Showing papers in "IEEE Geoscience and Remote Sensing Magazine in 2016"
TL;DR: A general framework of DL for RS data is provided, and the state-of-the-art DL methods in RS are regarded as special cases of input-output data combined with various deep networks and tuning tricks.
Abstract: Deep-learning (DL) algorithms, which learn the representative and discriminative features in a hierarchical manner from the data, have recently become a hotspot in the machine-learning area and have been introduced into the geoscience and remote sensing (RS) community for RS big data analysis. Considering the low-level features (e.g., spectral and texture) as the bottom level, the output feature representation from the top level of the network can be directly fed into a subsequent classifier for pixel-based classification. As a matter of fact, by carefully addressing the practical demands in RS applications and designing the input?output levels of the whole network, we have found that DL is actually everywhere in RS data analysis: from the traditional topics of image preprocessing, pixel-based classification, and target recognition, to the recent challenging tasks of high-level semantic feature extraction and RS scene understanding.
1,625 citations
TL;DR: A critical review of the recent advances in DA approaches for remote sensing is provided and an overview of DA methods divided into four categories: 1) invariant feature selection, 2) representation matching, 3) adaptation of classifiers, and 4) selective sampling are presented.
Abstract: The success of the supervised classification of remotely sensed images acquired over large geographical areas or at short time intervals strongly depends on the representativity of the samples used to train the classification algorithm and to define the model. When training samples are collected from an image or a spatial region that is different from the one used for mapping, spectral shifts between the two distributions are likely to make the model fail. Such shifts are generally due to differences in acquisition and atmospheric conditions or to changes in the nature of the object observed. To design classification methods that are robust to data set shifts, recent remote sensing literature has considered solutions based on domain adaptation (DA) approaches. Inspired by machine-learning literature, several DA methods have been proposed to solve specific problems in remote sensing data classification. This article provides a critical review of the recent advances in DA approaches for remote sensing and presents an overview of DA methods divided into four categories: 1) invariant feature selection, 2) representation matching, 3) adaptation of classifiers, and 4) selective sampling. We provide an overview of recent methodologies, examples of applications of the considered techniques to real remote sensing images characterized by very high spatial and spectral resolution as well as possible guidelines for the selection of the method to use in real application scenarios.
331 citations
TL;DR: In this paper, the main theoretical Gaussian Process (GPs) developments in the field of biogeophysical parameter retrieval are reviewed, considering new algorithms that respect signal and noise characteristics, extract knowledge via automatic relevance kernels, and allow applicability of associated uncertainty intervals to transport GP models in space and time that can be used to uncover causal relations between variables and can encode physically meaningful prior knowledge via radiative transfer model (RTM) emulation.
Abstract: Gaussian processes (GPs) have experienced tremendous success in biogeophysical parameter retrieval in the last few years. GPs constitute a solid Bayesian framework to consistently formulate many function approximation problems. This article reviews the main theoretical GP developments in the field, considering new algorithms that respect signal and noise characteristics, extract knowledge via automatic relevance kernels to yield feature rankings automatically, and allow applicability of associated uncertainty intervals to transport GP models in space and time that can be used to uncover causal relations between variables and can encode physically meaningful prior knowledge via radiative transfer model (RTM) emulation. The important issue of computational efficiency will also be addressed. These developments are illustrated in the field of geosciences and remote sensing at local and global scales through a set of illustrative examples. In particular, important problems for land, ocean, and atmosphere monitoring are considered, from accurately estimating oceanic chlorophyll content and pigments to retrieving vegetation properties from multi- and hyperspectral sensors as well as estimating atmospheric parameters (e.g., temperature, moisture, and ozone) from infrared sounders.
185 citations
TL;DR: This article provides a holistic view of generic data fusion concepts and their applicability to the remote sensing domain and a review of current activities in the field of Earth observation is given.
Abstract: Characterized by a certain focus on the heavily discussed topic of image fusion in its beginnings, sensor data fusion has played a significant role in the remote sensing research community for a long time. With this article, we aim to provide a short overview of established definitions, targeting a generalized understanding of the topic. In addition, a review of the state of the art of remote sensing data fusion research is given. By bringing together the conventional view expressed in the classical data fusion community and a review of current activities in the field of Earth observation, this article provides a holistic view of generic data fusion concepts and their applicability to the remote sensing domain.
161 citations
TL;DR: A brief overview of the challenges in monitoring land-cover changes from the perspective of machine learning and some of the recent advances in machine learning that are relevant for addressing them are discussed.
Abstract: Monitoring land-cover changes is of prime importance for the effective planning and management of critical, natural and man-made resources. The growing availability of remote sensing data provides ample opportunities for monitoring land-cover changes on a global scale using machine-learning techniques. However, remote sensing data sets exhibit unique domain-specific properties that limit the usefulness of traditional machine-learning methods. This article presents a brief overview of these challenges from the perspective of machine learning and discusses some of the recent advances in machine learning that are relevant for addressing them. These approaches show promise for future research in the detection of land-cover change using machine-learning algorithms.
81 citations
TL;DR: The advantages of using machine-learning techniques in ground-based image analysis via three primary applications: segmentation, classification, and denoising are demonstrated.
Abstract: Ground-based whole-sky cameras have opened up new opportunities for monitoring the earth?s atmosphere. These cameras are an important complement to satellite images by providing geoscientists with cheaper, faster, and more localized data. The images captured by whole-sky imagers (WSI) can have high spatial and temporal resolution, which is an important prerequisite for applications such as solar energy modeling, cloud attenuation analysis, local weather prediction, and more. Extracting the valuable information from the huge amount of image data by detecting and analyzing the various entities in these images is challenging. However, powerful machine-learning techniques have become available to aid with the image analysis. This article provides a detailed explanation of recent developments in these techniques and their applications in ground-based imaging, aiming to bridge the gap between computer vision and remote sensing with the help of illustrative examples. We demonstrate the advantages of using machine-learning techniques in ground-based image analysis via three primary applications: segmentation, classification, and denoising.
51 citations
TL;DR: This article presents the life cycle of big, linked, and open EO data and shows how to support their various stages using the software stack developed by the EU research projects TELEIOS and the Linked Open EO Data for Precision Farming.
Abstract: Big Earth-observation (EO) data that are made freely available by space agencies come from various archives. Therefore, users trying to develop an application need to search within these archives, discover the needed data, and integrate them into their application. In this article, we argue that if EO data are published using the linked data paradigm, then the data discovery, data integration, and development of applications becomes easier. We present the life cycle of big, linked, and open EO data and show how to support their various stages using the software stack developed by the European Union (EU) research projects TELEIOS and the Linked Open EO Data for Precision Farming (LEO). We also show how this stack of tools can be used to implement an operational wildfire-monitoring service.
36 citations
TL;DR: The ICESat-2 mission invested in an applications program aimed at innovatively applying the data in a variety of fields to maximize the use of data products after launch and to provide early insight into the range of potential uses of the mission data.
Abstract: NASA's Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2) mission, scheduled to launch no later than April 2018 (and currently slated for October 2017), is being developed to continue the multiyear observations of the earth's surface elevation, ice, and clouds started by ICESat. To increase the use of the satellite data after launch, the ICESat-2 mission invested in an applications program aimed at innovatively applying the data in a variety of fields. The program provides a framework for building a broad and well-defined user community during the prelaunch period to maximize the use of data products after launch and to provide early insight into the range of potential uses of the mission data. Ideas and research on how altimetry data will be used for decision making arise from the end users; therefore, the ICESat-2 mission is extending itself through its applications program.
26 citations
TL;DR: An overview of some of climate science's big data problems and the technical solutions being developed to advance data publication, climate analytics as a service, and interoperability within the Earth System Grid Federation (ESGF), which is the primary cyberinfrastructure currently supporting global climate research activities.
Abstract: The knowledge we gain from research in climate science depends on the generation, dissemination, and analysis of high-quality data. This work comprises technical practice as well as social practice, both of which are distinguished by their massive scale and global reach. As a result, the amount of data involved in climate research is growing at an unprecedented rate. Some examples of the types of activities that increasingly require an improved cyberinfrastructure for dealing with large amounts of critical scientific data are climate model intercomparison (CMIP) experiments; the integration of observational data and climate reanalysis data with climate model outputs, as seen in the Observations for Model Intercomparison Projects (Obs4MIPs), Analysis for Model Intercomparison Projects (Ana4MIPs), and Collaborative Reanalysis Technical Environment-Intercomparison Project (CREATE-IP) activities; and the collaborative work of the Intergovernmental Panel on Climate Change (IPCC). This article provides an overview of some of climate science's big data problems and the technical solutions being developed to advance data publication, climate analytics as a service, and interoperability within the Earth System Grid Federation (ESGF), which is the primary cyberinfrastructure currently supporting global climate research activities.
25 citations
TL;DR: This work states that big data technologies are being widely practiced in Earth sciences and remote sensing communities to support EO data access, processing, and knowledge discovery.
Abstract: Recent trends on big Earth-observing (EO) data lead to some questions that the Earth science community needs to address. Are we experiencing a paradigm shift in Earth science research now? How can we better utilize the explosion of technology maturation to create new forms of EO data processing? Can we summarize the existing methodologies and technologies scaling to big EO data as a new field named earth data science? Big data technologies are being widely practiced in Earth sciences and remote sensing communities to support EO data access, processing, and knowledge discovery. The data-intensive scientific discovery, named the fourth paradigm, leads to data science in the big data era [1]. According to the definition by the U.S. National Institute of Standards and Technology, the data science paradigm is the "extraction of actionable knowledge directly from data through a process of discovery, hypothesis, and hypothesis testing" [2]. Earth data science is the art and science of applying the data science paradigm to EO data.
25 citations
TL;DR: There are opportunities for how citizens can become part of an integrated EO monitoring system in the framework of the European Union (EU) space program, including Copernicus and other monitoring initiatives.
Abstract: Citizens are providing vast amounts of georeferenced data in the form of in situ data collections as well as interpretations and digitization of Earth-observation (EO) data sets. These new data streams have considerable potential for supporting the calibration and validation of current and future products derived from EO. We provide a general introduction to this growing area of interest and review existing crowdsourcing and citizen science (CS) initiatives of relevance to EO. We then draw upon our own experiences to provide case studies that highlight different types of data collection and citizen engagement and discuss the various barriers to adoption. Finally, we highlight opportunities for how citizens can become part of an integrated EO monitoring system in the framework of the European Union (EU) space program, including Copernicus and other monitoring initiatives.
TL;DR: This report addresses the spectrum issues faced by active science sensors, primarily radars, and makes recommendations to government, industry, and the remote sensing community going forward.
Abstract: The scientific users of radio frequencies must contend with the fact that the spectrum is becoming increasingly crowded, which is in large measure due to the advent of advanced affordable electronics and mobile wireless technology. The growing demand for bandwidth has sparked increased discussions in the microwave remote sensing community of how to respond to this crowded spectrum environment and how to deal with the consequent issues of radio frequency interference (RFI). The National Research Council (NRC) published a study in 2010, "Spectrum Management for Science in the 21st Century" [1], that examined the increasing difficulties encountered by passive microwave measurements in the presence of the expanding worldwide commercial and governmental occupancy of the radio spectrum. The challenges faced by passive sensors also have been summarized in a 2014 IEEE Geoscience and Remote Sensing Magazine article [2]. Recognizing that active microwave sensors also face spectrum-related issues, NASA later commissioned the NRC to perform a similar study, "A Strategy for Active Remote Sensing Amid Increased Demand for Radio Spectrum," which was recently published in July 2015 [3]. (In this article, the report will be abbreviated as the NRC Active Sensing Report.) This report addresses the spectrum issues faced by active science sensors, primarily radars, and makes recommendations to government, industry, and the remote sensing community going forward. The report considers multiple types of active sensors including ground-based operational weather radars, ionospheric sensing radar, and radar astronomy. This article focuses on spectrum topics related primarily to Earth remote sensing from aircraft and spacecraft.
TL;DR: This special issue aims at providing an updated, refreshing view of current developments in machine learning, with five articles that present snapshots of the recent advances in machine-learning methodologies for remote sensing and geosciences.
Abstract: Machine learning has become a standard paradigm for the analysis of remote sensing and geoscience data at both local and global scales. In the upcoming years, with the advent of new satellite constellations, machine learning will have a fundamental role in processing large and heterogeneous data sources. Machine learning will move from mere statistical data processing to actual learning, understanding, and knowledge extraction. The ambitious goal is to provide responses to the challenging scientific questions about the earth system. This special issue aims at providing an updated, refreshing view of current developments in the field. For this special issue, we have collected five articles that present snapshots of the recent advances in machine-learning methodologies for remote sensing and geosciences.
TL;DR: This work will demonstrate how EO data life cycles benefit from the proximity of data management and application scientists and from the extensive operational experience gathered over time.
Abstract: The German Satellite Data Archive (D-SDA) at the German Aerospace Center (DLR) has been managing largevolume Earth-observation (EO) data in the context of EOmission payload ground segments (PGSs) for more than two decades. Hardware, data management, processing, user access, long-term preservation, and data exploitation expertise are under one roof and interact closely. Upcoming EO-mission PGSs benefit as much from the comprehensive expertise, close interaction, and integrated infrastructure as do in-house scientific application projects requiring access, processing, and archiving of large-volume EO data. Using a number of examples, we will demonstrate how EO data life cycles benefit from the proximity of data management and application scientists and from the extensive operational experience gathered over time.
TL;DR: In this paper, the authors present two methods for measuring water-surface backscattering signature and estimating the near-surface wind vector over water using airborne weather radar (AWR).
Abstract: It is widely known that spaceborne synthetic aperture radar (SAR) can provide sea-surface wind field maps that are very useful for applications where knowledge of the sea-surface wind at fine scales is crucial, such as the analysis of waves, ocean circulation, marine meteorology, and the relationships between oceanic and atmospheric systems. Perhaps it is less well known that airborne weather radar (AWR) may also be used for the same purpose. In this article, we present two methods for measuring water-surface backscattering signature and estimating the near-surface wind vector over water using AWR. The latter operates in the ground-mapping mode as a scatterometer. An estimate of the azimuth normalized radar cross section (NRCS) curve of the water surface is obtained from an aircraft's circular or rectilinear flight. The wind vector is retrieved from the azimuth NRCS curve. In this article, some measuring recommendations and algorithms are proposed.
TL;DR: The current status and challenges in the fusion of remotely sensed displacement measurements are addressed and some potential ways to deal with heterogeneous data types and to assimilate remote sensing data into physical models to realize near-real-time displacement monitoring are proposed.
Abstract: Nowadays, data fusion constitutes the key subject in numerous applications of remotely sensed displacement measurements, with the increasing availability of remote sensing data and the requirement of improving the measurement accuracy. This article addresses the current status and challenges in the fusion of remotely sensed displacement measurements. An overview is given to discuss the remote sensing sources and techniques extensively used for displacement measurement and the recent development and achievement of displacement measurements fusion. The fusion between displacement measurements and the integration of a geophysical model are discussed. The fusion strategies and uncertainty propagation approaches are illustrated in two main applications: 1) the surface displacement measurements fusion to retrieve surface displacement with a reduced uncertainty in case of redundancy, with larger spatial extension, or of a higher level in case of complementarity, and 2) the surface displacement measurements fusion to estimate the geometrical parameters of a physical deformation model in case of redundancy and complementarity. Finally, the current status and challenges of remotely sensed displacement measurements fusion are highlighted. Moreover, some potential ways to deal with heterogeneous data types and to assimilate remote sensing data into physical models to realize near-real-time displacement monitoring are proposed.
TL;DR: The five articles in this special section address the key topics related to big data that refer to data volume, velocity, variety, veracity, and value.
Abstract: The five articles in this special section address the key topics related to big data that refer to data volume, velocity, variety, veracity, and value.
TL;DR: The NASA Jet Propulsion Laboratory (JPL) is a national research facility that carries out cuttingedge earth science missions as mentioned in this paper, and is a pioneer in the use of remote sensing for science of the oceans, atmosphere and solid earth.
Abstract: The NASA Jet Propulsion Laboratory (JPL) is a national research facility that carries out cuttingedge earth science missions. JPL developed the first U.S. Earth-orbiting science spacecraft and is a pioneer in the use of remote sensing for science of the oceans, atmosphere, and solid earth. Explorer I was the first U.S. Earth-orbiting spacecraft. It followed the Soviet Sputniks 1 and 2 but carried James Van Allen's Geiger counter, which upended space physics with the discovery of the radiation belts now named for him [1]. Explorer I also carried a micrometeoroid detector. JPL developed atmospheric temperature instruments for the Nimbus series of weather satellites, built microwave and infrared instruments to help gain an understanding of stratospheric ozone depletion, ocean circulation, and surface winds, and flew Seasat, which carried the first civilian synthetic aperture radar. This article gives a general overview of recent, current, and near-future earth science missions led by JPL, highlighting a few of the many measurements that are transforming our understanding of the processes governing the Earth's atmosphere, oceans, land surfaces, and climate.
TL;DR: The 2016 Data Fusion Contest as discussed by the authors was organized by the Image Analysis and Data Fusion Technical Committee (IADF TC) of the IEEE Geoscience and Remote Sensing Society (GRSS), which aims at providing a challenging image analysis opportunity including multitemporal, multiresolution, and multisensor fusion.
Abstract: The 2016 Data Fusion Contest, organized by the Image Analysis and Data Fusion Technical Committee (IADF TC) of the IEEE Geoscience and Remote Sensing Society (GRSS), aims at providing a challenging image analysis opportunity including multitemporal, multiresolution, and multisensor fusion. The 2016 contest involves two data modalities acquired over the city of Vancouver, Canada (49?15?N 123?6?W): a very high-resolution multitemporal sequence and an ultrahigh definition (UHD) video from space acquired from the International Space Station (ISS). The data were acquired and provided by Deimos Imaging and UrtheCast (Figure 1).
TL;DR: A parallel uncertainty-aware drainage basin delineation algorithm and a multinode GPU compute unified device architecture (CUDA) implementation along with scalability benchmarking shows a nearly linear strong scalability.
Abstract: Processing high-resolution digital elevation models (DEMs) can be tedious due to the large size of the data. In uncertainty-aware drainage basin delineation, we apply a Monte Carlo (MC) simulation that further increases the processing demand by two to three orders of magnitude. Utilizing graphics processing units (GPUs) can speed up the programs, but their on-chip random access memory (RAM) limits the size of the DEMs that can be processed efficiently on one GPU. Here, we present a parallel uncertainty-aware drainage basin delineation algorithm and a multinode GPU compute unified device architecture (CUDA) implementation along with scalability benchmarking. All of the computations are run on the GPUs, and the parallel processes communicate using a message-passing interface (MPI) via the host central processing units (CPUs). The implementation can utilize any number of nodes, with one or many GPUs per node. The performance and scalability of the program have been tested with a 10-m DEM covering 390,905 km2, i.e., the entire area of Finland. Performing the drainage basin delineation for the DEM with different numbers of GPUs shows a nearly linear strong scalability.
Journal Article•
TL;DR: Calibration and validation determine the quality and integrity of the data provided by spaceborne imaging spectroscopy sensors and have enormous downstream impacts on the accuracy and reliability of products generated from these sensors.
Abstract: Calibration is the process of quantitatively defining a system's responses to known, controlled signal inputs, and validation is the process of assessing, by independent means, the quality of the data products derived from those system outputs [1]. Similar to other Earthobservation (EO) sensors, the calibration and validation of spaceborne imaging spectroscopy sensors is a fundamental underpinning activity. Calibration and validation determine the quality and integrity of the data provided by spaceborne imaging spectroscopy sensors and have enormous downstream impacts on the accuracy and reliability of products generated from these sensors.
TL;DR: The IEEE Geoscience and Remote Sensing Society (GRSS) Instrumentation and Future Technologies Technical Committee (IFT-TC) as discussed by the authors seeks to foster international cooperation to advance the state-of-the-art in remote sensing instruments and technologies that improve knowledge for the betterment of society and the global environment.
Abstract: The IEEE Geoscience and Remote Sensing Society (GRSS) Instrumentation and Future Technologies Technical Committee (IFT-TC) seeks to foster international cooperation to advance the state of the art in geoscience and remote sensing instrumentation and technologies that improve knowledge for the betterment of society and the global environment. The mission of the IFT-TC is to facilitate, engage, and coordinate GRSS members and communities at large to assess the current state of the art in remote sensing instruments and technology, identify new instrument concepts and relevant technology trends, and recognize enabling technologies for future instruments.
TL;DR: In this paper, the authors investigated whether the position of the moon relates in any significant way to the occurrence of earthquakes and determined whether earthquakes exhibit nonuniform behavior with regard to the earth-centered angle between the moon and the earthquake.
Abstract: This study investigated whether the position of the moon relates in any significant way to the occurrence of earthquakes. The objective of the study was to determine whether earthquakes exhibit nonuniform behavior with regard to the earth-centered angle between the moon and the earthquake. The study examined this angular separation for a set of 229 California earthquakes occurring between 1769 and 2004, ranging in magnitude from 5.2 to 8.25. Cross-correlation analysis with a matching function was employed to compensate for data scatter and sparsity and to search for underlying nonuniform behavior.