Professional Documents
Culture Documents
We are soliciting articles on the following topics as well as entertaining alternative suggestions from
interested contributors:
The deadline for submission of initial manuscripts is Monday, July 1st, 2024. Final upload of accepted paper
proofs is September 1st, 2024. All production ready manuscripts will be posted on IEEE Xplore Early Access and
will appear in final paginated sequence in the Special Issue – scheduled for final release in October 2024.
Contributed papers should be targeted at ten pages but review and special invited papers can be longer. All
submissions will be reviewed in accordance with the normal procedures of the journal. Please tag uploaded
papers as “Special Issue” through our Author Portal.
We hope you will consider contributing to this Special Issue of IEEE Journal of Microwaves and continue to
support the journal through your regular research submissions.
FEATURES
21 Airborne Lidar Data Artifacts
by Wai Yeung Yan
SCOPE
IEEE Geoscience and Remote Sensing Magazine (GRSM) will
inform readers of activities in the IEEE Geoscience and
Remote Sensing Society, its technical committees,
and chapters. GRSM will also inform and educate
readers via technical papers, provide information on
international remote sensing activities and new satellite
missions, publish contributions on education activities,
industrial and university profiles, conference news, book
reviews, and a calendar of important events.
IEEE Geoscience and Remote Sensing Magazine (ISSN 2473-2397) is published provided the per-copy fee indicated in the code is paid through the Copyright Clear-
quarterly by The Institute of Electrical and Electronics Engineers, Inc., IEEE ance Center, 222 Rosewood Drive, Danvers, MA 01923 USA; 2) pre-1978 articles
Headquarters: 3 Park Ave., 17th Floor, New York, NY 10016-5997, +1 212 419 without fee. For all other copying, reprint, or republication information, write to:
7900. Responsibility for the contents rests upon the authors and not upon the Copyrights and Permission Department, IEEE Publishing Services, 445 Hoes Lane,
IEEE, the Society, or its members. IEEE Service Center (for orders, subscrip- Piscataway, NJ 08854 USA. Copyright © 2023 by the Institute of Electrical and Elec-
tions, address changes): 445 Hoes Lane, Piscataway, NJ 08854, +1 732 981 tronics Engineers, Inc. All rights reserved. Application to Mail at Periodicals Postage
0060. Individual copies: IEEE members US$20.00 (first copy only), nonmem- Prices is pending at New York, New York, and at additional mailing offices. Cana-
bers US$110.00 per copy. Subscription rates: included in Society fee for each dian GST #125634188. Canada Post Corporation (Canadian distribution) publica-
member of the IEEE Geoscience and Remote Sensing Society. Nonmember tions mail agreement number 40013885. Return undeliverable Canadian addresses
subscription prices available on request. Copyright and Reprint Permissions: to PO Box 122, Niagara Falls, ON L2E 6S8 Canada. Printed in USA.
Abstracting is permitted with credit to the source. Libraries are permitted to
photocopy beyond the limits of U.S. Copyright Law for private use of patrons: IEEE prohibits discrimination, harassment, and bullying. For more information,
1) those post-1977 articles that carry a code at the bottom of the first page, visit http://www.ieee.org/web/aboutus/whatis/policies/p9-26.html.
I n line with what I did for the June issue, I will use my
IEEE Geoscience and Remote Sensing Magazine (GRSM)
editorial on one hand to summarize the content of the
operating phase. Still considering radar systems but
focusing on Interferometric Synthetic Aperture Radar
(InSAR), [A2] reviews the phase linking algorithms that
current issue and, on the other hand, to introduce a fea- have been proposed with respect to their efficiency, pre-
ture of this magazine that may not be well known to cision, and usefulness with respect to final applications.
(or understood by) all our readers. Specifically, I will de- As a preprocessing tool for various SAR applications,
scribe the possibility to publish special issues in GRSM. these algorithms well match the topic of the following
As mentioned in my first editorial this year, GRSM ac- article [A3], which focuses on the many different types
cepts proposals for special issues. In June I spent some of artifacts that affect light detection and ranging (lidar)
time to explain why articles should be submitted in a datasets. The comprehensive analysis of these artifacts
white paper format; here I will go a little bit into explain- and the most common techniques to reduce them is a
ing how white papers and special issues go together and very good introduction of preprocessing techniques ap-
what types of special issues are welcome in GRSM. plied to lidar data. Finally, the last technical article in
the issue is devoted to datasets coming from multiple
IEEE GEOSCIENCE AND REMOTE SENSING sources and their relationship with deep learning tech-
MAGAZINE ISSUE CONTENT niques [A4]. Indeed, Earth observation (EO) datasets
Let us first start, however, by describing the contents of are very different from what is usually considered for
this issue, which is full of valuable technical articles, algorithm development in computer vision, and a better
as well as interesting and timely columns providing in- knowledge of the datasets that are available and their
sights about what happens in our Society. Indeed, this features is a way to strengthen the link between the
issue includes four technical articles and five columns. GRSS and the computer vision community without los-
As you have seen, I am pushing to have more columns ing the link with the physics at the very basis of EO data.
than in the past years because columns are meant to The columns are actually well connected to the
provide information about the activities of the Society fourth technical article by means of the first two of
and are very valuable for IEEE Geoscience and Remote them because they are reports about new datasets that
Sensing Society (GRSS) members. Columns have both are available, one for deep learning and the other one for
an informative and an engaging purpose: readers are classification purposes. The first column [A5] provides
informed about what happens in the various GRSS details about a large-scale, global, multimodal, and mul-
committees and working groups and find out how they tiseasonal corpus of satellite imagery from multispectral
could be involved, if interested. and radar sensors. The second column [A6] introduces
The technical content of this issue encompasses an open dataset that tackles the burned area delineation
hardware and software and refers to multiple sensors problem using prefire and postfire Sentinel-2 acquisi-
and systems. The first article [A1] introduces the struc- tions of California forest fires that took place starting in
ture, history, and performance of the Cosmos-1500 real 2015. The following two columns refer to activities per-
aperture radar, with an interesting description of the formed in the framework of the recent 2023 IEEE Inter-
challenges faced by its developers in the designing and national Geoscience and Remote Sensing Symposium.
In [A7], the interested reader will find the highlights of
Digital Object Identifier 10.1109/MGRS.2023.3304503
the awards and opening sessions during the first day of
Date of current version: 19 September 2023 the conference. The next column [A8] summarizes the
FIGURE 1. On 17 June 2023, the GRSS signed an MoU with FIGURE 2. On 17 June 2023, the GRSS signed an MoU with its
its IEEE Sister Society, the AP-S. On the left, Dr. Mariko Burgin, IEEE Sister Society, the MTT-S. On the left, Dr. Mariko Burgin,
president of the GRSS. On the right, Dr. Stefano Maci, president president of the GRSS. On the right, Dr. Nuno Borges Carvalho,
of the AP-S. president of the MTT-S.
Pen. Chukotka
(b)
Ejected
Reactor Reactor
Core Container
5.8 m Attitude Thruster Pod
3,495
892
1,075 257
Reactor Radiator Propulsion
System Main Spacecraft Side-Looking Radar Antenna,
to Raise Including Radar Payload One on Each Side of the Spacecraft f = 8.2 GHz
19.542-MHz Transmitter in this Part Reactor 166-MHz Transmitter 0 5
in this Part (m)
(a)
(b)
FIGURE 7. (a) RORSAT concept and configuration. (Source: [28].) (b) A demonstration copy at the DB “Arsenal” in St. Petersburg showing
the nuclear reactor at the forward end and two unfolded SLR antennas in the rear. (Source: [24].)
FIGURE 8. (a) “Legenda” satellite receiver antennas on the Pyotr Velikiy nuclear-powered USSR cruiser (a)—white radome on a vertical
structure at the side of the tower—and (b) on the sunken Moskva cruiser (b)—similar light-grey radome just above the rear launch tube.
To eliminate the shadowing of antennas, the same equipment was placed at the portside.
The whole USSR orbital naval reconnaissance and reactor into a high “graveyard” orbit and instead crashed
targeting system “Legenda” carried the same stigma of over Canada’s Great Slave Lake, contaminating a wide area.
hopeless gigantomania and kamikaze spirit. The USSR The debris was examined by U.S. Lawrence Livermore Na-
electronics of that time were quite backward and unreli- tional Laboratory scientists, which enabled them to get a
able; additionally, the signals backscattered from ships better understanding of the design and mission of ROR-
are accompanied by intensive clutter due to the sea SATs. This catastrophe led to a two-year break in RORSAT
waving and precipitation effects. To compensate for in- launches, used for improvements in their design. Still, an-
sufficient sensitivity and poor signal processing, USSR other similar accident happened at the beginning of 1983
developers used an extremely monstrous approach. First, with Cosmos-1402 when separate parts of the reactor fell
the RORSATs used small fast neutron nuclear reactors into the Indian and Atlantic Oceans.
(“Buk”) to provide the 3-kW power needed to feed the ra-
dar; second, they always flew at low orbits of 250–270 km
with a 65° inclination that made their lifetimes short,
less than two months on average. Even more—to enable
determining the direction and speed of the sea target
with such an incoherent sensor as SLR—a primitive but
efficient solution was found—nuclear RORSATs were
launched in pairs and placed into identical orbits with
a half-hour separation [28].
The combination of a low orbit and a nuclear power
source introduced a serious risk of accident or uncontrolled
reentry [29].
“To counter the problem, each RORSAT consisted of (a)
three major components: the payload and propul-
sion section, the reactor, and a disposal stage used to
lift the reactor into a higher orbit, with an altitude of
900 km, at the end of the mission.”
Each of at least 33 reactors launched in 1975–1988 con-
tained more than 30 kg of weapon-grade (enriched to 90%)
uranium-235, besides the sodium-potassium coolant. This
means that presently about 940 kg of highly enriched ura-
nium and a further 15 tons, mostly shaped as tens of thou-
sands of radioactive coolant droplets, 0.6–2 cm in diameter, (b)
orbit Earth [29].
There were several accidents of the malfunctioning FIGURE 9. (a) and (b) “Legenda” receiver antenna in a radome on
of RORSATs that triggered public attention to the dan- the top of the coning tower of the USSR Oskar type submarines
ger they presented. On 24 January 1978, five years before (one of them was the ill-fated Kursk, which exploded in 2000)
Kalmykov’s success, Cosmos-954 failed to throw its nuclear (a) in the harbor and (b) at sea. (Sources: [32], [33].)
TABLE 1. AIRBORNE LIDAR DATA ARTIFACTS AND THEIR EFFECTS ON THE DERIVATIVES.
Adverse Potential
Effects Causes
Airborne Lidar
Time
H
First Return
hb Second Return
Nadir s
FIGURE 2. (a) The occluded region, s, is blocked by an elevated object. Here, H is the flying altitude, db is the horizontal distance between
the nadir point and the object, and hb is the height of the object. (b) A laser beam penetrates the tree crown, resulting in multiple returns.
The final return may cause an appearance of a data void due to occlusion caused by the overstory and/or a data pit in a CHM.
Elevation (m)
1,550 1,064 532 1,550 1,064 532 1,550 1,064 532 1,550 1,064 532
(nm) (nm) (nm) (nm)
Elevation (m) Intensity SLIER Land–Water Classification
138 146 0 4096 Low High Water Land
(a) (b) (c) (d)
FIGURE 6. (a) Laser dropouts found in a water region, particularly at the swath edges. (b) A high variance of the intensity values on the
water surface. (c) High SLIER values computed on the water region, particularly on laser channels of 1,064 and 1,550 nm. (d) The use of the
SLIER for land–water classification [36].
(d)
378
Height (m)
0 0.2 km
243
FIGURE 7. (a) Laser dropouts cause data voids along an inland river. (b) The implementation of a scan line void-filling algorithm implement-
ed at the close-to-nadir region and swath edges. (c) The resulting hydroflattened DEM. (d) The corresponding 3D view of the original DEM
with unpleasant triangular facets along the river. (e) The hydroflattened DEM [44].
SWATH GAPS not be the case for systems equipped with rotating prisms.
Another explanation of void appearance can be attributed The change of swath width also happens if the flight alti-
to the gaps found between swaths. In common practice, a tude changes during the flight mission. This may also lead
15%–50% overlap between adjacent data strips is a rule of to the formation of a gap between two swaths. Second, data
thumb during mission planning [13], [14], [45]. Combin- gaps may be found between two data strips, which can be
ing a number of partially overlapping data strips can lead ascribed to the accuracy of the GNSS and inertial measure-
to a seamless mapping of the study region. Putting aside ment unit (IMU). Nevertheless, data gaps may be particu-
the possibility of poor flight planning, adjacent flight lines larly obvious along the vertical direction in the overlapping
may have gaps due to various internal and external factors. region. Such a case also would lead to stripe artifacts in the
Instantaneous flight conditions, such as air turbulence or a resulting DEM. Strip adjustment should therefore be ap-
change of heading direction, likely cause undesired move- plied to remove system errors and biases so as to improve
ments of the aircraft, resulting in a crooked flight trajec- the geometric accuracy. Details can be found in the “Stripe
tory. It is a common problem not only in an airborne lidar Artifacts in Digital Elevation Models” section.
system but also in a UAS platform. As demonstrated in Fig-
ure 8, a portion of the airborne lidar data strip is twisted
with a curvy swath due to possible turbulence. A notable
void is thus found between the swaths, a scenario consid-
ered an unintentional cause of data voids. On the other
hand, a change of system settings or system error/bias may
also lead to the presence of swath gaps. This situation can
be further broken down into the following two scenarios.
First, a significant change of terrain height, particularly
on rugged mountainous terrain, may alert the swath of the
airborne lidar if the flying altitude remains unchanged.
Combining two adjacent strips under this circumstance
may lead to the appearance of a swath gap. Although
topographic airborne lidar systems operated with oscillat-
ing mirrors are capable of maintaining a consistent swath FIGURE 8. A swath gap with an annulus-shaped sector due to a
width via adjusting the degree of angular rotation, this may crooked flight trajectory.
Transmitter’s Laser
Transmitter’s Axis is Not Parallel
Receiver’s FOV Beam
Axis to the Receiver’s axis
Falls Receiver’s FOV Mirror Accelerates
Cross Section of Outside
Receiver's FOV Receiver’s
Cross-Section View
Laser
FOV
Beam
FIGURE 11. The imperfect alignment of a laser beam’s axis and receiver’s FOV in a specific direction, causing intensity banding [66].
Asphalt
Nadir Surface
Half of
Swath Width Footprint Striping at Overlapping
Striping Found in
Swath Edges Region
Swath Edge the Entire
Overlapping Region
(a) (b) (c)
FIGURE 12. (a) Laser attenuation found at long ranges and large incident angles, resulting in smaller intensity values at the swath edges
compared to those at the nadir region. (b) Striping found at the swath edges when combining two overlapping data strips. (c) Striping found
in the entire overlapping region, due to different flight altitudes.
(a)
Swath Edge
(b)
0 40 m
(c)
FIGURE 13. Two overlapping swaths of a multispectral airborne lidar dataset, one with a large scan angle at the bottom and the other with a
small scan angle (near the nadir region) at the top. The (a) original intensity, (b) original intensity + LSLC (removal of the intensity banding),
and (c) original intensity + LSLC + overlap intensity correction (further removal of the striping at the swath edge).
FIGURE 16. Speckle intensity noise appears in different types of water regions, such as (a) inland rivers, (b) natural shores, (c) inland lakes,
and (d) rocky shores.
FIGURE 18. Noisy 3D returns caused by (a) smoke, (b) unknown flying objects, (c) cloud, and (d) solar background noise or rain.
of data voids (refer to the “Occlusions/Shadows” section). one may simply investigate the distance between any pair
Instantaneous rainfall or snow storms during airborne li- of closest points or construct a voxel grid to assess the lo-
dar surveys generate random scattered returns, which are cal density [7] so as to look for the outliers. If end users
challenging to remove manually. Similar scenarios have are unable to have the original point clouds on hand, a
been reported in photon-counting lidar systems, such as despiking algorithm, such as the despike virtual defores-
the Leica SPL-100, where the system’s detector is sensitive tation algorithm [106] or spike-free algorithm [107], can
to solar background photons [7]. As a result, the solar back- be applied to the triangular area network (TIN) for re-
ground noise contaminates the collected point clouds. Un- moving the spiky features and negative blunders as well
like the previously mentioned unwanted returns caused by as smoothening the terrain.
clouds or smoke, random noisy returns are challenging to
remove manually. PITS AND SPIKES IN CANOPY HEIGHT MODELS
There are two ways to remove the outliers when The penetrating property of lasers and the ability of li-
one faces the preceding situations. Users can perform de- dar systems to probe and reconstruct environments in
noising on the point cloud, based on different 3D spa- 3D have allowed scientists to adopt airborne lidar data
tial filters [102]; clustering approaches, such as principal to retrieve/understand forest structures and properties.
component analysis [103] and density-based spatial clus- Although airborne lidar systems have been proved to be
tering of applications with noise [104]; or deep learning superior to optical imaging sensors in fine-scale forestry
approaches [105]. Unlike those point clouds collected by studies [108], [109], the collected datasets have certain
low-cost sensors, the density of outliers, in most cases, is imperfections disrupting the retrieval of forest inventory
significantly lower than the point cloud itself. Therefore, parameters. Aside from the occlusion effects caused by
foliage and upper canopies, as mentioned in the “Occlu-
sions/Shadows” section, the retrieved CHMs often suffer
from randomly distributed data pits and spikes [see Fig-
ure 19(a)]. These unnatural pits cause a high fluctuation
of elevations in the CHMs, which may influence the es-
timation of tree heights, crown boundaries, basal areas,
and stand volumes or induce errors in treetop detection
[107], [110], [111].
Due to the natural variation of canopies and irregu-
larities in tree heights, the emitted laser beams usually
produce a number of returns from the upper canopies,
(a) (b) tree trunks, branches, and foliage before reaching the
understory or even hitting the ground [see Figure 2(b)].
FIGURE 19. (a) A CHM suffering from data pits. (b) The CHM after Occasionally, laser beams may not interact with any
the data pits are removed. overstory and simply penetrate deeply into the crown,
INTRODUCTION
InSAR has become an increasingly popular tool for high-precision deformation
monitoring, due to its ability to detect small changes over time and provide a
unique view of Earth’s surface [1]. One critical challenge in InSAR is extracting
TABLE 1. NOMENCLATURE USED IN THIS ARTICLE. where E denotes spatial averaging, cnm is the coherence of
the nmth interferometric pair, v n = E {S 2n}, z n is the mul-
ABBREVIATION EXPANSION
tilook interferometric phase for the nth acquisition, and
CAESAR Component extraction and selection synthetic
aperture radar z nm = z n - z m is the multilook interferometric phase of ynm.
CNN Convolutional neural network Equation (4) requires reevaluating the phase estimation
ComSAR Compressed persistent scatterers and distributed in a stack of SAR images. The selection and weighting of the
scatterers interferometric SAR algorithm interferograms can affect the accuracy of the reconstructed
DL Deep learning
CRLB Cramér‒Rao lower bound
11°0′0″N 11°10′0″N
DS Distributed scatterer
N
EMI Eigendecomposition-based maximum likelihood
estimator of interferometric phase
ESA European Space Agency
ESR Equivalent single reference +30
PL Phase linking
PS Persistent scatterer
PSDS PSs and DSs
PSI PS interferometry
PU Phase unwrapping
RNN Recurrent NN
0 5 10 20
SAR Synthetic aperture radar km
t
with element c nm .
t ) + ln ^det (G) h- .
= argmin # tr (HG -1 H H C (9)
G, H
R = HGH H (8) m
/
n = 1 m= n+1
e
ng
t0 t cept, a vertical axis reference must
Ra
1 tN–1 ξ
be established, which enables height
s
os
measurements with respect to ter-
Cr
30 m
Sl rain elevation. The ground surface is
an z h
tR t often called the “zero-meter layer.”
imu 15 m
an Az x However, the ground phase contri-
ge
r 0m bution must be separated from the
O vegetation phase to prevent it from
influencing the 3D focusing.
y
(c) (d) These two points can be made by
removing the ground phase contri-
bution in the tomographic data [23].
FIGURE 3. A comparison of traditional SAR and TomoSAR acquisitions. (a) and (b): Traits of tra- The ground phases are determined
ditional SAR. (c) and (d): Traits of TomoSAR. (a) A SAR acquisition. (b) A SAR resolution cell. (c) A not only by terrain height zg but also
TomoSAR acquisition. (d) A TomoSAR resolution cell. The figure was adapted from [29, Fig. 2]. by the phase disturbances h deriv-
ing from the platform motion. In a
Consequently, TomoSAR processing enables us to ob- formula, { ground = k z z g + h, where k z = 4rb n /m sin iR n is the
tain the cross-range distribution of the SAR scene’s reflec- height-to-phase factor and i is the local incidence angle.
tivity at every range and azimuth location. By doing so, The multipolarimetric multibaseline covariance matrix
we can obtain a 3D image that provides comprehensive W can be approximated by retaining the first two terms
information on the reflectivity of a forest in three dimen- of the sum of the Kronecker products [33]. In a formula,
sions. This information can be used to derive valuable W . C G 7 R G + C V 7 R V , where R and C are referred to as
forest structure characteristics, such as height and bio- interferometric information and polarimetric information, re-
mass [23], [29], [30]. spectively, and G and V are associated with ground-only
and volume-only contributions, respectively. PL is a fun-
damental component in facilitating the retrieval of this
ground phase contribution from RG. Indeed, applying
z PL to forested areas allows for representing forest scatter-
N
30 m ing in terms of the “equivalent point target,” with well-
x
defined distances from the radar in different trajectories.
15 m This allows for simultaneous target and radar position
estimation, after which platform motion can be corrected
0m with subwavelength accuracy [20]. Figure 4 presents an
example of SAR and TomoSAR imaging in the Paracou
tropical forest site (French Guiana, South America).
0 0.5 1
PERSISTENT SCATTERERS AND DISTRIBUTED
y
(a) (b) SCATTERERS TECHNIQUE
The PSDS technique is an approach that leverages the phase
FIGURE 4. A comparison of traditional SAR and TomoSAR imaging. change over time of both PS and DS targets [1]. The tech-
(a) A traditional SAR image from the Paracou, French Guiana, forest nique involves two main steps: 1) PL and signal decorrela-
site. (b) TomoSAR layers, with each related to a certain height tion removal and 2) estimation of parameters of interest.
above the ground. The figure was adapted from [29, Fig. 4(b)]. PSDS refers to techniques that exploit the time series phase
1,000
The DS target is known for its low average temporal co-
herence, primarily due to geometrical and temporal decor- 500
relation phenomena [3]. As a result, this target often has
a low signal-to-noise ratio, making it challenging to work 0
with. However, enhancing the DS target’s signal-to-noise
ratio and treating it as a PS target is possible by identify- –500
ing pixels within a neighborhood that exhibit similar be-
–1,000
havior. These similar pixels are called SHPs. They can be 2011 2012 2013 2014 2015
identified using a two-sample Kolmogorov–Smirnov [11] Acquisition Date
or Baumgartner–Weiss–Schindler [40] test on the ampli- (c)
tude-based time series of the current pixel and its neigh-
bors within a specified window. The pixels with a similar FIGURE 5. Interferogram networks. (a) The single master network
cumulative probability distribution are grouped as “broth- in PSI processing. (b) The subset network consists of interfero-
ers,” resulting in a family of SHPs [see Figure 6(a)]. A DS grams with short spatial and temporal baselines in SBASs.
candidate is identified if it has a sufficient number of SHPs (c) The fully connected network in the PSDS technique.
20 0.8
40 0.6
0.4
60
0.2
80
0
20 40 60 80
(b)
2
4
6
8 (c)
10
2 4 6 8 10
(a)
(d)
FIGURE 7. The ComSAR algorithm. (a) A full coherence matrix in the PSDS technique. The data are divided into ministacks with 10 images
to improve the process efficiency. (b) The PL technique is employed to compress each ministack. This generates linked phases that enable
a coherent focus on the stack subset, resulting in a compressed image representing the first 10 images. This compression procedure is
repeated on the following 10 images, creating nine compressed images. These compressed images can be utilized to link prior ministacks
with new acquisitions and reconstruct the full phase time series without the need to recalculate everything. ComSAR can work with full and
compressed time series, but the (c) compressed version typically outperforms the (d) full time series version [16].
–π +π
(a) (b)
8
Frequency
2
(d) (e)
0
0 0.2 0.4 0.6 0.8 1
Coherence
(c)
FIGURE 8. Mexico interferograms. (a) A raw 670-day interferogram. (b) A compressed 670-day interferogram. (c) The coherence distribution.
(d) and (e) Zoomed-in versions of (a) and (b), respectively. The figure is adapted from [54, Fig. 3].
(a) (b)
(c) (d)
(e) (f)
–π 0 +π
FIGURE 9. A synthetic example of PL using DL. (a) A simulated deformed signal for interferograms, using the first acquisition as the refer-
ence image. (b) Interferograms after adding decorrelation noise. (c) The results of the MLE method using all the interferograms.
(d) The residuals of the MLE method [i.e., the difference between (a) and (c)]. (e) The results of the DL method using the U-Net model.
(f) The residuals of the DL method [i.e., the difference between (a) and (e)].
7,358
Change Detection
Superresolution/Pansharpening
Object Detection 76
Classification
Semantic Segmentation
191
Change Detection and Classification
352
1,386 Object Detection and Classification
Object Detection and Semantic Segmentation 338 102
260
Classification and Semantic Segmentation 341
Others 44 45 98
359
273 263
312
20
282
361 48 264
356 158 1
25997
50
47 281
208 287 288 151 54 49
256
40 160 46
325 373
Volume (GB)
199 332
292 132 320 235
331 236
107
35 42 262 371
13 36
255 41 311 303
53 127
216 349 217
73 114 37 5719
286 152 280 328 218
145
141 272370 172
224 244 251
22 77 294 119 327
220 223 229 111 270 368
109 314 206 261 38 90 175
219 269 8 283
147 333 335 243
27 266 61
12 274 18
169 309 239138
334 161 300 173
106 143
148 124 39 358 205253 174
82 339 189 231 32 201
142 291 308 301
128 364 5 221 242 135 17
366 55 168 91
30 78 240 101
234 200 336 171
153 296 207
290 3 33 15 321
69 131 6306 277 94 121 194
130 113 157 350 265
2 71 79 52 118 211
9 4 210 230 250
355 209 238 233 317
28 29 75 110 154 293 86 192 275 357
1 104 182 225 156 241 87 323 139
72 232 276
31 184 226 318 83 295 237 170 163
348
95
203 190 193 304
363 11 155 16 248 56
316 67 150 43
14 247
63 149 186 60
108 185 34 271 202 372
284
80 187 144 188 120
116
25 285 93
367
64 298 165
23 66 362
146 84 299 166 167
322
24 105 245
26 297 134
22
254 85
51
21 222 329
65 289
0
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
Publication Year
FIGURE 1. Distribution of remote sensing datasets over the years. The x-axis shows the publication year (the datasets are placed within the
region of their publication year with a small random offset to minimize visual overlap in the graph), while the y-axis represents the volume
of each dataset in gigabytes on a logarithmic scale. The circle radius indicates the dataset size in terms of the number of pixels. Colors
denote the type of task addressed by a dataset. Each circle is accompanied by an index, allowing for identification of the dataset in the
database (see Table 2), which provides further information.
TABLE 1. ALTHOUGH A THOROUGH ANALYSIS OF LIDAR DATASETS IS BEYOND THE SCOPE OF THIS SURVEY, WE DO PROVIDE AN
OVERVIEW OF SEVERAL EXAMPLE DATASETS. POINT CLOUD DATASETS ARE ANOTHER LARGE GROUP OF BENCHMARK DATA THAT ARE
WIDELY USED IN THE LITERATURE AND INDUSTRY. WITHIN EO THE MOST COMMON SOURCE FOR POINT CLOUD DATA ARE LIDAR SEN-
SORS THAT USE LIGHT IN THE FORM OF LASER PULSES TO MEASURE THE DISTANCE TO THE SURFACE. THE PRIMARY SOURCES ARE
AIRBORNE LASER SCANNING (ALS), TERRESTRIAL LASER SCANNING (TLS), AND MOBILE LASER SCANNING (MLS) DEVICES. OTHER
SOURCES OF POINT CLOUDS AND 3D DATA INCLUDE PHOTOGRAMMETRIC METHODS (STRUCTURE FROM MOTION, MULTI-VIEW STE-
REO, AND DENSE MATCHING APPROACHES) AND TOMOGRAPHIC SAR. AS 3D DATA TYPICALLY COME WITH FEATURES THAT ARE VERY
DIFFERENT FROM 2D IMAGE DATA, SUCH DATASETS ARE BEYOND THE SCOPE OF THIS ARTICLE. NEVERTHELESS, TABLE 1 PROVIDES A
SHORT LIST OF EXAMPLE LIDAR/POINT CLOUD DATASETS FOR INTERESTED READERS.
SEPTEMBER 2023
INDEX TASK PLATFORM SENSOR TYPE NAME DATE TIMESTAMPS IMAGES SIZE SIZE CLASSES (MB)
1 Change detection Aerial Multiple DFC21-MSD 2021 Multiple 2,250 4,000 36,000,000,000 15 325,000
10 Change detection Satellite Multiple DFC09 2009 Multiple Two 98 — — —
11 Change detection Satellite Multispectral OneraCD 2018 Multiple 24 600 8,640,000 Two 489
15 Change detection Satellite Optical LEVIR-CD 2020 Single 637 1,024 667,942,912 One 2,700
21 Classification Aerial Hyperspectral Indian Pines 2000 Single One 145 21,025 16 Six
22 Classification Aerial Hyperspectral Salinas 2000 Single One 365 111,104 16 27
24 Classification Aerial Hyperspectral Kennedy Space Center 2005 Single One 550 311,100 13 57
34 Classification Drone Optical AIDER 2019 Single 2,645 240 152,352,000 Four 275
40 Classification Satellite Multiple BigEarthNet-MM 2019 Single 590,326 × 12 120 1.02E + 11 19 121,000
52 Classification Satellite Multispectral EuroSAT 2018 Single 27,000 × 13 64 1,437,696,000 10 1,920
63 Classification Satellite Optical University of California, 2010 Single 2,100 256 137,625,600 21 317
Merced
69 Classification Satellite Optical AID 2017 Single 10,000 600 3,600,000,000 30 2,440
76 Classification Satellite Optical FMoW 2018 Single 523,846 — 1.08E + 12 63 3,500,000
(Continued)
67
68
TABLE 2. ALTHOUGH THE COMPLETE COLLECTION OF 380 DATASETS IS TOO EXTENSIVE TO BE INCLUDED IN THE PRINT VERSION OF THIS ARTICLE, IT CAN BE FOUND AT [85]. HERE WE
INCLUDE A PORTION OF THIS LIST, WHICH CONTAINS ALL THE DATASETS MENTIONED IN THE ARTICLE, ALONG WITH THEIR INDEX IN THE COMPLETE DATABASE. (Continued)
SEPTEMBER 2023
form their own category. Examples in the “Others” category Figure 2 provides further insights into the size distri-
are datasets on cloud removal, visual question answering, bution of available datasets in terms of the 1) number
and parameter-estimation tasks such as hurricane wind of pixels and 2) data volume in gigabytes for typical
speed prediction, satellite pose estimation, and vegetation remote sensing tasks. Illustrating these two different
phenological change monitoring. The dashed line (“—”) il- aspects allows a deeper understanding of the nature
lustrates an exponential growth of benchmark datasets cre-
ated by and for the EO community.
This map on the evolution of remote sensing datasets How to Measure the Size of a Dataset
offers several interesting insights: In this article, we look at the size of datasets from the following two perspectives:
◗◗ The beginnings: In addition to the first IEEE GRSS Data Fu- 1) Size: data volume in terms of the number of spatial pixels. We count the num-
sion Contest in 2006 (Table 2; number 316), there are a few ber of pixels in the highest-available image resolution while ignoring mul-
other pioneering datasets that have fostered ML research tiband, multichannel, and multisensor data. In other words, pixels are only
applied to remote sensing data in its early stages, e.g., counted once in the spatial coverage provided by the dataset.
•• Hyperspectral datasets [(Indian Pines, Salinas Valley, 2) Volume: data volume in terms of storage. The amount of disk space required
and Kennedy Space Center) (Table 2; 21, 22, and 24)]: for a dataset is a proxy for image resolution and the provided modalities (e.g.,
Published before 2005, these datasets triggered the multiple bands and sensor types).
ML era in remote sensing. Covering a very small area Figure S1 highlights the different factors that affect the volume and size of a
on the ground and having a very small number of dataset: the number of bits per pixel (radiometric resolution), number of spectral
pixels, such datasets are not suitable for training DL bands (spectral resolution; i.e., red, green, blue; multispectral; or hyperspectral),
models (or have to be used with excessive caution). number of images during a specific time period (temporal resolution), and num-
On the other hand, due to their rich hyperspectral ber of pixels per unit area (spatial resolution). As mentioned previously, the size is
information, they are still being used for tasks such directly related to the unique number of ground-projected resolution cells. A
as dimensionality reduction and feature extraction. larger dataset in terms of size corresponds to images with higher resolutions or
•• The University of California, Merced dataset (Table 2; broader coverage.
63) [7]: Published in 2010, it was the first dataset
dedicated to scene classification.
•• ISPRS Potsdam/Vaihingen dataset (Table 2; 219 and 220)
[8]: Published in 2012, it was initially intended to bench- Satellites
mark semantic segmentation approaches tailored to
aerial imagery. Later, it was also used for other tasks, e.g.,
single-image height reconstruction (e.g., in [9]).
•• SZTAKI-AirChange dataset (Table 2; 105) [10]: Pub-
Airplanes t1 t2 ... tn
lished in 2011, it was one of the earliest datasets de-
signed for object detection.
All of those pioneering datasets have seen massive use in
Drones
the early ML-oriented EO literature. It is interesting to note Time Series
that pansharpening, scene classification, semantic segmen- Multispectral Hyper-
tation, and object detection were the first topics in remote Width Spectral
sensing to be addressed using ML-based methodologies. t
igh Volume
◗◗ The DL boom: As discussed by several review articles on He
DL and AI applied to EO [11], [12], [13], the year 2015 Resolution
marked the beginning of the DL boom in the EO com-
munity. This is well reflected by a significantly rising
number of datasets published from that year onward. It is Images Projected
furthermore confirmed by the fact that the dataset sizes, onto Ground Surface
both in terms of spatial pixels and data volume, started
to increase significantly from approximately that time.
◗◗ The diversity of tasks: From the early days to the present,
ML-oriented EO datasets have been designed for a mul-
titude of different tasks. The historical evolution depict-
ed in Figure 1 further shows that object detection and
semantic segmentation are the most popular tasks, with FIGURE S1. A schematic illustration of the proposed size measure
a significant increase of datasets dedicated to minor- used to characterize datasets i.e., pixels are only counted once in
ity categories (denoted as “Others”) from roughly 2019 the spatial coverage provided by a dataset. For a more detailed
to the present. This indicates that the rise of DL in EO definition, see “How to Measure the Size of a Dataset.”
broadens the overall application scope of the discipline.
0.16% 1.12%
1.92% 14.53%
21.07%
8.52%
22.81%
0.65%
29.94% 13.87%
57.31%
20.19% 0.24%
0.48%
0.34% 0.14%
2.52%
(a) (b)
FIGURE 2. Distribution of EO dataset sizes over typical remote sensing tasks, expressed in (a) volume and (b) size, as defined in Figure S1
and “How to Measure the Size of a Dataset.” In (a), with 30%, object detection is the predominant task, followed by semantic segmentation.
In (b), in contrast to Figure 2(a), semantic segmentation is the prevailing task, illustrating that corresponding datasets involve more complex
scenarios such as leveraging multiple sensors or spectral bands. Object detection and semantic segmentation are the dominant image
analysis tasks in ML-centered EO.
4.63% HS
25% 7.41%
SAR
17.9% 20.99%
6.48% 15.43%
8.33%
Aerial Multiple Multiple Multiple
MS
Drone
FIGURE 3. A distribution of available EO datasets over different platforms, sensor types, and number of acquisition times. Single-image red,
green, blue (RGB) images acquired by satellites are clearly the dominating modality. MS: multispectral; HS: hyperspectral.
Class/SemSeg
SemSeg
SemSeg
eg
Others
SemS
Class
OD
OD
ss
OD hers
ers
/Cla
OD
/Cla
Oth
Ot
Se
OD
ss
mS
OD
Clalass
eg
/C
ss
Ot
OD
he
rs
Se
m
Se
Multiple
g
Cl
ss
MS
OD as
HS
SAR
/C O s la
C
Se Oth las D
RGB
m er s
Se s
CD g
B
/C CD CD
RG
las
M
s
ult
eg
ipl
mS
e
Se
Cla g
ss RG ers Se
B OthD/Sem
OD
Cla OD
ss/S C
emS Seg
eg
le Sem
OD ltip
Mu SR
Dron
OD/S OD
emS
eg
e
M
ul
MS
tip
Others
l
le
MS ria SemSeg
Ae
SR
HS
Others
OD
Class
CD
SemSeg
SemSeg
Others
CD SAR
s OD
Clas
llite
le
Sate
ltip
Mu
Seg
em
ss/S Cla
ss
Cla
ers
Oth
Se
eg mS
mS eg
Se
RGB
CD
SR
Ot
he
rs
ss
Cla
OD
Seg
Class/Sem
FIGURE 4. A distribution of tasks between sensors and platforms. Platforms are in the inner ring, sensors are distributed in the middle ring,
and the outer ring shows different tasks per sensor. OD: object detection; CD: change detection; SemSeg: semantic segmentation; SR: super-
resolution; class: classification; MS: multispectral; HS: hyperspectral.
HS
CD CD/Class Class Class/SemSeg OD OD/ClassOD/SemSeg Others SR SemSeg CD CD/Class Class Class/SemSeg OD OD/ClassOD/SemSeg Others SR SemSeg
2019 2020
MS Multiple RGB SAR
HS
CD CD/Class Class Class/SemSeg OD OD/ClassOD/SemSeg Others SR SemSeg CD CD/Class Class Class/SemSeg OD OD/ClassOD/SemSeg Others SR SemSeg
2021 2022
MS Multiple RGB SAR
HS
CD CD/Class Class Class/SemSeg OD OD/ClassOD/SemSeg Others SR SemSeg CD CD/Class Class Class/SemSeg OD OD/ClassOD/SemSeg Others SR SemSeg
Tasks Tasks
(b)
FIGURE 5. Different tasks [CD, classification (class), semantic segmentation (SemSeg), object detection (OD), superresolution (SR), and
pansharpening] and task combinations (denoted by “/”) make use of very different platforms and sensors. (a) Combinations of different
platforms, sensors, and tasks accumulated over the years. (b) Combinations of different platforms, sensors, and tasks for different years.
MS: multispectral; HS: hyperspectral.
80
Geographic Diversity of Datasets
70
South America
4%
60
North
50 America
18% Multiple
Other Global Locations
40 Europe 41% 12%
25%
21%
30 Asia
Synthetic
10%
20 Australia 4%
1% Africa
10 5%
0
Brazil
Australia
Sweden
Global
United States
Multiple Locations
China
Synthetic
Germany
France
Europe
Italy
Canada
Asia
Africa
Rwanda
Greenland
Switzerland
India
Peru
South Africa
South America
Spain
Netherlands
Japan
Greece
Alaska
Belgium
Gulf of Mexico
United Kingdom
UAE
Sudan
Austria
Benin
Slovenia
Scotland
Denmark
Poland
Ecuador
Norway
North America
New Zealand
Estonia
Finland
Mexico
Mali
Malawi
Kenya
Ghana
Hungary
West Africa
FIGURE 6. A geographic distribution of EO benchmark datasets (which provided clear location information). UAE: United Arab Emirates.
108 CD 108
97 97
Superresolution/Pansharpening
87 87
78 Object Detection 78
69 Classification 69
61 Semantic Segmentation 61
54 Change Detection and Classification 54
48 48
Object Detection and Classification
42 42
Number of Classes
1
6
28
109
357
1,000
2,000
6,000
14,000
29,000
59,000
112,000
206,000
365,000
627,000
1,048,000
1,708,000
2,719,000
4,242,000
6,492,000
FIGURE 7. The number of classes provided by the reference data of a given dataset not only varies for different tasks (e.g., object detection
is dominated by datasets with only a single class) but also with (a) publication year and (b) the number of images of a dataset.
Volume (MB)
ActiveFireL8 South Sudan Crop Type
AIRS SEN1-2
(b)
(c)
(d)
Volume (MB)
FMoW
So2Sat LCZ42
DND-SB
Rwanda Crop Type
BrazilDAM
Other
ZueriCrop
SAR OVERHEAD
Kag. Satellite
-based
NWPUBCS
BCSS
Hurricane
RSSCN7
WHU-RS19
Sea
UC MNIST
IceDamage
-Merced
Bijie Dataset
SanFranBay_Ships_Planet
Landslide
SIRI-WHU
USTC_SmokeRS
RSI-
SAT-4
SAT-6
Airbus Wind PatternNet
Turbines
MLRSNet Patches
BigEarthNet-MM EuroSAT
RSI-CB256
MASATI-v2
AID
LandCoverNet Au Kag. Find a Car Park
SeaIceClassification
Kag. Planet Forest CV4A Kenya
DeepGlobe -Road
VEDAI
SloveniaLandCover ADVANCE
Kaggle Cloud Detection
MultiScene
LandCoverNet Af AiRound CV-BrCT
LandCoverNet SA S2-Agri
xView ICONES-HSI
ARGO
AnthroProtect
LandCoverNet Eu
LandCoverNet As DENETHOR
LandCoverNet NA
FIGURE 11. Relative volume distribution among datasets addressing scene classification.
FIGURE 12. Sample image patches of all 10 classes covered in the EuroSAT dataset.
red, green, blue (RGB). The FMoW-Full is in TIFF format 200 GB and includes all multispectral data converted to
and contains four- and eight-band multispectral imag- RGB in JPEG format. Examples of the classes in the da-
ery with a high spatial resolution resulting in 3.5 TB of taset include flooded roads, military facilities, airstrips,
data, while the FMoW-RGB has a much smaller size of oil and gas facilities, surface mines, tunnel openings,
shipyards, ponds, and towers (see
Figure 14 for examples). The FMoW
dataset has a number of important
characteristics, such as global di-
versity, multiple images per scene
captured at different times, multi-
spec t ral imager y, and metadata
linked to each image.
Airport Airport Hangar Airport Terminal Amusement Park Aquaculture Archaeological Site Barn
Border Checkpoint Burial Site Car Dealership Construction Site Crop Field Dam Debris or Rubble
Educational Factory
Electric Substation Fire Station Flooded Road Fountain Gas Station
Institution or Power Plant
Parking lot
Lighthouse Military Facility Nuclear Power Plant Office Building Oil or Gas Facility Park
or Garage
Recreational
Place of Worship Police Station Port Prison Racetrack Railway Bridge
Facility
Residential Unit Residential Unit Road Bridge Runway Shipyard Shopping Mall Smokestack
Solar Farm Space Facility Stadium Storage Tank Surface Mine Swimming Pool Tollbooth
Tower Tunnel Opening Waste Disposal Water Treatment Wind Farm Zoo
Facility
FIGURE 14. Sample image patches for several classes from the FMoW dataset.
WHUS2-CD+
WHU Building Kag. Airbus Ship Detection
RampBuilding
DOTA v2.0
TopoBoundary
INRIA
BIRDSAI
Other
DOTA v1.5
AirbusTree
Urban OilAMD
SZTAKI
SAR_VD
NWPU
Storage
EoCCrownAirChange
-VHR10
NASAdebris
AIR-SARShip
ALCD
Iceberg
RSOD
WAMI Detection
-2
Detection
DIRSIG
PlanesNet
SAR Ship
MTARSI
HRSID
SRSDD Detection
-v1.0
SaRNet
UAVOD-10
TaS LEVIR-Ship
COWC DOTA v1.0 SSDD
SIMD
DSSDD
ITCVD
Bridges
built-structure
CARPK -count
VisDrone NYCPlanimetric
AU-AIR
DSTL3B SeabirdsDetection
MSAR
Oil Storage Tanks
ArcticSeals Satellite Pool Detection
SpaceNet-4 SWIM ShipRSImageNet
Agriculture-Vision
DroneCrowd PKLot
AFO
TGRS-HRRSD EU Flood
VISO DSTL16BRoads
Massachusetts
NEON Tree HRSC2016
Dataset
LS-SSDD
RarePlanes
Flying Airplanes
FIGURE 15. Relative volume distribution among the datasets addressing object detection.
state rasters, and the dataset is intended to support the The DOTA dataset is available in three different versions:
development and evaluation of ML approaches for de- DOTA-v1.0, DOTA-v1.5, and DOTA-v2.0. The image size
tecting “dark vessels” not visible to conventional moni- in the initial version ranges from 800 × 800 pixels to
toring systems. 4,000 × 4,000 pixels, with 188,282 object instances with
◗◗ DOTA [54] is one of the most popular and largest ob- various scales and angular orientations and a total of 15
ject detection datasets (Table 2; 109 and 114) in terms object categories. DOTA-v1.5 adds various small objects
of labeled object instances. It includes 2,806 images ac- (fewer than 10 pixels) and a new container crane catego-
quired from Google Earth (GE) and the China Center for ry with 402,089 instances, whereas DOTA-v2.0 adds two
Resources Satellite Data and Application [31], [32], [33]. categories, airport and helipad, with 11,268 images and
1,793,658 instances, respectively.
Some image samples of the DOTA
VH_dB VV_dB Mask dataset are presented in Figure 17.
CD
CD in remote sensing aims to iden-
tify temporal changes by analyzing
multitemporal satellite images of the
same location. CD is a popular task
in EO as it fosters monitoring envi-
ronmental changes through artifi-
(a) cial or natural phenomena. Figure 2
WindDirection WindQuality WindSpeed shows that the number of dedicated
CD datasets is small compared to
other applications. Figure 18 shows
that the available data are dominated
by the DFC20 dataset (Track MSD),
which focuses on semantic CD, fol-
lowed by xView2, which tackles
building damage assessment in the
context of natural disasters. We chose
(b) LEVIR-CD as a recent dataset exam-
ple and the Onera Satellite Change
FIGURE 16. (a) An example image stack of dual-polarimetric SAR images and a water mask and Detection dataset as one of the first
(b) several wind properties from the xView3-SAR dataset. large-scale datasets.
Volume (MB)
DFC21-MSD
Synthetic and Real
LEVIR-CD
Other Second
HRSCD DFC09
SYSU-CD DSIFN
S2Looking
MtS-WH
S2MTcityPair
OneraCD
MUNO21
AIST Building CD
AICD
xView2
FIGURE 18. Relative volume distribution among the datasets addressing change detection.
FIGURE 20. Examples of annotated samples from the Onera Satellite Change Detection dataset.
FIGURE 21. Sample images from the Proba-V dataset. Each sample consists of one high-resolution (HR) and several low-resolution (LR)
images, each with a quality map (QM) showing which pixels are concealed (e.g., through clouds and so on).
(107 GB), as Figure 1 shows (Table 2; 325). This dataset WORKING WITH REMOTE SENSING DATASETS
contains high-resolution images from Airbus SPOT 6 This section provides guidance on how to leverage avail-
and 7 along with 16 temporally matched low-resolu- able datasets to their full potential. Two of the main
tion images from Sentinel-2 satellites. The high-resolu- difficulties caused by information asymmetry (i.e., the in-
tion images are over five spectral bands: the panchro- formation imbalance between the data providers and the
matic band at a 1.5-m pixel resolution and RGB, and data users) [5] have found suitable datasets and easy pro-
NIR bands at 6 m per pixel. The low-resolution ranges totyping of ML approaches using such datasets. Here we
from 10 m per pixel to 60 m per pixel (Figure 23). discuss resources to gain an overview of existing datasets
In total, the dataset covers an area of approximately and download actual data, but we also provide examples
10,000 km 2 and attempts to represent all types of land of EO-oriented ML programming libraries.
use across the world. Notably, the dataset contains
nonsettlement and underrepresented locations such DATA AVAILABILITY
as illegal mining sites, settlements of persons at risk, Data availability covers two distinct aspects: on the one
and so on. hand, access to the curated benchmark datasets, i.e., how
such datasets are made available to
the public. This section prov ides
several e xamples of the most
common data sources. On the other
hand, the actual noncurated measure-
ments as acquired by the different
sensors such as satellites, planes, and
UAVs are often available too. Many
data providers offer either their com-
plete database for public access (e.g.,
the European Copernicus Open Ac-
cess Hub [60]) or at least portions
of their image acquisitions (e.g., via
Open Data Programs such as those
from Maxar [61] and Capella Space
FIGURE 23. Sample images from the WorldStrat dataset. (Source: [37].) [62] or through scientific proposals
TABLE S1. AN OVERVIEW OF THE IEEE GRSS DATA FUSION CONTESTS FROM 2006 TO 2023.
and interactive manner (see Figure 25 for an example of data. As PyTorch [39] and TensorFlow [40] (note that
the user interface). For more information, see “The IEEE Keras is now a part of TensorFlow) are the most widely
Geoscience and Remote Sensing Society Earth Observa- used DL frameworks, we focus mainly on the introduc-
tion Database.” tion for the existing libraries, using these two DL frame-
works as the back end.
EO-ORIENTED ML LIBRARIES 1) TorchGeo [72] is an open source PyTorch-based li-
Most of the existing ML libraries are developed for clas- brary, which provides datasets, samplers, transforms,
sic computer vision tasks, where the input image is usu- and pretrained models specific to geospatial data [41].
ally single-channel (grayscale) or with RGB bands. EO The main goal of this library is to simplify the pro-
datasets are often of large volumes with highly diverse cess of interacting with complex geospatial data and
data types, different numbers of spectral bands, and make it easier for researchers to train ML models for
spatial resolutions, as illustrated in Figure S1. The code EO tasks. Figure 26 provides an example of sampling
base for processing such data samples is often highly pixel-aligned patch data from heterogeneous geo-
complex and difficult to maintain. One approach to spatial data layers using the TorchGeo package. As
increase readability, reusability, and maintainability is different layers usually have different coordinate
to modularize the code and encapsulate different tasks systems and spatial resolutions, patches sampled
by decoupling processing the dataset (data loaders, vi- from these layers in the same area may not be pixel
sualization, preprocessing, and so on) and applying ML aligned. Therefore, in a practical application sce-
models (training, prediction, model selection, evalua- nario, researchers need to conduct a series of pre-
tion, and so forth). A major challenge in training ad- processing operations such as reprojecting and resa-
vanced ML models for EO tasks is the implementation mpling of the geospatial data before training ML
of an easy-to-use, yet efficient data loader explicitly de- models, which is time consuming and laborious.
signed for geoscience data that loads and preprocesses To address this challenge, TorchGeo provides data
a complex dataset and produces an iterable list of data loaders tailored for geospatial data, which support
samples in a customizable way. transparently loading data from heterogeneous geo-
This section introduces several well-known packages spatial data layers with relatively simple code, per
designed explicitly for geoscience and remote sensing the following example:
A more detailed introduction about the supported ning the model. It supports loading raster data from
geospatial datasets in TorchGeo can be found in [41]. local files or cloud services. Necessary preprocessing,
2) RasterVision [73] is an open source Python framework like reprojecting and resampling, is also conducted auto-
that aims to simplify the procedure for building DL- matically. Keras Spatial supports sample augmentation
based computer vision models on satellite, aerial, and
other types of geospatial data (including oblique drone
Landsat 8 Scene
imagery). It enables users to efficiently construct a DL EPSG: 32617
A
pipeline, including training data preparation, model C
training, model evaluation, and model deployment,
without any expert knowledge of ML. Specifically,
RasterVision supports chip classification, object detec-
tion, and semantic segmentation with the PyTorch back
end on both CPUs and GPUs, with built-in support D
for running in the cloud using AWS. The framework
is extensible to new data sources, tasks (e.g., instance
segmentation), back ends (e.g., Detectron2), and cloud
providers. Figure 27 provides the pipeline of the Raster- B Cropland Data Layer
Vision package. A more comprehensive tutorial for this EPSG: 5072
Deployment
Inputs RasterVision Pipeline
Batch
Images
Predictions
ANALYZE CHIP TRAIN PREDICT EVAL BUNDLE
Live
Labels
Predictions
Dataset Training Training Validation Evaluation Model
Metrics Chips Model Predictions Metrics Bundle
Custom
AOI
Integrations
FIGURE 27. An illustration of the pipeline from the RasterVision package. AOI: area of interest. (Source: https://github.com/azavea/raster-vision.)
Lidar
SAR
Multi-/Hyperspectral Unsupervised CD
Different Geolocations
t1 tn
Object Detection
Annotations
Semantic Segmentation
Spring Summer Autumn Winter Instance Segmentation
Different/Multiple Times of the Year
Supervised CD
FIGURE 28. An illustration that shows the authors’ view of the paramount properties that an ideal benchmark dataset needs to satisfy,
including the type of tasks, sensors, temporal constraints, and geolocalization.
downloaded by everyone who wanted to work with them. boom has not only led to a rise in dataset numbers but also
With the hundreds of datasets available today and many a large increase in size (as in spatial coverage and resolution
more published every year, this cannot be maintained. but also with respect to multimodal and multitemporal im-
Concepts such as findability, accessibility, interoperabili- agery) and a diversity of application tasks. Furthermore, this
ty, and reuse (FAIR) (see, for example, [45]) were proposed development has led to the implementation of dedicated
years ago and are still of high relevance. Data catalogs software packages and meta databases that help interested
such as EOD (see the “Working With Remote Sensing Da- users develop solutions for their applications. Eventually,
tasets” section) are a first step toward structure datasets we drew the conclusion that one of the critical challenges
that are scattered among different data hosts. ARD (see, in dataset design for EO tasks is the strong heterogeneity of
e.g., [46]), for example, in the form of data cubes [47], possible sensors, data, and applications, which has led to
and efforts to homogenize meta-information, e.g., in the a jungle of narrow-focused datasets. Although one of the
form of datasheets [48], will continue to evolve into new trends in DL is certainly the consideration of smaller, well-
standardized data formats. The trend of datasets growing curated, and task-specific datasets, another direction is the
in terms of size and volume (see the “Evolution of EO- creation of a generic, nearly sensor- and task-agnostic data-
Oriented ML Datasets” section) as well as the need for base similar to the well-known ImageNet dataset used in
global data products will soon put a stop to the current Computer Vision. Such a generic dataset will be especially
practice of downloading datasets and processing them lo- valuable in the pretraining of large high-capacity models
cally. Working with data on distributed cloud services will with worldwide applicability.
create new requirements regarding data formats but also
lead to new standards and best practices for processing. AUTHOR INFORMATION
Finally, the goal of any ML-centered dataset is to train an MICHAEL SCHMITT (michael.schmitt@unibw.de) received
ML model. These models should be treated similar to the his Dipl.-Ing. degree in geodesy and geoinformation, his
data they originated from, i.e., they should follow stan- Dr.-Ing. Degree in remote sensing, and his habilitation
dardized data formats and FAIR principles. in data fusion from the Technical University of Munich
(TUM), Germany, in 2009, 2014, and 2018, respectively.
SUMMARY AND CONCLUSION Since 2021, he has been a full professor for Earth obser-
This article discussed the relevance of ML-oriented datas- vation at the Department of Aerospace Engineering of the
ets in the technical disciplines of EO and remote sensing. University of the Bundeswehr Munich, 85577 Neubiberg,
An analysis of historical developments shows that the DL Germany. Before that, he was a professor of applied geodesy
SSL4EO-S12
A large-scale multimodal, multitemporal dataset for self-supervised learning in Earth observation
TRANSFER LEARNING
The pretrained models were transferred to various down-
stream tasks. For scene classification, we evaluated EuroSAT
[19] (single-label land cover classification), BigEarthNet [13]
(multilabel land cover classification), and So2Sat-LCZ42
FIGURE 2. Geographical distribution of SSL4EO-S12 dataset. [20] (local climate zone classification, culture-10 version).
BENCHMARK RESULTS
CLASSIFICATION
COMPARISON OF SELF-SUPERVISED
LEARNING METHODS
We first benchmarked different SSL methods through
linear probing on EuroSAT, BigEarthNet, and So2Sat-
LCZ42. As detailed in Table 3, all methods outperformed
(a)
random initialization by a substantial margin. As ex-
pected, linear probing on BigEarthNet with all labels
performs worse than fully supervised training. Prom-
isingly, the gap stays below 5%. On small datasets like
BigEarthNet with 10% labels or EuroSAT, linear probing
provides results comparable to supervised training with-
in approximately ±1%. The trends are slightly different
for So2Sat-LCZ42, where the training and testing sets are
built upon different cities with a challenging geographi-
cal split. Because of this significant domain shift, adding
labeled training data does not necessarily improve the
testing performance. In fact, fitting the training data dis-
tribution does not guarantee out-of-distribution gener-
(b)
alization. Nevertheless, the best pretrained models with
linear probing beat the supervised baseline by at least 1% FIGURE 3. Image patches without (a) and with (b) overlap filter-
up to about 4%. ing in Tokyo metropolitan area. We plotted red circles of radius
Furthermore, we benchmarked fine-tuning results in 1.32 km (132 pixels) for better visualization.
Table 4. All self-supervised methods outperform super-
vised learning with a margin from 1% to 6%. Top SSL
models score 99.1% on EuroSAT (MoCo/DINO) and over TABLE 2. 100 EPOCH TRAINING TIME OF THE STUDIED
SSL METHODS.
90% on BigEarthNet (MoCo/DINO). Comparing linear
probing and fine-tuning results, one interesting phe- MOCO DINO MAE DATA2VEC
nomenon shows up: in linear probing, contrastive meth- ResNet50 18 h 25 h — —
ods (MoCo and DINO) consistently score better than ViT-S/16 24 h 25 h 7h 14 h
their image-masking (MAE and data2vec) counterparts.
TABLE 3. LINEAR PROBING RESULTS FOR EUROSAT, BIGEARTHNET (BE), AND SO2SAT-LCZ42 (10% AND 100% LABELS).
PRETRAIN DATASET EUROSAT BE-10% BE-100% BE Label Percentage 1% 10% 50% 100%
ImageNet (RGB) [4] 86.4 70.5 71.8 Linear 75.9 82.1 82.7 84.2
ViT-S/16 RN50
SeCo (RGB) [4] 89.5 74.5 76.3 Fine-Tune 80.3 86.2 87.7 91.8
SEN12MS (RGB) 94.9 76.6 79.6 Supervise 75.7 83.4 85.2 88.7
SSL4EO-S12 (RGB) 96.6 80.1 82.3 Linear 78.2 82.3 83 83.1
SeCo* (all bands) 89.2 73.7 76.6 Fine-Tune 78.9 86.1 88.2 89.9
SEN12MS (all bands) 95.5 79.6 82.1 Supervise 69.3 81.3 84.9 87.4
BigEarthNet (all bands) 94.4 80.6 83.9
SSL4EO-S12 (all bands) 98.0 82.1 84.2
FIGURE 4. BigEarthNet (BE) performance depending on the
Italics means cited from the literature. Bold values indicate best per column performance.
*SeCo is meant to have 200,000 geographical patches as described in the article, but the number of labels available to train downstream task. We report
available data at https://github.com/ServiceNow/seasonal-contrast has only about linear probing and fine-tuning results with ResNet50 and ViT-S/16
160,000 patches, which may affect our reproduced performance.
encoders pretrained using MoCo-v2. Bold values indicate best per
column performance.
bel splits. While single-modality pretraining already works Augment 77.6 96.2
well for both Sentinel-2 and Sentinel-1 data, pretraining Bold values indicate best per column performance.
CONCLUSION
In this work, we present SSL4EO-S12, a large-scale mul-
timodal, multitemporal unlabeled dataset for SSL in EO.
An extensive benchmark on various SSL methods and
RS applications proves the promising benefits of the pro-
posed dataset.
(a) SSL4EO-S12 has some limitations: 1) there’s little cover-
age of polar regions, 2) geographical bias exists due to cloud
filtering, 3) it is not strictly free of geospatial overlap, and
4) medium-resolution radar and multispectral images are a
limited subset of EO data. Despite these limitations, we be-
lieve SSL4EO-S12 renders a valuable basis to advance self-
supervised pretraining and large-scale data mining in RS.
ACKNOWLEDGMENT
This work is jointly supported by the Helmholtz Association
through the Framework of Helmholtz Artificial Intelligence
(b) (Grant ZT-I-PF-5-01)–Local Unit “Munich Unit @Aeronau-
tics, Space and Transport (MASTr)” and Helmholtz Excel-
FIGURE 5. t-SNE visualization of EuroSAT image representations. lent Professorship “Data Science in Earth Observation–Big
One color represents one class. (a) Random-encoded features; Data Fusion for Urban Research”(Grant W2-W3-100); by the
(b) SSL-encoded features. SSL-encoded features are well clustered German Federal Ministry of Education and Research in the
even without label information. framework of the International Future Artificial I ntelligence
CaBuAr
California burned areas dataset for delineation
California 42
42
Satellite Tiles
40
40
Latitude (°)
Latitude (°)
38 38
36 36
34 34
32
–124 –122 –120 –118 –116 –114 –124 –122 –120 –118 –116 –114
Longitude (°) Longitude (°)
(a) (b)
Map tiles by Stamen Design, CC BY 3.0 -- Map data (C) OpenStreetMap contributors
FIGURE 1. (a) The satellite tiles coverage: the California administrative boundaries (red) versus the satellite tiles of the proposed dataset
(blue). (b) The location of the wildfires (red) inside the California boundaries (blue).
TASKS
The proposed dataset can be used
as a benchmark for different tasks FIGURE 3. A sample of postfire RGBs and masks with the associated comments.
1 1
Burned Pixels per Image (%)
0.8 0.8
0.6 0.6
0.4 0.4
0.2 0.2
0 0
0 1 2 3 4 0 1 2 3 4
Fold Fold
(a) (b)
FIGURE 4. The burned pixels percentage per image per fold: (a) setting 1 and (b) settings 2 and 3.
©IGARSS 2023
sessions” were introduced to get more of
the community involved. Also, the struc-
ture of the oral session was changed to FIGURE 3. Technical committee cochairs Dr. Rashmi Shah (right) and Dr. David Kunkee
include more discussion among the com- (left) presenting the technical program.
munity members: presentations were
planned to be 12 min long, and within a session there was After Dr. Coughlin’s presentation, GRSS President Dr.
only at the end of all presentations a 15-min slot in which Mariko S. Burgin gave a warm welcoming address to all at-
to ask questions. tendees and reported on the activities of the GRSS (see
Dr. Rashmi recommended the TIE events and YP events. Figure 5). The GRSS is one of 39 IEEE Societies, and it is a truly
She also spoke about the 13 tutorials with more than 300 global community. It has nearly 5,000 members in 144 coun-
participants that were held in the run-up to the conference tries and is organized into 128 Chapters with 12 ambassadors
and were a great success. all over the world. The GRSS is governed by the GRSS AdCom.
After Dr. Rashmi spoke, 2023 IEEE President-Elect
Dr. Tom Coughlin’s introduction followed (see Figure 4).
He sees IEEE as a resource for technology decisions.
As technology of all sorts drives the world’s economy,
this is something that is very important to be aware of.
IEEE is the largest technical professional organization
in the world, its members are involved in all aspects
of technology creation and use, its research powers
patents, and it creates the world’s technical standards.
©IGARSS 2023
©IGARSS 2023
FIGURE 7. IEEE Fellow Award recipient Prof. Jonathan Li (middle)
with Dr. Mariko Burgin (right) and IEEE President-Elect Dr. Tom
Coughlin (left). FIGURE 8. IEEE Fellow Award recipient Prof. Gabriele Moser.
the University of Colorado Boulder in 1997 and also holds Engineering and the Engineering Institute of Canada and is
a B.S. degree from the Rensselaer Polytechnic Institute and the recipient of more than 20 prestigious awards.
an M.S. degree from Stanford University. He is a fellow The third Fellow recognition was received by Prof. Ga-
of the Institute of Navigation. From 2018 to 2022 he was briele Moser with the following citation (see Figure 8): “For
editor-in-chief of GRSM. contributions to pattern recognition in remote sensing.”
The second Fellow recognition went to Prof. Jonathan Li Prof. Gabriele Moser is a full professor of telecommu-
with the citation (see Figure 7) “For contribution to point nications at the University of Genoa. His research activ-
cloud analytics in lidar remote sensing.” ity is focused on pattern recognition and image processing
Prof. Jonathan Li received his Ph.D. degree in geomatics methodologies for remote sensing and energy applica-
engineering from the University of Cape Town, South Africa, tions. He served as chair of the GRSS Image Analysis and
in 2000. Prof. Jonathan Li is currently a professor of geomat- Data Fusion Technical Committee (IADF TC) from 2013 to
ics and systems design engineering 2015 and as IADF TC cochair from
with the University of Waterloo, 2015 to 2017. He was publication
Canada. His main research interests cochair of IGARSS 2015, technical
include artificial intelligence (AI)- program cochair of the GRSS Earth-
based 3D geospatial information Vision workshop at the 2015 IEEE
extraction from Earth observation Conference on Computer Vision
images and lidar point clouds, pho- and Pattern Recognition (CV PR),
togrammetry and pointgrammetry and coorganizer of the second edi-
for high-definition map generation, tion of EarthVision at CVPR 2017.
3D vision, and GeoAI for digital Since 2019, he has been the head of
twin cities. He has coauthored more the M.Sc. program in Engineering
than 530 publications, more than for Natural Risk Management at the
330 of which were published in University of Genoa. Since 2021, he
refereed journals. He has also pub- has been a member of the national
lished papers in flagship conferenc- evaluation committee for national
es in computer vision and AI. His scientific qualification (Abilitazione
publications have received more Scientifica Nazionale) as a full pro-
than 15,000 Google citations with fessor in the telecommunications
an h-index of 65. He has supervised field in Italy.
©IGARSS 2023
more than 120 master’s and Ph.D. The fourth Fellow recognition
students as well as postdoctoral fel- was presented to Prof. Ping Yang
lows to completion. He is a fellow FIGURE 9. IEEE Fellow Award recipient Prof. w it h t he c itat ion (see Fig ure 9)
of both the Canadian Academy of Ping Yang. “For s e m i n a l cont r ibut ion s to
FIGURE 10. IEEE Fellow Award recipients Prof. Gabriele Moser (second from left), Prof. Jonathan Li, Prof. James Garrison, and Prof. Ping
Yang (second from right) with GRSS President Dr. Mariko Burgin (right) and IEEE President-Elect Dr. Tom Coughlin (left).
©IGARSS 2023
the deputy project scientist of the
NASA Soil Moisture Active Passive
mission from 2013 January to Sep- FIGURE 11. 2023 IEEE GRSS Outstanding Service Award recipient Dr. Simon Yueh.
tember 2013, and the Soil Moisture
Active Passive Project scientist since October 2013. He has affiliation is required. The awardee receives a certificate
been the principal/coinvestigator of numerous NASA and and a plaque.
U.S. Department of Defense research projects on remote The 2023 GRSS Education Award was presented to
sensing of ocean salinity, ocean wind, terrestrial snow, Prof. Shutao Li with the citation (see Figure 12) “In recog-
and soil moisture. He has authored four book chapters nition of his significant educational contributions to geo-
and published more than 300 publications and presenta- science and remote sensing.”
tions. He received the 2021 J-STARS Prize Paper Award, Shutao Li received his B.S., M.S., and Ph.D. degrees
the 2014, 2010, and 2002 GRSS Transactions Prize Paper from Hunan University, Changsha, China, in 1995,
awards, the 2000 Best Paper Award at IGARSS 2000, and 1997, and 2001, respectively. Prof. Shutao Li has been a
the 1995 GRSS Transactions Prize Paper award for a paper full professor with the College of Electrical and Informa-
on polarimetric radiometry. He received the JPL Lew Allen tion Engineering, Hunan University, since 2004 and is
Award for Excellence in 1998, the Ed Stone Award in 2003, currently the vice rector of Hunan University. Prof. Li’s
the NASA Exceptional Technology Achievement Medal in current research interests include remote sensing image
2014, and the NASA Outstanding Public Leadership Med- processing, pattern recognition, AI, and applications
al in 2017. He was an associate edi- in environmental observation, re-
tor of Radio Science from 2002 to source investigation, and precise
2006 and editor-in-chief of TGRS agriculture. He has authored or co-
from 2018 to 2022. authored more than 300 refereed
journal and international confer-
IEEE GRSS EDUCATION AWARD ence papers. He has received more
The Education Award was estab- than 28,000 citations in Google
lished to recognize an individual S c hola r ( h-i nde x : 79) a nd wa s
who has made significant educa- selected as a Clarivate Analytics’
tional contributions to the field of Global Highly Cited Researcher in
GRSS. In selecting the individual, 2018–2022. For his scientific re-
the factors considered are signifi- search contributions, he received
cance of the educational contribu- two Second-Grade State Scientific
tion in terms of innovation and and Technological Progress Awards
the extent of its overall impact. of China (in 2004 and 2006), a Sec-
The contribution can be at any ond Prize of the National Natural
level, including K-12, undergradu- Science Award by the State Council
ate, and graduate teaching, profes- of China (in 2019), and two First
sional development, and public Prize Hunan Provincial Natural
©IGARSS 2023
outreach. It can also be in any form Science Awards (in 2017 and 2022).
(e.g., textbooks, curriculum devel- Prof. Li is the founder and head of
opment, and educational program FIGURE 12. 2023 IEEE GRSS Education Award the Hunan Provincial Key Labo-
initiatives). GRSS membership or recipient Prof. Shutao Li. rator y of Visual Perception and
©IGARSS 2023
FIGURE 13. 2023 GRSS Industry Leader Award recipient FIGURE 14. IEEE GRSS Fawwaz Ulaby Distinguished Achievement
Robbie Schingler. Award recipient Prof. Howard Zebker.
(a) (b)
FIGURE 19. AdCom participants at the GRSS AdCom meeting prior to IGARSS 2023: (a) AdCom at large. (b) Executive AdCom members.
the summer season, Elia Saikaly put the original team one in the attempt of climbing Mount Everest. The other
back together because the misfortune of their friends was side of the problem is not only the foreigner’s ambitions,
a shock for all of them, and this time they were able to but also the ambitions of the locals, for whom being in
find the bodies of the three rope companions from Paki- business around Mount Everest is a way out of poverty.
stan, Chile, and Iceland. Now local people are trying to fight for insurance and re-
Elia Saikaly was on Mount Everest again this year. He tirement plans.
shared with us reflective thoughts about the reasons why Another issue with Mount Everest is the environmen-
people are dying on Mount Everest. The media oversim- tal disaster caused by the climbers. They leave their trash
plify the reasons and blame traffic as only this season 478 behind, which is very often the result of poor planning
foreign climbers received government permission to climb because no manpower is accounted for to leave no trace
Everest—this is more tourists than usual. Elia Saikaly in- behind. A further issue is the overcommercialization
terviewed mountaineers and families who had lost some- of Mount Everest with clients who are interested only in
speed records and multiple peaks. As people are hovered
by helicopters across the valley, more inexperience and
negligence can be observed. People on the inside of the
country have to change this, but as storytellers, bystanders
and supporters can also help get the word out to finally
find positive solutions.
At the end of his speech, Elia Saikaly addressed the ef-
fects of climate change. The ice and snow layer is also get-
ting thinner on Mount Everest. He compared a photograph
taken by Sir George Mallory in 1921 from a glacier in Tibet
and another from 2009 taken from the same position. Ac-
cording to calculations, the glacier is 300 feet less dense.
Finally, Elia Saikaly reminded us of the responsibility of
the storyteller. Everyone can communicate their message
and share their work with the world, and the possibilities
are unlimited.
TAP.
CONNECT.
NETWORK.
SHARE.
Connect to IEEE–no matter where you are–with the IEEE App.
Stay up-to-date Schedule, manage, or Get geo and interest-based
with the latest news join meetups virtually recommendations
Data System
Applications
Landsat Sentinel
Unsupervised Foundation
Fine-Tuning
Learning Models
Harmonization
Analysis Segmentation
(NDVI and Classification (Burn Scars,
so on) Flood) Other
Harmonized
Use Cases
Landsat and
Sentinel
FIGURE 1. The practical data science workflow used during the summer school session (green boxes indicate hands-on experience). NDVI:
Normalized Difference Vegetation Index.
The international nature of the participants further ing the TiTiler services and Christopher Phillips from the
emphasized the global relevance and impact of data sci- University of Alabama in Huntsville for putting together
ence in today’s interconnected world. The session adopted training datasets for fine-tuning the GeoFM. The session re-
a practical approach, centering on real-world application sources are available at https://github.com/NASA-IMPACT/
scenarios. The participants actively engaged in hands-on summer-school-2023. The GeoFM, fine-tuned models, and
exercises and practical demonstrations. The session aimed data are openly available at https://huggingface.co/ibm
to foster collaboration and create a platform for exchanging -nasa-geospatial.
ideas among participants. The success of the session high-
lighted the value of collaboration, hands-on experience, REFERENCES
and the application of foundational models in geospatial [1] “HDCRS summer school 2023.” GRSS-IEEE. Accessed: Jul. 16,
analysis. Furthermore, the hands-on interactions enabled 2023. [Online]. Available: https://www.grss-ieee.org/community/
the organizers to document the necessary steps to be taken groups-initiatives/high-performance-and-disruptive
from data science tools, platforms, and compute infrastruc- -computing-in-remote-sensing-hdcrs/hdcrs-summer
ture requirements, enabling large-scale collaborative proj- -school-2023/
ects to analyze Earth-observation data (see Figure 2). [2] “Open-source science initiative.” NASA Science. Accessed:
Jul. 20, 2023. [Online]. Available: https://science.nasa.gov/
CONCLUSION open-science-overview
The specialized session on data science organized by NASA, [3] National Academies of Sciences, Engineering, and Medicine,
IBM Research, and the University of Iceland showcased the Data Science for Undergraduates: Opportunities and Options.
power of collaboration, hands-on experience, and the ap- Washington, DC, USA: The National Academies Press, 2018.
plication of foundational models in geospatial analysis. As [4] “HLS overview,” United States Geological Survey, Valley Drive
data science continues to shape the future, it is crucial to Reston, VA, USA. Accessed: Jul. 16, 2023. [Online]. Available: https://
empower professionals with the necessary skills and knowl- lpdaac.usgs.gov/data/get-started-data/collection-overview/
edge to harness the potential of data-driven decision mak- missions/harmonized-landsat-sentinel-2-hls-overview/
ing. The session stands as an important milestone in this [5] “eoAPI-raster.” Cloudfront.net. Accessed: Jul. 16, 2023. [Online].
journey, reflecting a commitment to enhance the capacity Available: https://d1nzvsko7rbono.cloudfront.net/docs
for data science and push the boundaries of what is achiev- [6] “IBM’s new geospatial foundation model.” IBM. Accessed:
able. Due to the success of the session, we have plans to Jun. 30, 2023. [Online]. Available: https://research.ibm.com/
continue organizing the session with a data science theme blog/geospatial-models-nasa-ai
during future summer schools. [7] R. Bommasani et al., “On the opportunities and risks of
foundation models,” 2021, arXiv:2108.07258.
ACKNOWLEDGMENT [8] Science Managed Cloud Environment. Accessed: Jul. 17,
Our heartfelt gratitude goes to the ESI TC for its generous 2023. [Online]. Available: https://smce.nasa.gov/
sponsorship of the workshop and the GRSS for its ongoing [9] M. Maskey et al., “Dashboard for earth observation,” in Advances
support. Additionally, we would like to express our special in Scalable and Intelligent Geospatial Analytics, 1st ed. Boca Raton,
thanks to the University of Iceland for hosting the event FL, USA: CRC Press, 2023.
and for the invaluable assistance provided by the dedicated [10] “Interactive computing for your community.” 2i2c. Accessed:
volunteers. A special mention goes to Sean Harkins from Jul. 17, 2023. [Online]. Available: https://2i2c.org/
Development Seed for his invaluable support in facilitat- GRS
Remote sensing (RS) image understanding aims at extracting valuable information and acquiring knowledge from remotely sensed
data, and artificial intelligence (AI) plays a significant role. With the increased data availability and the development of techniques for
data interpretation –particularly, deep learning (DL) techniques– the past few years have witnessed a tremendous growth of
research efforts focused on the visual interpretation of remote sensing images. Such techniques have made significant
breakthroughs in multiple domains, such as scene classification, object detection, feature extraction and recognition, and land-
use/land-cover mapping, to name a few. Nevertheless, there are still several challenges in this field, mostly related to the robustness
and transferability of interpretation approaches, the efficient perception and understanding of RS mages, the effective fusion and
utilization of multi-modal RS data, etc. This special issue that is aimed at investigating new techniques, algorithms and architectures
that can be used to overcome the above-mentioned challenges and bring together the state-of-the-art research in this field.
Articles submitted to this special issue of the IEEE Geoscience and Remote Sensing Magazine must contain significant
relevance to geoscience and remote sensing and should have noteworthy tutorial/review value. Selection of invited papers
will be done on the basis of 4-page White papers, submitted in double-column format. These papers must discuss the
foreseen objectives of the paper, the importance of the addressed topic, the impact of the contribution, and the authors’
expertise and past activities on the topic. Contributors selected on the basis of the White papers will be invited to submit
full manuscripts. Manuscripts should be submitted online at http://mc.manuscriptcentral.com/grsm using the Manuscript Central
interface. Prospective authors should consult the site http://ieeexplore.ieee.org/servlet/opac?punumber=6245518 for guidelines and
information on paper submission. Submitted articles should not have been published or be under review elsewhere. All submissions
will be peer reviewed according to the IEEE and Geoscience and Remote Sensing Society guidelines.
BENEFITS:
• Rapidly disseminate your research findings
Follow @TechRxiv_org
Learn more techrxiv.org Powered by IEEE