this study were discovered and downloaded automatically using
OpenSearch and OData scripting capabilities offered by the Copernicus Open Access [33] and
Copernicus Services [34] hubs operated by the European Space Agency. The query for download
consisted of searching for all Sentinel-1A IW-GRDH products with a non-empty intersection with the
Mediterranean Sea during the period 3 October 2014–30 September 2016 as detailed in Section 2.2. The
downloaded files are unzipped during ingestion for subsequent use by SUMO.
The analysis of the revisit times with the joint Sentinel-1A + Sentinel-1B (Section 4.1) was done by
performing a similar query to that described above but retrieving only the necessary metadata files.
To accommodate a wide variety of legacy software written in different programming languages
with often incompatible software library dependencies, the JEODPP implements virtualisation
technologies. The typical overhead encountered with operating system virtualisation is avoided
thanks to operating system level virtualisation (also called lightweight virtualisation) based on Docker
containerization [35]. Each application runs within a Docker container isolating the user-space instance
to prevent conflict with applications running side-by-side in different containers. A Docker container
is launched from a Docker image that contains all the required software and associated libraries. The
SUMO software relies on programs written in Java so it was enough to create a Docker image equipped
with Java run time environment.
Another essential component of the JEODPP is its workload manager that efficiently distributes
the requested computing jobs to the processing servers by taking into account the resources required
by each job, their priority, and the current load of each node. This is achieved with the HTCondor
workload manager given its suitability to manage massive amounts of parallel jobs with little need
for inter-job communication [36]. For instance, the ship detection with SUMO boils down to the
independent detection performed on each input Sentinel-1 image followed by a subsequent analysis
restricted to the detected ships. This can be optimally addressed by switching off any multithread
computation and launching as many Docker containers as the number n of available cores. In this
case, each container processes sequentially a number of Sentinel 1 images equal to the total number of
files to process divided by n. The results of this approach are presented in Section 4.2. Finally, besides
accommodating scientific workflows written in a variety of languages over time (legacy software), the
JEODPP also offers interactive visualisation and analysis capabilities through a web interface. This
was not used in the present study given that the analysis was fully automatic and that modules for
digesting the detection performed at scene level were already developed prior to the platform. This
will be addressed in the future given that it would further ease collaboration and knowledge sharing
with other projects running on the JEODPP.

3.3. Integrated Maritime Surveillance Platform: Blue Hub

The Blue Hub is JRC’s maritime surveillance R&D platform [37]. Among other functionalities,
this platform receives and stores ship reporting data (such as AIS) and ship detection data from
SUMO, and fuses these data. During the fusion process, the reporting data are interpolated or
extrapolated to the SAR image acquisition time. A shift in the image azimuth direction is applied to the
interpolated reported positions based on their reported speed and the imaging geometry, to account
for the speed-induced shift that the ship will experience in the image during the SAR image formation
process [8]. This is necessary for an accurate collocation of the SAR detection and the interpolated
reported position. A SAR detection and an interpolated reported position are deemed to be correlated
if they are geographically closer than 500 m, and may still be correlated if the distance between them is
under 5000 m. The fusion process outputs the list of interpolated reported positions, the list of SAR
detections that are correlated to interpolated reported positions, the list of uncorrelated SAR detections,
and the list of uncorrelated interpolated reported positions.
Remote Sens. 2017, 9, 678 8 of 20

3.4. Recurrent Targets Analysis

A framework in the context of maritime surveillance to identify so-called recurrent detections
(or targets), defined as those detections that appear in the same location in different acquisitions, has
been presented in [38]. This framework exploits the availability of multiple images of the same area
acquired at different times, and it has been applied in this study. Identification of recurrent maritime
detections is important because, often, they are the result of image artefacts (ambiguities from strong
land-based scatterers) and are, therefore, false alarms. In other cases, they indicate real fixed structures
that can also be classified as false alarms (e.g., small unmapped islets or port structures) or structures
that are undesired as targets in some maritime surveillance applications (e.g., buoys or oil platforms).
As mentioned in Section 3.1, SUMO flags the detections that are deemed to be azimuth ambiguities in
an individual image based on the deterministic distance between such an ambiguity and its originating
target. However, SUMO will fail to flag azimuth ambiguities produced by strong scatterers outside the
image and does not attempt to flag range ambiguities. For such ambiguities caused by fixed structures,
this recurrence framework is very valuable. Of all the ship detections in this study, 20% have been
identified as recurrent, and 25% of those have been categorized as ambiguities (see Section 4.4); these
high numbers underline the added value this analysis can bring to SAR-based maritime surveillance.
For completeness, an abridged description of the framework is presented here (see [38] for more
details). Figure 4 shows the block diagram. SAR images are presented, one at a time, to the framework
(input to diagram block A in Figure 4), and the objective is to produce a list of reliable non-recurrent
detections (i.e., to exclude the recurrent detections) for the set of images (one of the outputs of
block C in the figure), and to populate the databases that contain the recurrent targets (block D),
the fixed structures (block F) and the ambiguities (block G). Depending on the application, only
recurrent detections identified as ambiguities will be excluded from the list of detections, retaining
targets like platforms; in other applications, all recurrent targets will be excluded. This choice is
represented as the dashed lines providing input to block B. An image’s metadata is needed to select
the recurrent targets or ambiguities that are relevant to that image (block B), because the location
of the ambiguities is deterministically governed by the observation geometry (azimuth direction,
determined by relative orbit number and left/right antenna pointing) and by sensor parameters (Pulse
Repetition Frequency, determined by mode and beam). The discrimination of recurrent targets in
fixed structures or ambiguities (block E) is done using the idea that the geographic position of a fixed
structure does not depend on the observation geometry or sensor parameters, whereas the geographic
position of an ambiguity does depend on them. Consequently, fixed structures will be visible (and
probably detected unless the structure shows very non-isotropic properties) in the same position for
any observation geometry or set of sensor parameters, but ambiguities will only occur in a specific
geographic position for one combination of geometry and sensor parameters.
This framework simplifies somewhat in this study because all the images are of the same mode
(Sentinel-1 IW). Therefore, the discrimination of recurrent targets in fixed structures or ambiguities is
solely based on the observation geometry, which can be represented as a single number (the relative
orbit number, an integer from 1 to 175) since Sentinel-1’s antenna points always to the right. In the
study, the definition of recurrence is purely based on location: detections in two images are said
to represent the same target if they are separated by less than 50 m. This value has been chosen
empirically: it is short enough to minimise the risk that two different ships in two different images
will be classified as recurrent, and at the same time it still allows for some perturbation in the detected
position of large fixed structures. Furthermore, detection in at least three different images is needed to
classify those detections as recurrent. A source of recurrence is defined as a geographic location in
which recurrent detections appear. As an example, if there are 10 detections in almost exactly the same
geographic location (to within 50 m) and each detection is in a different image, the 10 detections will
be classified as recurrent, and the location marked by the centroid of the 10 detections will become
one source of recurrence. A source of recurrence (and the recurrent detections associated to it) will be
categorised as a fixed structure if it has been detected in at least two different observation geometries.
Remote Sens. 2017, 9, 678 9 of 20
Remote Sens. 2017, 9, 678 9 of 20

Iftwo different
geometric observation
diversity in thegeometries.
acquisitionsIf is
geometric
missing diversity
(i.e., if thatin position
the acquisitions
has onlyisbeen
missing (i.e., if
imaged inthat
one
position has only been imaged in one geometry), it has been decided to categorise that
geometry), it has been decided to categorise that source of recurrence as fixed structure as well. In any source of
recurrence
other asimaged
case (i.e., fixed structure as geometries,
in multiple well. In anybut other case only
detected (i.e., in
imaged
one of in multiple
them), geometries,
a source but
of recurrence
detected only in one of them), a source of recurrence
(and the associated detections) will be categorised as an ambiguity. (and the associated detections) will be
categorised as an ambiguity.

Figure4.4.Block
Figure Blockdiagram
diagramofofthe
therecurrent
recurrenttarget
targetanalysis
analysis framework.
framework. When a new new image
image is presented
presented
(bottomleft),
(bottom left),ship
shipdetection
detectionisisapplied
applied(block
(blockA)A)and
andthe
therelevant
relevant recurrent
recurrent targets are selected
selected (B)
(B) and
and
removed(C),
removed (C),leaving
leaving only
only reliable
reliable detections
detections(bottom
(bottomright). AllAll
right). thethe
detections areare
detections used to populate
used the
to populate
recurrent targets DB (D), which are classified (E) in fixed structures (F) and ambiguities
the recurrent targets DB (D), which are classified (E) in fixed structures (F) and ambiguities (G). (G).

4.4.Results
Resultsand
andDiscussion
Discussion

4.1.Sentinel-1
4.1. Sentinel-1Coverage
Coverageofofthe
theMediterranean
MediterraneanSea
Sea

An Ananalysis
analysisofofthe theSentinel-1
Sentinel-1coverage
coveragein inthe
theAoI
AoI isis presented
presented next. To show the full full monitoring
monitoring
capabilitiesofofthe
capabilities theSentinel-1
Sentinel-1constellation,
constellation, aa time
time window
window from from 29 November 2016 2016 toto 99 February
February
2017 (i.e., 72 days, six complete orbit cycles) was chosen for this analysis. During that window, both
2017 (i.e., 72 days, six complete orbit cycles) was chosen for this analysis. During that window, both
Sentinel-1Aand
Sentinel-1A and Sentinel-1B
Sentinel-1B were
were probably
probably at full
at full capacity
capacity and and operating
operating on anonacquisition
an acquisition plan
plan likely
tolikely to be maintained
be maintained in the future.
in the future. FigureFigure 5 displays
5 displays the average
the average revisitrevisit time
time (in (in days)
days) duringduring that
that time
time window for each geographic location when: (a) only Sentinel-1A acquisitions
window for each geographic location when: (a) only Sentinel-1A acquisitions are considered; and (b) are considered;
andSentinel-1A
both (b) both Sentinel-1A and Sentinel-1B
and Sentinel-1B images are images
used. areThisused. This
figure figure describes
describes more accurately
more accurately the expectedthe
expected monitoring
Sentinel-1 Sentinel-1 monitoring
capabilitiescapabilities
in the AoIinthan the Figure
AoI than 2. Figure
This is2.soThis is so Figure
because because2 Figure
includes 2
includes information from the Sentinel-1A production ramp-up phase, when
information from the Sentinel-1A production ramp-up phase, when the satellite was not yet operating the satellite was not yet
atoperating
full capacityat full
and capacity
modesand EWmodes
and SM EW andbeing
were SM were being sometimes
sometimes used.
used. It has to Itbehas to be out
pointed pointed
that
out that the images in the 29 November 2016 to 9 February 2017 window
the images in the 29 November 2016 to 9 February 2017 window were only used for this coverage were only used for this
coverage
analysis and analysis
were not and were not
analysed in analysed
the maritimein the maritime surveillance
surveillance study, whichstudy,
spans which
from 3 spans
October from
20143
October 2014 to 30 September 2016. In addition, as in Section 2.2, images with
to 30 September 2016. In addition, as in Section 2.2, images with less than 10% overage over sea were less than 10% overage
over
not sea were not included.
included.
Figure 5 shows how the average revisit time halves when both satellites are included in the
Figure 5 shows how the average revisit time halves when both satellites are included in the
analysis. Common average observation periodicities for the two-unit constellation are 1.5 days
analysis. Common average observation periodicities for the two-unit constellation are 1.5 days (areas
(areas in deep red colour in Figure 5b, where two ascending and two descending swaths of each
in deep red colour in Figure 5b, where two ascending and two descending swaths of each satellite
satellite overlap in each 12-day orbit cycle), two days (in light red, three swaths of each satellite
overlap in each 12-day orbit cycle), two days (in light red, three swaths of each satellite overlap),
overlap), three days (in orange, two swaths of each satellite overlap) and four days (in light orange,
three days (in orange, two swaths of each satellite overlap) and four days (in light orange, imaged
imaged in one swath of Sentinel-1A and in two of Sentinel-1B). Common average observation
in one swath of Sentinel-1A and in two of Sentinel-1B). Common average observation periodicities
periodicities of Sentinel-1A alone are three days (areas in orange in Figure 5a, where two ascending
of Sentinel-1A alone are three days (areas in orange in Figure 5a, where two ascending and two
and two descending swaths of Sentinel-1A overlap in each 12-day orbit cycle), four days (in light
descending swaths of Sentinel-1A overlap in each 12-day orbit cycle), four days (in light orange, three
orange, three swaths overlap), six days (in light green, two swaths overlap) and 12 days (in green,
swaths overlap), six days (in light green, two swaths overlap) and 12 days (in green, imaged in one
imaged in one swath only). The northern part of the Med is imaged more often, while the sensors
Remote Sens. 2017, 9, 678 10 of 20

swath only). The northern part of the Med is imaged more often, while the sensors remain frequently
Remote Sens. 2017, 9, 678 10 of 20
off (especially Sentinel-1A) over the southeastern part of the Med. It has to be remembered that, at
theremain
Med Sea latitudes,
frequently offall(especially
the acquisitions of the over
Sentinel-1A) Sentinel-1 satellites take
the southeastern partplace at around
of the Med. It 6:00 a.m.
has to be or
6:00 p.m. local time.
remembered that, at the Med Sea latitudes, all the acquisitions of the Sentinel-1 satellites take place
at around 6:00 a.m. or 6:00 p.m. local time.

Figure
Figure5. 5.Average
Average revisit
revisit time (in
(in days)
days)between
betweenconsecutive:
consecutive: (a) Sentinel-1A
(a) Sentinel-1A acquisitions;
acquisitions; and (b) and
(b)Sentinel-1A
Sentinel-1A+ Sentinel-1B,
+ Sentinel-1B, from
from 29 November
29 November 2016 2016 to 9 February
to 9 February 2017
2017 (72 (72 6days,
days, 6 complete
complete orbit
cycles,
orbit absolute
cycles, orbitorbit
absolute number fromfrom
number 14,160 to 15,209
14,160 for Sentinel-1A,
to 15,209 and from
for Sentinel-1A, 3177 to
and from 4226
3177 to for
4226
forSentinel-1B;
Sentinel-1B;1381
1381Sentinel-1A
Sentinel-1A images
imagesandand
1523 Sentinel-1B
1523 images).
Sentinel-1B Calculated
images). with with
Calculated a grid size of
a grid size
◦.
0.05°.
of 0.05

Figure 6 plots the daily number of acquisitions used in the maritime surveillance study and the
Figure 6 plots the daily number of acquisitions used in the maritime surveillance study and the
area coverage (time frame from 3 October 2014 to 30 September 2016, only Sentinel-1A). An increase
area coverage (time frame from 3 October 2014 to 30 September 2016, only Sentinel-1A). An increase
in the number of acquisitions and the stabilisation of the observation plan is seen as the Sentinel-1A
in the numberwent
operations of acquisitions
through the and the stabilisation
ramp-up phase. Gaps of arethe observation
observed (e.g., inplan is seen
January 2015asand
theJune
Sentinel-1A
2016)
operations went through the ramp-up phase. Gaps are observed (e.g., in January
when the satellite was unavailable. Since the second half of July 2016, the operations have stabilised2015 and June 2016)
when
in a 12-day periodic pattern, with 17–24 daily acquisitions. Using an average image size of 250 × 170 in
the satellite was unavailable. Since the second half of July 2016, the operations have stabilised
a 12-day
km2, theperiodic pattern, ranges
area covered with 17–24
fromdaily acquisitions.
700,000 to 1,000,000 Usingkman average
2 per day; image
this is,size of 250 ×a 170
however, km2 ,
small
theoverestimate
area coveredofranges 2
the seafrom
area 700,000
covered to 1,000,000
since some ofkm the per day;contain
images this is,land
however,
parts. a small overestimate
of the sea area covered since some of the images contain land parts.
4.2. JEODPP Results
4.2. JEODPP Results
The main SUMO processing has been conducted on 11,647 Sentinel-1A products that constitute
The main
a data volume SUMO processing
of around 19 TBhas in been conductedmode.
uncompressed on 11,647
OneSentinel-1A
single core products
has been that constitute
assigned by
a data volume
product of around
resulting in 47219concurrent
TB in uncompressed mode.
jobs; this was Onecluster
the full single capacity
core has at been
theassigned by product
time of workflow
resulting in 472
execution. Theconcurrent jobs;time
total elapsed this was theSUMO
of the full cluster capacity
processing wasat the time of workflow
approximately execution.
2 h. Such fast
Theprocessing enables
total elapsed timetheofexploration
the SUMO of the parameter
processing space in view of2determining
was approximately h. Such fastoptimal settings.
processing enables
Some scripts
the exploration of thetoparameter
extract and aggregate
space in viewmeta-information
of determining optimalfrom the Sentinel-1 products were
settings.
written in Matlab. An additional container having the Matlab Runtime
Some scripts to extract and aggregate meta-information from the Sentinel-1 products Environment installed was
were written
deployed for their execution. Due to the lightness of the specific process
in Matlab. An additional container having the Matlab Runtime Environment installed was deployed and in order to avoid the
overhead of launching and stopping many containers in a short time range, a number of 25 products
Remote Sens. 2017, 9, 678 11 of 20

for their
Remote execution.
Sens. 2017, 9, 678 Due to the lightness of the specific process and in order to avoid the overhead
Remote Sens. 2017, 9, 678
11 of 20
11 of 20
of launching and stopping many containers in a short time range, a number of 25 products was
was associated
associated to everyto every job, resulting
job, resulting in 466in 466 concurrent jobs. The duration of this process was 15 min
was associated to every job, resulting inconcurrent
466 concurrentjobs.jobs.
The The
duration of this
duration process
of this waswas
process 15 15
min minand
and
theand the respective
respective volume
the respective volume of
of theof
volume the
output output data
data (csv
the output (csv
data files) files) reached
reached
(csv files) the
the number
reached number
the number of 30
of 30ofMB. MB.
30 MB.
Figure
Figure
Figure 77 shows
7 shows the
the CPU
shows CPU
CPU workload
the workload of the
of the JEODPP
workload of the cluster
JEODPP
JEODPP cluster
while whilethe
processing
cluster while processing the two
two aforementioned
processing the two
aforementioned
processes. Note that
aforementioned processes. Note
as indicated
processes. Note that
bythat as
the very indicated
low waitby
as indicated by the
time,
theno very low
bottleneck
very wait
low wait time, no
is experienced bottleneck
to access is
time, no bottleneck is
the
experienced
experienced
11,647 Sentinel-1to access
to access the
products 11,647
the thanks Sentinel-1
11,647 Sentinel-1
to the CERNproducts
products thanks to the
thanks to the
EOS distributed CERN
CERN
file system EOS
EOS distributed
distributed
deployed file system
file system
on the JEODPP.
deployed
deployedon onthetheJEODPP.
JEODPP.

Figure 6. 6.Daily
Figure Dailynumber
numberof ofSentinel-1A
Sentinel-1A acquisitions used in
acquisitions used in this study
study andarea
area covered.The
The gaps
Figure 6. Daily number of Sentinel-1A acquisitions used in this
this study and
and areacovered.
covered. Thegaps
gaps
indicate periods of satellite unavailability. Average image area: 250 × 170 km 2.
indicateperiods
indicate periods of
of satellite
satellite unavailability.
unavailability. Average
Average image
image area:
area: 250
250××170
170km
km.2.
2

(a) (b)
(a) (b)
Figure 7. JEODPPcluster
cluster CPU
CPU workload
workload(472
(472cores):
cores):(a) forfor
thethe
SUMO workflow applied to 11,647 S1
Figure7.7.JEODPP
Figure JEODPP cluster CPU workload (472 cores): (a)(a)for SUMO
the SUMO workflow
workflow applied
applied to 11,647
to 11,647 S1
products
S1products (19
products(19 TB);
(19TB); and
TB);and (b)
and(b) for
(b)for the
forthe Matlab
Matlab script-based
the Matlab script-basedmeta-information
meta-informationextraction.
extraction.The
Theblue
blueline
line
script-based meta-information extraction. The blue line
denotes the
denotes load of the user-triggered processes
andand
thethe orangelineline represents
thethe system waiting
denotesthetheload
loadofof
thethe
user-triggered
user-triggered processes
processes and orange
the orange represents
line represents system waiting
the system time.
waiting
time. Horizontal
Horizontal axis
axis is axis is time
timeisintime
HH:MMin HH:MM and vertical
and vertical axis
axis isaxis
per isis per
cent. cent.
time. Horizontal in HH:MM and vertical per cent.

4.3. Ship Detection and Correlation Results

4.3.
4.3.Ship
ShipDetection
DetectionandandCorrelation
CorrelationResults
Results
Table 2 presents the overall detection and correlation results of the study. The total number of
Table22presents
Table presentsthe theoverall
overalldetection
detection and
and correlation
correlation results of the study. The The total
total number
numberof of
detections in the 11,573 (excl. RFI) images was 667,746. SUMO automatically assigned the lowest
detectionsininthe
detections the11,573
11,573(excl.
(excl. RFI)
RFI) images
images waswas 667,746.
667,746. SUMO
SUMO automatically
automatically assigned
assigned thethe lowest
lowest
reliability level to 59,420 of those. These detections are assumed to be false alarms (in most cases they
reliabilitylevel
reliability leveltoto59,420
59,420ofofthose.
those. These
These detections
detections are assumed to be falsefalse alarms
alarms (in
(inmost
most cases
cases they
they
are azimuth ambiguities) and are excluded from the fusion process. The number of reliable
areazimuth
are azimuth ambiguities)
ambiguities) and and
are are excluded
excluded from from
the the fusion
fusion process. process.
The The of
number number
reliableofdetections
reliable
detections was thus 608,326. Most of these (599,278) were detected in dual-pol VV + VH or HH + HV
detections
was
products. was
thus 608,326. thus
Of all Most
the 608,326.
targets Most
of these of these
(599,278)
detected (599,278)
were
in these were
detected
products, indetected
98% dual-pol inVV
dual-pol
+ VH
were detected VVHH
or
in the + VH or HH
+ HV
cross-pol + HV
products.
channel
products. Of all the targets detected in these products, 98% were detected in the
(HV or VH), while 54% were detected in the co-pol channel (HH or VV). These results show that the cross-pol channel
(HV or VH), while
contribution of the 54% were
co-pol detected
channel in study
in this the co-pol channel
was low, i.e., (HH
most or VV). detected
targets These results
in theshow
co-polthat
werethe
contribution of the co-pol channel in this study was low, i.e., most targets detected in the co-pol were
Remote Sens. 2017, 9, 678 12 of 20

Of all the targets detected in these products, 98% were detected in the cross-pol channel (HV or VH),
while 54% were detected in the co-pol channel (HH or VV). These results show that the contribution of
the co-pol channel in this study was low, i.e., most targets detected in the co-pol were also detected
in the cross-pol. This was expected given the high detection threshold used in the co-pol channels,
a measure that was needed to reduce the number of false alarms, as mentioned in Section 3.1.
A total of 818,493 interpolated reported positions were counted within the image footprints at
the image acquisition times. The fusion stage returned 366,081 correlations between detections and
interpolated reported positions. Therefore, 242,245 ship detections and 452,412 interpolated positions
remain uncorrelated. In percentage terms, 40% of all the ship detections (with higher than lowest
reliability level) remain uncorrelated. Many of these uncorrelated detections are expected to be real
ships that either do not transmit AIS or are located in areas of low AIS coverage (in this case their
messages will not be received even if transmitted). Many others are detections that are flagged as
recurrent; since they are either ambiguities or fixed structures that often do not transmit AIS, it is
natural that they are not correlated. They are further analysed in the next section. Some of the
uncorrelated detections are likely to be reporting ships in areas of good AIS coverage for which the
fusion process fails to associate the ship detection with the interpolated position. This is expected due
to the inherently uncertain nature of the fusion, which relies on a spatio-temporal interpolation, and
especially affects ships that report with a low frequency. This situation also results in an uncorrelated
reported position. Finally, some of the uncorrelated detections are expected to be false alarms (e.g., sea
clutter), although this number is probably low due to the high detection thresholds used in this study.
In terms of interpolated reported positions, 55% of them are not correlated to any ship detection.
Most of these are likely to be ships docked at ports (i.e., inside the buffered land mask), which are
areas where SUMO makes no attempt at detection, or small reporting boats that are beyond SUMO’s
fully-automatic detection capabilities (with the high detection thresholds employed in this study) or
even beyond Sentinel-1’s imaging capabilities.
The quoted correlated/uncorrelated fractions are averages over the entire AoI. However, the AIS
coverage is not uniform across the AoI. In the regions of good AIS coverage, which are the ones closer
to the European coast, the correlated fraction will be higher, and vice versa.

Table 2. Summary of ship detection and correlation results (excluding off Israel area).

Number Fraction Fraction

Ship detections 667,746
Detections with the lowest reliability level 59,420
Detections with higher reliability level 608,326 100%
Interpolated reported positions 818,493 100%
Detections correlated to interpolated reported positions 366,081 60% 45%
Uncorrelated detections 242,245 40%
Uncorrelated interpolated reported positions 452,412 55%

4.4. Recurrent Targets Analysis

Table 3 presents the results of the recurrent target analysis. Of the 608,326 reliable detections
returned by SUMO, 122,984 are found to be recurrent and 485,342 non-recurrent. The percentage
of detections that show recurrence is thus 20%. Among the recurrent detections, 75% of them are
classified as fixed structures and 25% as ambiguities. In terms of percentage of the total number
of reliable detections, 15% of them are classified as fixed structures and 5% as ambiguities. These
numbers demonstrate the high rate of occurrence of recurrent detections in the AoI: one in every
five reliable detections is recurrent. Most of these recurrent targets are of no interest for maritime
surveillance purposes: the ones classified as ambiguities are false alarms, and many of the detections
classified as fixed structures are of little importance in most applications (e.g., oil and gas platforms,
buoys, aquaculture equipment). The Mediterranean is a relatively small sea (compared to the world’s
Remote Sens. 2017, 9, 678 13 of 20

oceans) completely surrounded by a basin that is highly populated and has many large ports and
built up coastal sections. This explains to some extent the high frequency of recurrent detections:
ambiguities (range and azimuth) are mostly seen near the coasts as they are primarily produced by
coastal man-made structures, and many fixed structures are associated with ports (e.g., buoys, mooring
points). On the other hand, the Med’s coastline is very well mapped and the tides are of low amplitude.
These factors reduce the number of land masking errors (e.g. significant changes between high and
low tides, reefs exposed at low tide) that can result in recurrent false alarms. In a smaller scale study
in the Western Indian Ocean [22], it was estimated that 12% of the reliable detections were recurrent.
In that case, most of the recurrent detections happened in wetlands and reefs that were not correctly
land masked. It is thought that few such land masking errors occur in the Med.

Table 3. Summary of the recurrence analysis (excluding off Israel area).

Number Fraction Fraction Fraction Fraction

Ship detections (only reliable) 608,326 100%
Non-recurrent detections 485,342 80% 100%
Correlated 320,566 53% 66%
Uncorrelated 164,776 27% 34%
Recurrent detections 122,984 20% 100%
Fixed structures 91,757 15% 75%
Ambiguities 31,227 5% 25%
Sources of recurrence 12,300 100%
Fixed structures 9565 78%
Ambiguities 2735 22%

Among the non-recurrent detections, 34% of them remain uncorrelated, while 66% are correlated
to interpolated reported positions.
Table 3 also shows the total number of the sources of recurrence (12,300), as well as the sources
categorised as fixed structures (9565) and as ambiguities (2735). These numbers give an estimate
of how many locations with fixed structures outside the OpenStreetMap used in SUMO and with
ambiguities from strong land-based scatterers (not already flagged by SUMO) exist in the Med and
can be detected in Sentinel-1 images.

4.5. Density Maps

Maps were computed applying a smoothing kernel to the points. In the different maps, a point
can be a SAR ship detection, an interpolated reported position or a source of recurrence. To compute
the density maps in this section, the points are weighted by the inverse of the number of images,
excluding RFI-labelled images, at each point’s location.
Figure 8 shows the density map of all the ship detections with high reliability level. The main
shipping lanes are clearly visible in the map, as are the high concentration of ships outside many
coastal cities and ports. It has to be remembered that this map has been created aggregating snapshots
taken at 6:00 a.m. and 6:00 p.m. local time only; this temporal filtering hides traffic that only occurs
outside those times. Note that, in Figures 8–12, the results that extend into the Black Sea are not so
statistically robust as they are based on a few images only (see Figure 2).
Figure 9 presents the density map of interpolated reported positions. The main difference between
Figure 8 (detections) and Figure 9 (reported positions) is the high density of reported positions near
the coast of Spain, France and Italy that are not detected in the SAR images. These reported positions
likely correspond to small boats.
Figure 10 maps the uncorrelated, non-recurrent detections. This map evidences the presence of
a large number of ships for which no position reports are available. The map closely reflects the AIS
area coverage: many of the detections in areas with good AIS coverage are excluded from the map, but
detections in areas with bad coverage remain.
Remote
RemoteSens. 2017,9,
Sens.2017, 9,678
678 14ofof20
14 20
Remote Sens. 2017, 9, 678 14 of 20

Figure
Figure
Figure Density
8.8.8.
Density map
Densitymap
mapof all
ofofall the
allthereliable
the reliabledetections
reliable detections(608,326).
detections (608,326).Area
Areaoff
Area offIsrael
off Israelwith
Israel withhigh
with highconcentration
high concentrationof
concentration
ofof RFI-related
RFI-related
RFI-related detections
detections isisindicated
indicated
is indicated
detections aswhite
as aas aa white
white polygon.
polygon.
polygon.

Figure 9. Density map of interpolated reported positions (818,493). Area off Israel with high
Figure 9.9. Density
Figure Density map of interpolated
interpolated reported positions (818,493). Area
positions (818,493). Areaoff
off Israel
Israel with
with high
high
concentration of RFI-related detections is indicated as a white polygon.
concentration
concentrationof
of RFI-related
RFI-related detections
detections is
is indicated
indicated as a white polygon.

Figure 10. Density map of uncorrelated, non-recurrent ship detections (164,776). Area off Israel with
high 10.
Figure
Figure concentration
10. Density of RFI-related
Density map
map of detections
of uncorrelated,
uncorrelated, is indicatedship
non-recurrent as adetections
white polygon.
detections (164,776).Area
(164,776). Areaoff
offIsrael
Israelwith
with
high concentration of RFI-related detections is indicated
high concentration of RFI-related detections is indicated as a white polygon.
Figure 11 displays the weighted geographic distribution of: (a) all the sources of recurrence; (b)
theFigure
sources
11categorised
displays theasweighted
fixed structures; anddistribution
geographic (c) the sources
of: (a)categorised as ambiguities.
all the sources The
of recurrence; (b)
weight applied to each point in these maps is calculated as the ratio between the number
the sources categorised as fixed structures; and (c) the sources categorised as ambiguities. Theof images
weight applied to each point in these maps is calculated as the ratio between the number of images
 Remote Sens. 2017, 9, 678 15 of 20

in which the source of recurrence has been detected over the number of images in which the source
could have been detected, i.e., all the images in the case of the fixed structures, only the images with
Remote Sens. 2017, 9, 678 15 of 20
a given observation geometry in the case of the ambiguities. It is seen that many of the small blobs of
SAR ship detections in Figure 8 correspond to recurrent detections.

(a)

(b)

(c)

Figure 11. Weighted maps of: (a) sources of recurrence (12,300); (b) sources classified as fixed
Figure 11. Weighted maps of: (a) sources of recurrence (12,300); (b) sources classified as fixed structures
structures (9565); and (c) sources classified as ambiguities (2735). In all maps, the sources of
(9565); and (c) sources classified as ambiguities (2735). In all maps, the sources of recurrence are
recurrence are weighted by the ratio between the number of images in which the source has been
weighted
detectedbyover
the the
rationumber
betweenof the number
images of images
in which in which
the source couldthe source
have beenhas been detected
detected Area off over
Israelthe
number of images in which the source could have been detected Area off Israel with
with high concentration of RFI-related detections is indicated as a white polygon. high concentration
of RFI-related detections is indicated as a white polygon.
 the coast are in the north Adriatic Sea, in the Gulf of Gabès (off the coast of Tunisia and Libya) and
off Egypt, areas with intense offshore oil & gas activities, which often employ fixed platforms. The
remaining detections are near the coast. After examination of a few of these coastal places, it has
been concluded that many of these detections correspond to structures near ports (e.g., buoys) or to
aquaculture facilities. However, busy anchor areas east of Malta and south of Istanbul also result in
Remote Sens. 2017, 9, 678 16 of 20
recurrent detections that are categorised as fixed structures. Most of these detections are real ships
anchored at exactly the same position in multiple images. They are, therefore, not fixed structures.

Figure12.12.Density
Figure Density map
map ofof non-recurrent
non-recurrent ship
ship detections
detections (485,342).
(485,342). Area
AreaoffoffIsrael
Israelwith
withhigh
high
concentration
concentration ofofRFI-related
RFI-relateddetections
detectionsisisindicated
indicated as
as a white polygon.
polygon.

The map of ambiguities presents detections distributed fairly uniformly in an area extending up
Figure 11 displays the weighted geographic distribution of: (a) all the sources of recurrence; (b) the
to 200 km off the coastline. The detections closest to the coast (e.g., near Istanbul) are expected to be
sources categorised as fixed structures; and (c) the sources categorised as ambiguities. The weight
azimuth ambiguities not flagged by SUMO. Some of these detections have been visually verified to
applied to each point in these maps is calculated as the ratio between the number of images in which
correspond to the third or even forth order replica of very bright coastal targets, while SUMO only
thechecks
sourcefor
ofthe
recurrence
first and has
secondbeenreplicas.
detected over the further
Detections numberaway of images in which
from the coast arethelikely
sourcetocould have
be range
been detected, i.e., all the images in the case of the fixed structures, only the
ambiguities of coastal or near-coastal cities. The most intense clusters of such ambiguities originate images with a given
observation geometry
from the cities in the case
of Thessaloniki, of the ambiguities.
Bologna, Rome and BarcelonaIt is seen that cities
(these manyare of indicated
the smallin blobs of 1).
Figure SAR
ship
Thedetections
density ofintheFigure 8 correspond
detections is highertoinrecurrent
the northerndetections.
part of the sea, which is the most populated
Looking at the map of fixed structures, the
and built up area of the Med basin. The density of detections only high concentrations
is much lower of in
detections away from
the less populated
thesouthern
coast are in the north Adriatic Sea, in the Gulf of Gabès (off the coast
part. This corroborates the assumption that most of these detections are range ambiguities, of Tunisia and Libya)
and off Egypt,
since areas with
only man-made intense offshore
structures oil & gas
are generally activities,
strong enoughwhich often employ
to generate visible andfixed platforms.
detectable
The remaining
replicas. The detections
maximum distanceare near ofthethese
coast. After examination
detections of a fewalso
from the coastline of these coastal
matches the places, it has
theoretical
distance
been of thethat
concluded rangemanyambiguities in Sentinel-1
of these detections IW images.
correspond to Spain’s
structures Mediterranean
near ports (e.g.,coastbuoys)
is a very
or to
clear example
aquaculture of this:However,
facilities. the limit ofbusythe detections
anchor areasreplicates
east ofalmost
Maltaperfectly
and south theofcoastline
Istanbulatalsoa distance
result in
of 150–200
recurrent km to that
detections the east. At the Med’s
are categorised latitudes,
as fixed the range
structures. Mostdirection in the SARare
of these detections images is
real ships
approximately 15° off east–west. This has to be remembered when looking
anchored at exactly the same position in multiple images. They are, therefore, not fixed structures. at the detections near
Morocco andofAlgeria:
The map even though
ambiguities presents they are very distributed
detections close to the coastline, they arein
fairly uniformly still
anlikely to be rangeup
area extending
to ambiguities
200 km off the originating
coastline. fromThecoastal structures
detections more
closest than
to the 100 (e.g.,
coast km awaynearto the westare
Istanbul) or to the eastto
expected ofbe
the detection.
azimuth ambiguities not flagged by SUMO. Some of these detections have been visually verified to
Finally,
correspond Figure
to the third12or maps
eventhe non-recurrent
forth order replica detections. It can coastal
of very bright be interpreted
targets, as the SUMO
while difference
only
between all the detections (Figure 8) and the recurrent detections (not shown, but very similar to
checks for the first and second replicas. Detections further away from the coast are likely to be range
Figure 11a) and gives a much better representation of the true shipping patterns than Figure 8.
ambiguities of coastal or near-coastal cities. The most intense clusters of such ambiguities originate
Figure 13 displays close up sections of Figure 12 south of Sicily and around Sardinia. Many shipping
from the cities of Thessaloniki, Bologna, Rome and Barcelona (these cities are indicated in Figure 1).
details are visible, related to traffic routes (north of Tunisia), off-shore supply (Gulf of Gabès),
The density of the detections is higher in the northern part of the sea, which is the most populated
anchor area (east of Malta), and fishing activities (around Sardinia and Malta).
and built up area of the Med basin. The density of detections is much lower in the less populated
southern part. This corroborates the assumption that most of these detections are range ambiguities,
since only man-made structures are generally strong enough to generate visible and detectable replicas.
The maximum distance of these detections from the coastline also matches the theoretical distance of
the range ambiguities in Sentinel-1 IW images. Spain’s Mediterranean coast is a very clear example of
this: the limit of the detections replicates almost perfectly the coastline at a distance of 150–200 km
to the east. At the Med’s latitudes, the range direction in the SAR images is approximately 15◦ off
east–west. This has to be remembered when looking at the detections near Morocco and Algeria: even
though they are very close to the coastline, they are still likely to be range ambiguities originating from
coastal structures more than 100 km away to the west or to the east of the detection.
Remote Sens. 2017, 9, 678 17 of 20

Finally, Figure 12 maps the non-recurrent detections. It can be interpreted as the difference
between all the detections (Figure 8) and the recurrent detections (not shown, but very similar to
Figure 11a) and gives a much better representation of the true shipping patterns than Figure 8. Figure 13
displays close up sections of Figure 12 south of Sicily and around Sardinia. Many shipping details are
visible, related to traffic routes (north of Tunisia), off-shore supply (Gulf of Gabès), anchor area (east of
Malta), and fishing activities (around Sardinia and Malta).
Remote Sens. 2017, 9, 678 17 of 20

(a) (b)

Figure
Figure 13. 13. Close
Close upup sections:(a)(a)South
sections: SouthofofSicily;
Sicily; and
and (b)
(b) around
around Sardinia
Sardinia of
of the
the density
density map
map of
of
non-recurrent ship detections (Figure
non-recurrent ship detections (Figure 12). 12).

5. Conclusions
5. Conclusions
The free, full, and open nature of the Sentinel-1 SAR images gives access to unprecedented
The free, full, and open nature of the Sentinel-1 SAR images gives access to unprecedented
volumes of data. This presents opportunities, but these will not come to fruition unless the
volumes of data.
challenges thatThis
alsopresents
come with opportunities,
the huge data but these are
volume willconfronted.
not come toThis fruition
paperunless the challenges
has presented tools
thatand
alsoplatforms
come withdeveloped
the huge dataat the JRC to deal with these challenges, and the resultsand
volume are confronted. This paper has presented tools thatplatforms
can be
developed
achievedatforthemaritime
JRC to deal with theseTwo
surveillance. challenges, and the
years’ worth of results
Sentinel-1thatimagery
can be achieved for maritime
of the Mediterranean
surveillance. Two years’ worth of Sentinel-1 imagery of the Mediterranean
Sea (around 11,500 products) have been automatically analysed with SUMO (a ship detection Sea (around 11,500 products)
tool)
haveonbeen automatically
the JEODPP platform analysed with SUMO
with a processing time(a of
ship
<2 detection tool) onofthe
h. The versatility theJEODPP
JEODPP platform
platform waswith
a processing
key to thetime of <2ofh.this
success Theanalysis.
versatility of thethe
Indeed, JEODPP
JEODPP platform
enableswas key toand
complex the consolidated
success of thisscientific
analysis.
Indeed, the JEODPP
workflows such as enables
SUMO complex
developedand byconsolidated
domain expertsscientific workflows
over the yearssuchto be asdeployed
SUMO developed
at scale
by domain
without experts
embarking over into
the years to be deployed
the code rewriting, atanscale withoutthat
operation embarking
can proveinto very
the code rewriting,
lengthy and
challengingthat
an operation without guarantee
can prove very of success.
lengthy Thechallenging
and ship detections (more
without than 600,000)
guarantee have then
of success. Thebeen
ship
fused with
detections shipthan
(more reporting
600,000)data (AIS)
have on been
then the Blue Hub
fused maritime
with surveillance
ship reporting dataplatform.
(AIS) on the Blue Hub
maritimeThe automatic platform.
surveillance analysis of such a large set of images taken under diverse weather conditions
means that certainanalysis
The automatic steps have to be ataken
of such largetosetprevent the detection
of images taken under of toodiverse
many false alarms.
weather Besides
conditions
the exclusion of images with too much RFI, the most important adaptation
means that certain steps have to be taken to prevent the detection of too many false alarms. Besides is raising the detection
the thresholds
exclusion of in particular
images with in the
tooco-pol
muchchannels.
RFI, the mostConsequently,
importantsmall boats are
adaptation is not detected.
raising Further
the detection
work would be needed to better discard false alarms fully automatically
thresholds in particular in the co-pol channels. Consequently, small boats are not detected. Further at lower detection
workthresholds.
would be needed to better discard false alarms fully automatically at lower detection thresholds.
On the other hand, the availability of repeat observations allows for the identification of
On the other hand, the availability of repeat observations allows for the identification of recurrent
recurrent detections, most of which are false alarms or otherwise of no interest for maritime
detections, most of which are false alarms or otherwise of no interest for maritime surveillance
surveillance purposes. The high presence of these recurrent detections (about 20% of all the reliable
purposes. The high presence of these recurrent detections (about 20% of all the reliable detections)
detections) indicates the value added by this analysis, especially in coastal areas. In fact, the
indicates the value added by this analysis, especially in coastal areas. In fact, the recurrent targets
recurrent targets analysis teaches us that when analysing a SAR image on an individual basis, a
analysis teaches us that when analysing a SAR image on an individual basis, a significant fraction of
significant fraction of the detections may be ambiguities, which would never be recognised because
the the
detections
ambiguity may be ambiguities,
source is not insidewhich would
the image never beunnoticed.
or remains recognised Inbecause theover
this study, ambiguity
the Med source
Sea,
the fraction is 5 % on average and higher than that in areas near urbanised coasts.
The results show that the big dataset of Sentinel-1 images can be used to map shipping activity,
revealing the main shipping lanes, ports, anchor areas, off-shore supply and fishing activities. Even
if limited to the detection of medium and large ships, comparison with AIS data shows the presence
of many ships for which no position reports are available.
Remote Sens. 2017, 9, 678 18 of 20

is not inside the image or remains unnoticed. In this study, over the Med Sea, the fraction is 5 % on
average and higher than that in areas near urbanised coasts.
The results show that the big dataset of Sentinel-1 images can be used to map shipping activity,
revealing the main shipping lanes, ports, anchor areas, off-shore supply and fishing activities. Even if
limited to the detection of medium and large ships, comparison with AIS data shows the presence of
many ships for which no position reports are available.

Acknowledgments: Contains modified Copernicus Sentinel data [2014–2017]. Satellite AIS data from AISSat-1,
AISSat-2 and NORAIS were obtained courtesy of the Norwegian Coastal Administration and the Norwegian
Defence Research Establishment (FFI). MSSIS data were obtained courtesy of the Volpe Center of the US
Department of Transportation, the US Navy and SPAWAR. Terrestrial AIS from an Italian network of coastal
stations was obtained courtesy of the Italian Coast Guard. Land masking in SUMO was done using coastline data
derived from OpenStreetMap, © OpenStreetMap contributors.
Author Contributions: Santamaria, Alvarez and Greidanus designed the study and were responsible for SUMO
and Blue Hub. Syrris and Soille were responsible for the deployment of SUMO on the JEODPP. Argentieri adapted
SUMO to run on a Linux operating system. Santamaria carried out the scientific analysis and prepared the
manuscript, with contributions from Syrris and Soille for the JEODPP sections.
Conflicts of Interest: The authors declare no conflict of interest.

Abbreviations
The following abbreviations are used in this manuscript:

AIS Automatic Identification System

AoI Area of Interest
CERN European Organization for Nuclear Research
CFAR Constant False Alarm Rate
EOS Distributed file system developed by CERN
EU European Union
EW Extra Wide mode
FFI Norwegian Defence Research Establishment
GRDH Ground Range Detected High resolution
IW Interferometric Wide mode
JBOD Just a Bunch of Disks
JEODPP JRC Earth Observation Data and Processing Platform
JRC Joint Research Centre
LRIT Long Range Identification and Tracking
MSSIS Maritime Safety and Security Information System
SAR Synthetic Aperture Radar
SLC Single Look Complex
SM Stripmap mode
Search for Unidentified Maritime Objects (JRC’s algorithm for ship detection in satellite
SUMO
SAR images)
VMS Vessel Monitoring System
WV Wave mode
XML Extensible Markup Language

References
1. International Telecommunication Union. Technical Characteristics for an Automatic Identification System
Using Time-Division Multiple Access in the VHF Maritime Mobile Band. Available online: https://www.itu.
int/dms_pubrec/itu-r/rec/m/R-REC-M.1371-5-201402-I!!PDF-E.pdf (accessed on 29 June 2017).
2. International Maritime Organization. Long-Range Identification and Tracking System—Technical
Documentation. Available online: http://www.imo.org/en/OurWork/Safety/Navigation/Documents/
LRIT/1259-Rev-7.pdf (accessed on 29 June 2017).
Remote Sens. 2017, 9, 678 19 of 20

3. Food and Agriculture Organization (FAO) of the UN. FAO Technical Guidelines for Responsible Fisheries
Monitoring Systems. Available online: http://www.fao.org/3/a-w9633e.pdf (accessed on 29 June 2017).
4. Council Regulation (EC) No 1224/2009 of 20 November 2009 establishing a Community control system
for ensuring compliance with the rules of the common fisheries policy, Art. 11. Available online: http:
//eur-lex.europa.eu/legal-content/EN/ALL/?uri=CELEX:32009R1224 (accessed on 29 June 2017).
5. Vachon, P.W.; Kabatoff, C.; Quinn, R. Operational ship detection in Canada using RADARSAT. In Proceedings
of the IGARSS 2014, Quebec, Canada, 13–18 July 2014; pp. 998–1001.
6. European Maritime Safety Agency. The CleanSeaNet Service: Taking Measures to Detect and Deter Marine
Pollution. Available online: http://www.emsa.europa.eu/csn-menu.html (accessed on 29 June 2017).
7. Crisp, D.J. The State-of-the-Art in Ship Detection in Synthetic Aperture Radar Imagery; Information Sciences
Laboratory: Edinburgh, SA, Australia, 2004.
8. Greidanus, H.; Alvarez, M.; Santamaria, C.; Thoorens, F.X.; Kourti, N.; Argentieri, P. The SUMO ship detector
algorithm for satellite radar images. Remote Sens. 2017, 9, 246. [CrossRef]
9. Greidanus, H.; Kourti, N. Findings of the DECLIMS project—Detection and classification of marine traffic
from space. In Proceedings of the SEASAR 2006: Advances in SAR Oceanography from ENVISAT and ERS,
ESA-ESRIN, Frascati, Italy, 23–26 January 2006; pp. 23–26.
10. Stasolla, M.; Mallorqui, J.; Margarit, G.; Santamaria, C.; Walker, N. A comparative study of operational vessel
detectors for maritime surveillance using satellite-borne synthetic aperture radar. IEEE J. Sel. Top. Appl. Earth
Obs. Rem. Sens. 2016, 9, 2687–2701. [CrossRef]
11. Pelich, R.; Longépé, N.; Mercier, G.; Hajduch, G.; Garello, R. Performance evaluation of Sentinel-1 data
in SAR ship detection. In Proceedings of the 2015 IEEE International Geoscience and Remote Sensing
Symposium (IGARSS), Milan, Italy, 26–31 July 2015; pp. 2103–2106.
12. Friedman, K.S.; Wackerman, C.; Funk, F.; Pichel, W.G.; Clemente-Colon, P.; Li, X. Validation of a CFAR vessel
detection algorithm using known vessel locations. In Proceedings of the 2001 IEEE International Geoscience
and Remote Sensing Symposium (IGARSS), Sydney, NSW, Australia, 9–13 July 2001; pp. 1804–1806.
13. Greidanus, H.; Santamaria, C.; Alvarez, M.; Krause, D.; Stasolla, M.; Vachon, P.W. Non-reporting ship traffic
in the western Indian Ocean. In Proceedings of the ESA Living Planet Symposium, Prague, Czech Republic,
9–13 May 2016.
14. Santamaria, C.; Greidanus, H.; Fournier, M.; Eriksen, T.; Vespe, M.; Alvarez, M.; Arguedas, V.F.; Delaney, C.;
Argentieri, P. Sentinel-1 contribution to monitoring maritime activity in the Arctic. In Proceedings of the
ESA Living Planet Symposium, Prague, Czech Republic, 9–13 May 2016.
15. Sandirasegaram, N.; Vachon, P.W.; English, R.A.; Olsen, R.; Eriksen, T.; Hannevik, T.N. Validation of OceanSuite
Ship Detection in RADARSAT-2 Imagery with AISSat-1 Space-based and MSSIS Coastal AIS Data; Defence
Research and Department Canada: Ottawa, ON, Canada, 1 December 2013.
16. S-1 Mission Performance Center. Sentinel-1 Product Specification. Available online: https://sentinel.esa.int/
documents/247904/349449/Sentinel-1_Product_Specification (accessed on 30 June 2017).
17. Sentinel-1 Production Scenario. Available online: https://sentinel.esa.int/web/sentinel/missions/sentinel-
1/production-scenario (accessed on 30 June 2017).
18. Access to Sentinel Data. Available online: https://sentinels.copernicus.eu/web/sentinel/sentinel-data-
access/access-to-sentinel-data (accessed on 30 June 2017).
19. Sentinel-1 Observation Scenario. Available online: https://sentinel.esa.int/web/sentinel/missions/sentinel-
1/observation-scenario (accessed on 30 June 2017).
20. Copernicus Space Component Mission Management Team. Sentinel High Level Operations Plan (HLOP).
Available online: https://sentinel.esa.int/documents/247904/685154/Sentinel_High_Level_Operations_
Plan (accessed on 30 June 2017).
21. Greidanus, H.; Santamaria, C. First Analyses of Sentinel-1 Images for Maritime Surveillance; Publications Office
of the European Union: Luxembourg, 2014.
22. Santamaria, C.; Stasolla, M.; Fernandez Arguedas, V.; Argentieri, P.; Alvarez, M.; Greidanus, H. Sentinel-1
Maritime Surveillance: Testing and Experiences with Long-term Monitoring; Publications Office of the European
Union: Luxembourg, 2015.
Remote Sens. 2017, 9, 678 20 of 20

23. Rommen, B.; Barat, I.; Chabot, M.; Lambert, C.; Williams, D. Joint investigations on RADARSAT-2/
Sentinel-1A mutual RFI. In Proceedings of the Committee on Earth Observation Satellites (CEOS) Working
Group on Calibration and Validation (WGCV), SAR Subgroup, Synthetic Aperture Radar Workshop 2015,
Noordwijk, The Netherlands, 27 October 2015.
24. Meadows, P.; Pilgrim, A.; Piantanida, R.; Riva, D.; Miranda, N. Sentinel-1A radiometric calibration.
In Proceedings of the Committee on Earth Observation Satellites (CEOS) Working Group on Calibration and
Validation (WGCV), SAR Subgroup, Synthetic Aperture Radar Workshop 2015, Noordwijk, The Netherlands,
27 October 2015.
25. Skauen, A.N. Quantifying the tracking capability of space-based AIS systems. Adv. Space Res. 2016, 57,
527–542. [CrossRef]
26. Kourti, N.; Shepherd, I.; Greidanus, H.; Alvarez, M.; Aresu, E.; Bauna, T.; Chesworth, J.; Lemoine, G.;
Schwartz, G. Integrating remote sensing in fisheries control. Fish. Manag. Ecol. 2005, 12, 295–307. [CrossRef]
27. Li, F.K.; Johnson, W.T.K. Ambiguities in spaceborne synthetic aperture radar systems. IEEE Trans. Aerosp.
Electron. Syst. 1983, 19, 3.
28. OpenStreetMap. Available online: https://www.openstreetmap.org/ (accessed on 3 May 2017).
29. Soille, P.; Burger, A.; Rodriguez, D.; Syrris, V.; Vasilev, V. Towards a JRC earth observation data and processing
platform. In Proceedings of the 2016 Conference on Big Data from Space (BiDS’16), Santa Cruz de Tenerife,
Spain, 15–17 March 2016.
30. Soille, P.; Burger, A.; De Marchi, D.; Kempeneers, P.; Rodriguez, D.; Syrris, V.; Vasilev, V. A versatile
data-intensive computing platform for information retrieval from big geospatial data. Futur. Gen.
Comput. Syst. 2017, submitted.
31. Peters, A.; Janyst, L. Exabyte scale storage at CERN. Available online: https://cds.cern.ch/record/2134571/
files/Disk%20storage%20at%20CERN.pdf (accessed on 30 June 2017).
32. Peters, A.; Sindrilaru, E.; Adde, G. EOS as the present and future solution for data storage at CERN. Available
online: http://iopscience.iop.org/article/10.1088/1742-6596/664/4/042042 (accessed on 30 June 2017).
33. Copernicus Open Access Hub. Available online: https://scihub.copernicus.eu/ (accessed on 30 June 2017).
34. Copernicus Services Data Hub. Available online: https://cophub.copernicus.eu/ (accessed on 30 June 2017).
35. Merkel, D. Docker: Lightweight Linux containers for consistent development and deployment.
Available online: http://www.linuxjournal.com/content/docker-lightweight-linux-containers-consistent-
development-and-deployment (accessed on 30 June 2017).
36. Thain, D.; Tannenbaum, T.; Livny, M. Distributed computing in practice: The Condor experience.
Concurr. Comput. Pract. Exp. 2005, 17, 323–356. [CrossRef]
37. Mazzarella, F.; Alessandrini, A.; Greidanus, H.; Alvarez, M.; Argentieri, P.; Nappo, D.; Ziemba, L. Data fusion
for wide-area maritime surveillance. In Proceedings of the COST MOVE Workshop on Moving Objects at
Sea, Brest, France, 27–28 June 2013.
38. Santamaria, C.; Greidanus, H. Ambiguity discrimination for ship detection using Sentinel-1 repeat acquisition
operations. In Proceedings of the IGARSS 2015, Milan, Italy, 26–31 July 2015; pp. 2477–2480.

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).