The Experts below are selected from a list of 96 Experts worldwide ranked by ideXlab platform
Tuia Devis - One of the best experts on this subject based on the ideXlab platform.
-
RSVQA: Visual Question Answering for Remote Sensing Data
'Institute of Electrical and Electronics Engineers (IEEE)', 2020Co-Authors: Lobry Sylvain, Marcos Diego, Murray Jesse, Tuia DevisAbstract:This article introduces the task of visual question answering for Remote Sensing data (RSVQA). Remote Sensing images contain a wealth of information, which can be useful for a wide range of tasks, including land cover classification, object counting, or detection. However, most of the available methodologies are task-specific, thus inhibiting generic and easy access to the information contained in Remote Sensing data. As a consequence, accurate Remote Sensing Product generation still requires expert knowledge. With RSVQA, we propose a system to extract information from Remote Sensing data that is accessible to every user: we use questions formulated in natural language and use them to interact with the images. With the system, images can be queried to obtain high-level information specific to the image content or relational dependencies between objects visible in the images. Using an automatic method introduced in this article, we built two data sets (using low- and high-resolution data) of image/question/answer triplets. The information required to build the questions and answers is queried from OpenStreetMap (OSM). The data sets can be used to train (when using supervised methods) and evaluate models to solve the RSVQA task. We report the results obtained by applying a model based on convolutional neural networks (CNNs) for the visual part and a recurrent neural network (RNN) for the natural language part of this task. The model is trained on the two data sets, yielding promising results in both cases
-
RSVQA: Visual Question Answering for Remote Sensing Data
'Institute of Electrical and Electronics Engineers (IEEE)', 2020Co-Authors: Lobry Sylvain, Marcos Diego, Murray Jesse, Tuia DevisAbstract:This paper introduces the task of visual question answering for Remote Sensing data (RSVQA). Remote Sensing images contain a wealth of information which can be useful for a wide range of tasks including land cover classification, object counting or detection. However, most of the available methodologies are task-specific, thus inhibiting generic and easy access to the information contained in Remote Sensing data. As a consequence, accurate Remote Sensing Product generation still requires expert knowledge. With RSVQA, we propose a system to extract information from Remote Sensing data that is accessible to every user: we use questions formulated in natural language and use them to interact with the images. With the system, images can be queried to obtain high level information specific to the image content or relational dependencies between objects visible in the images. Using an automatic method introduced in this article, we built two datasets (using low and high resolution data) of image/question/answer triplets. The information required to build the questions and answers is queried from OpenStreetMap (OSM). The datasets can be used to train (when using supervised methods) and evaluate models to solve the RSVQA task. We report the results obtained by applying a model based on Convolutional Neural Networks (CNNs) for the visual part and on a Recurrent Neural Network (RNN) for the natural language part to this task. The model is trained on the two datasets, yielding promising results in both cases.Comment: 12 pages, Published in IEEE Transactions on Geoscience and Remote Sensing. Added one experiment and authors' biographie
Allan Weiner - One of the best experts on this subject based on the ideXlab platform.
-
a high performance Remote Sensing Product generation system based on a service oriented architecture for the next generation of geostationary operational environmental satellites
Remote Sensing, 2015Co-Authors: Satya Kalluri, James Gundy, Brian J Haman, Anthony Paullin, Paul Van Rompay, David Vititoe, Allan WeinerAbstract:The Geostationary Operational Environmental Satellite (GOES) series R, S, T, U (GOES-R) will collect Remote Sensing data at several orders of magnitude compared to legacy missions, 24 × 7, over its 20-year operational lifecycle. A suite of 34 Earth and space weather Products must be produced at low latency for timely delivery to forecasters. A ground system (GS) has been developed to meet these challenging requirements, using High Performance Computing (HPC) within a Service Oriented Architecture (SOA). This approach provides a robust, flexible architecture to support the operational GS as it generates Remote Sensing Products by ingesting and combining data from multiple sources. Test results show that the system meets the key latency and availability requirements for all Products.
Lobry Sylvain - One of the best experts on this subject based on the ideXlab platform.
-
RSVQA: Visual Question Answering for Remote Sensing Data
'Institute of Electrical and Electronics Engineers (IEEE)', 2020Co-Authors: Lobry Sylvain, Marcos Diego, Murray Jesse, Tuia DevisAbstract:This article introduces the task of visual question answering for Remote Sensing data (RSVQA). Remote Sensing images contain a wealth of information, which can be useful for a wide range of tasks, including land cover classification, object counting, or detection. However, most of the available methodologies are task-specific, thus inhibiting generic and easy access to the information contained in Remote Sensing data. As a consequence, accurate Remote Sensing Product generation still requires expert knowledge. With RSVQA, we propose a system to extract information from Remote Sensing data that is accessible to every user: we use questions formulated in natural language and use them to interact with the images. With the system, images can be queried to obtain high-level information specific to the image content or relational dependencies between objects visible in the images. Using an automatic method introduced in this article, we built two data sets (using low- and high-resolution data) of image/question/answer triplets. The information required to build the questions and answers is queried from OpenStreetMap (OSM). The data sets can be used to train (when using supervised methods) and evaluate models to solve the RSVQA task. We report the results obtained by applying a model based on convolutional neural networks (CNNs) for the visual part and a recurrent neural network (RNN) for the natural language part of this task. The model is trained on the two data sets, yielding promising results in both cases
-
RSVQA: Visual Question Answering for Remote Sensing Data
'Institute of Electrical and Electronics Engineers (IEEE)', 2020Co-Authors: Lobry Sylvain, Marcos Diego, Murray Jesse, Tuia DevisAbstract:This paper introduces the task of visual question answering for Remote Sensing data (RSVQA). Remote Sensing images contain a wealth of information which can be useful for a wide range of tasks including land cover classification, object counting or detection. However, most of the available methodologies are task-specific, thus inhibiting generic and easy access to the information contained in Remote Sensing data. As a consequence, accurate Remote Sensing Product generation still requires expert knowledge. With RSVQA, we propose a system to extract information from Remote Sensing data that is accessible to every user: we use questions formulated in natural language and use them to interact with the images. With the system, images can be queried to obtain high level information specific to the image content or relational dependencies between objects visible in the images. Using an automatic method introduced in this article, we built two datasets (using low and high resolution data) of image/question/answer triplets. The information required to build the questions and answers is queried from OpenStreetMap (OSM). The datasets can be used to train (when using supervised methods) and evaluate models to solve the RSVQA task. We report the results obtained by applying a model based on Convolutional Neural Networks (CNNs) for the visual part and on a Recurrent Neural Network (RNN) for the natural language part to this task. The model is trained on the two datasets, yielding promising results in both cases.Comment: 12 pages, Published in IEEE Transactions on Geoscience and Remote Sensing. Added one experiment and authors' biographie
Wanjun Zhang - One of the best experts on this subject based on the ideXlab platform.
-
research on generic optical Remote Sensing Products a review of scientific exploration technology research and engineering application
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021Co-Authors: Yang Liu, Xianyu Zuo, Junfeng Tian, Kun Cai, Wanjun ZhangAbstract:With the initial establishment of global earth observation system in various countries, more and more high-resolution Remote Sensing data of multisource, multitemporal, multiscale, and different types of satellites are obtained. It is urgent to explore the advanced basic theory of Remote Sensing information science, design high-performance generic key technologies of Remote Sensing information system and global positioning system, and study complex engineering system of Remote Sensing applications and geographic information system. In this article, the basic theory exploration, inversion technology research, and engineering application design and development of generic optical Remote Sensing Product (ORSP) are systematically reviewed. We classify the ORSP scientifically, review the main algorithms and application scope of 16 kinds of generic ORSP, and expound the validation and quality evaluation methods of ORSP in engineering application. Furthermore, we analyze the current core problems and solutions, and prospects for the state-of-the-art research and the future development trend of generic ORSP. This will provide valuable reference for scientific research and construction of high-resolution earth observation system.
Qing Xiao - One of the best experts on this subject based on the ideXlab platform.
-
advances in quantitative Remote Sensing Product validation overview and current status
Earth-Science Reviews, 2019Co-Authors: Qing Xiao, Jianguang Wen, Dongqin You, Andreas HueniAbstract:The roles of quantitative Remote Sensing Products in scientific research and quantitative applications lie in their ability to record the real states of earth surface features. Therefore, it is crucial to quantify the performance of Remote Sensing Products. However, validation is not straightforward due to the scale effects caused by surface heterogeneity and the spatial scale mismatch between satellite- and ground-based observations, but a challenging scientific issue in the field of Remote Sensing. Although validation works have been widely carried out in the past decades, it is difficult to reach an accordant and compelling conclusion about the performance of different satellite Products due to the inconsistencies in spatial and temporal extents, the type of in situ data sources, and the validation strategies, which hinder effective applications of Remote Sensing Products. Therefore, it is necessary to give an overview of validation, simultaneously to point out its problems and insufficiencies, and finally to put forward the suggestions for future researches. The in situ data acquisition, the upscaling method, the uncertainties implied in the validation process, together with future challenges, are included in this paper. It is expected to promote the validation technique development and improve the application accuracy of Remote Sensing Products.
-
a web based land surface Remote Sensing Products validation system lapvas application to albedo Product
International Journal of Digital Earth, 2018Co-Authors: Xingwen Lin, Jianguang Wen, Yong Tang, Dongqin You, Baocheng Dou, Xiaobo Zhu, Qing Xiao, Qing Huo LiuAbstract:ABSTRACTQuantitative Remote Sensing Product (QRSP) validation is a complex process to assess the accuracy and uncertainty independently using reference data with multiple land cover types and long-time series. A web-based system named as LAnd surface Remote Sensing Product VAlidation system (LAPVAS) is described in this paper, which is used to implement the QRSPs validation process automatically. The LAPAVS has two subsystems, the Validation Databases Subsystem and the Accuracy Evaluation Subsystem. Three functions have been implemented by the two subsystems for a comprehensive QRSP validation: (1) a standardized processing of reference data and storage of these data in validation databases; (2) a consistent and comprehensive validation procedure to assess the QRSPs’ accuracy and uncertainty; and (3) a visual process customization tool with which the users can register new validation data, host new reference data, and readjust the validation workflows for the QRSP accuracy assessment. In LAPVAS, more than...