The building damage status is vital to plan rescue and reconstruction after a disaster and is also hard to detect and judge its level. Most existing studies focus on binary classification, and the attention of the model is distracted. In this study, we proposed a Siamese neural network that can localize and classify damaged buildings at one time. The main parts of this network are a variety of attention U-Nets using different backbones. The attention mechanism enables the network to pay more attention to the effective features and channels, so as to reduce the impact of useless features. We train them using the xBD dataset, which is a large-scale dataset for the advancement of building damage assessment, and compare their result balanced F (F1) scores. The score demonstrates that the performance of SEresNeXt with an attention mechanism gives the best performance among single models, with the F1 score reaching 0.787. To improve the accuracy, we fused the results and got the best overall F1 score of 0.792. To verify the transferability and robustness of the model, we selected the dataset on the Maxar Open Data Program of two recent disasters to investigate the performance. By visual comparison, the results show that our model is robust and transferable.

In the era of big data, the explosive growth of Earth observation data and the rapid advancement in cloud computing technology make the global-oriented spatiotemporal data simulation possible. These dual developments also provide advantageous conditions for discrete global grid systems (DGGS). DGGS are designed to portray real-world phenomena by providing a spatiotemporal unified framework on a standard discrete geospatial data structure and theoretical support to address the challenges from big data storage, processing, and analysis to visualization and data sharing. In this paper, the trinity of big Earth observation data (BEOD), cloud computing, and DGGS is proposed, and based on this trinity theory, we explore the opportunities and challenges to handle BEOD from two aspects, namely, information technology and unified data framework. Our focus is on how cloud computing and DGGS can provide an excellent solution to enable big Earth observation data. Firstly, we describe the current status and data characteristics of Earth observation data, which indicate the arrival of the era of big data in the Earth observation domain. Subsequently, we review the cloud computing technology and DGGS framework, especially the works and contributions made in the field of BEOD, including spatial cloud computing, mainstream big data platform, DGGS standards, data models, and applications. From the aforementioned views of the general introduction, the research opportunities and challenges are enumerated and discussed, including EO data management, data fusion, and grid encoding, which are concerned with analysis models and processing performance of big Earth observation data with discrete global grid systems in the cloud environment.

Spatial vector data with high-precision and wide-coverage has exploded globally, such as land cover, social media, and other data-sets, which provides a good opportunity to enhance the national macroscopic decision-making, social supervision, public services, and emergency capabilities. Simultaneously, it also brings great challenges in management technology for big spatial vector data (BSVD). In recent years, a large number of new concepts, parallel algorithms, processing tools, platforms, and applications have been proposed and developed to improve the value of BSVD from both academia and industry. To better understand BSVD and take advantage of its value effectively, this paper presents a review that surveys recent studies and research work in the data management field for BSVD. In this paper, we discuss and itemize this topic from three aspects according to different information technical levels of big spatial vector data management. It aims to help interested readers to learn about the latest research advances and choose the most suitable big data technologies and approaches depending on their system architectures. To support them more fully, firstly, we identify new concepts and ideas from numerous scholars about geographic information system to focus on BSVD scope in the big data era. Then, we conclude systematically not only the most recent published literatures but also a global view of main spatial technologies of BSVD, including data storage and organization, spatial index, processing methods, and spatial analysis. Finally, based on the above commentary and related work, several opportunities and challenges are listed as the future research interests and directions for reference.

Due to the limitation of optical images that their waves cannot penetrate clouds, such images always suffer from cloud contamination, which causes missing information and limitations for subsequent agricultural applications, among others. Synthetic aperture radar (SAR) is able to provide surface information for all times and all weather. Therefore, translating SAR or fusing SAR and optical images to obtain cloud-free optical-like images are ideal ways to solve the cloud contamination issue. In this paper, we investigate the existing literature and provides two kinds of taxonomies, one based on the type of input and the other on the method used. Meanwhile, in this paper, we analyze the advantages and disadvantages while using different data as input. In the last section, we discuss the limitations of these current methods and propose several possible directions for future studies in this field.

Bibliometric analysis based on the Science Citation Index Expanded published by Thomson Scientific was carried out to identify the research status and future trends of remote sensing (RS) during 2010–2015. The analysis revealed the institutional, national, spatio-temporal, and categorical patterns in remote sensing research both from the WP (whole publications) viewpoint and the HCP (highly-cited publications) viewpoint. Statistical analysis results showed that remote sensing research almost doubled during 2010–2015. Environmental sciences comprised the most attractive subject category among remote sensing research. The International Journal of Remote Sensing was the most productive journal, and Remote Sensing of Environment published the most HCP among the 31 distributed journals. The productive ranking of countries was led by the U.S. both from the WP viewpoint and the HCP viewpoint, and CAS (Chinese Academy of Sciences) was the most productive institute both from the WP viewpoint and the HCP viewpoint with lower CPP (average number of citations per paper). Keyword analysis illustrated that model and algorithm research were the key points in RS during 2010–2015. RS data including Moderate-Resolution Imaging Spectroradiometer (MODIS), Landsat, synthetic aperture radar (SAR), and LiDAR (light detection and ranging) were the most frequently adopted, but the data usage of UAVs (unmanned aerial vehicles) and small satellites will be promoted in the future. With the development of data acquisition abilities, big data issues will become the challenges and hotspots of RS research, and new algorithms will continue to emerge.

Social media contains a lot of geographic information and has been one of the more important data sources for hazard mitigation. Compared with the traditional means of disaster-related geographic information collection methods, social media has the characteristics of real-time information provision and low cost. Due to the development of big data mining technologies, it is now easier to extract useful disaster-related geographic information from social media big data. Additionally, many researchers have used related technology to study social media for disaster mitigation. However, few researchers have considered the extraction of public emotions (especially fine-grained emotions) as an attribute of disaster-related geographic information to aid in disaster mitigation. Combined with the powerful spatio-temporal analysis capabilities of geographical information systems (GISs), the public emotional information contained in social media could help us to understand disasters in more detail than can be obtained from traditional methods. However, the social media data is quite complex and fragmented, both in terms of format and semantics, especially for Chinese social media. Therefore, a more efficient algorithm is needed. In this paper, we consider the earthquake that happened in Ya’an, China in 2013 as a case study and introduce the deep learning method to extract fine-grained public emotional information from Chinese social media big data to assist in disaster analysis. By combining this with other geographic information data (such population density distribution data, POI (point of interest) data, etc.), we can further assist in the assessment of affected populations, explore emotional movement law, and optimize disaster mitigation strategies.