Flood segmentation using synthetic aperture radar (SAR) images is essential for determining the extent of inundation areas, and informing subsequent management recommendations. However, existing networks for flood segmentation using single modality SAR images often face inherent challenges, including interference from terrain shadows and water-like surfaces, leading to degraded segmentation performance. In this study, we introduced a multi-level multi-modal fusion network (FloodNet), in which an Adaptive Gated Feature Fusion Module (AGFFM) is designed to integrate multi-modal features from Sentinel-1 SAR images, Digital Elevation Model (DEM) and Joint Research Centre Global Surface Water (JRC-gsw). Furthermore, we proposed a semantic consistency constraint strategy to alleviate the blurring of water edges during the prediction process. Experiments on two publicly available flood datasets, C2S-Flood and ETCI-Flood, demonstrate the competitive performance of the proposed FloodNet compared with other state-of-the-art single- and multi-modal networks. The code is available at https://github.com/SuperPixelPioneer/Flood-Net.

Nighttime light (NTL) data holds irreplaceable value in research related to human activities, sustainable development, etc. However, the temporal and spatial continuity of high-resolution NTL data is challenging to meet the demands of large-scale applications. Spatiotemporal data fusion methods, by integrating low-resolution and high-resolution data, can fill in missing high-resolution data. Nevertheless, previous spatiotemporal data fusion research has primarily focused on multispectral data and face challenges when directly applied to NTL data. This research proposes a spatiotemporal fusion method specifically for NTL data, named the "LIght ON spatiotemporal data fusion" (LION) model. Experimental results indicate that LION demonstrated potential in predicting abrupt changes in NTL.

The Universal Thermal Climate Index (UTCI) is a crucial temperature index for describing human thermal comfort. With the continuous advancement of earth observation technologies, it has become feasible to monitor hourly global UTCI at kilometer-level resolution. However, the computational efficiency of UTCI calculations limits the production and application of UTCI, particularly time-series UTCI application at fine resolution. To address the abovementioned issue, this letter proposes a CUDA UTCI (CUTCI) method based on the graphics processing unit (GPU). CUTCI leverages the parallel computing capabilities of GPUs and kernel fusion techniques to improve parallelization and mitigate overhead during calculation. Experimental results demonstrated that, in comparison to operational UTCI and Thermofeel UTCI, CUTCI significantly improved computational efficiency by over 250 times and 17 times, respectively. Production of a single-period global UTCI at 0.1° resolution consumed less than 0.2 seconds. Experimental results revealed that CUTCI holds practical value in supporting real-time UTCI analysis and historical big data analysis over long time series.

Spatiotemporal data fusion provides an efficacious strategy for addressing data gaps within time series datasets. This approach significantly enhances the feasibility of large-scale remote sensing applications by, for example, enabling the creation of seamless Data Cubes (SDC). Nevertheless, strict data input requirements and low computational efficiency of current methods severely limit the practicality of large-scale SDC production. In this study, we propose an efficient spatiotemporal data fusion method, the Fast Variation-based Spatiotemporal Data Fusion (FastVSDF) method. FastVSDF consists of 3 steps, i.e., unmixing, distributing global residuals, and distributing local residuals. In the unmixing process, FastVSDF introduces the fast abundant variation classification (FAVC) to mitigate sample imbalance and expedite the unsupervised classification. Then, the in-class Gaussian weight function is introduced to accelerate the distribution of local residuals by considering the classification to introduce the information on spectral similarity. Besides, FastVSDF employs Fast Guided Filter to combat the "block artifacts" of global residuals efficiently. Results show that FastVSDF demonstrated superior performance over Fit-FC, STARFM, RASDF, and FSDAF. More importantly, FastVSDF yields a remarkable improvement in computational efficiency, reducing predicting time by 43 to 573 times. As a practical application, we generated the Sentinel-2 SDC for the Yangtze River Basin, China. The fusion process for a single period’s Yangtze River Basin dataset was accomplished within 20 minutes, with an average of 3.85 seconds for each Sentinel-2 scene. Comprehensively considering the efficiency, accuracy, feasibility, and universality, FastVSDF demonstrates the practical potential for constructing large-scale and long-term SDC. Our code will be publicly available at https://github.com/ChenXuAxel/FastVSDF.

Personnel and boat detection in Unmanned Aerial Vehicles (UAVs) imagery plays a crucial role in Open Water Search and Rescue Missions. The diverse perspectives and altitudes of UAV images often result in significant variations in the imagery’s appearance and dimensions of personnel and boats, and the false detections arising from water surface flares are acknowledged as a great challenge as well. Existing deep learning-based detection methods employ convolutional blocks with fixed kernel sizes to extract features from the imagery at a fixed spatial scale, which will lead to missed and false detections, and severely affect detection accuracy when there are substantial differences in the appearance and size of the target objects. In this paper, a spatial scale adaptive real-time object detection neural network, namely YoloOW, was proposed to tackle the challenge of personnel and boat detection amidst the diverse UAV imagery, which comprises a feature extractor, a feature enhancer, and a postprocessor. The OaohRep convolutional block was proposed as a pivotal component in constructing the YoloOW and applied to the feature extractor and the feature enhancer. Compared with general convolution blocks, the OaohRep convolution block can extract image features across a wide range of spatial scales, show better scale adaptability, and achieve faster detection speed due to its unique merged convolution layer design. OaohRepBi-PAN was proposed in the feature enhancer, which imitated the architecture of the classic algorithm SIFT and was successfully applied to deep learning models, showing better scale adaptability. A novel UAV detection box filter (UDBF) module was proposed in the postprocessor, which can effectively remove false detections caused by water surface flares. Experimental results demonstrate that our YoloOW model achieves 37.18% mAP on the SeaDronesSee dataset, surpassing the baseline by 8.43%. This notable improvement positions our model at the first of the leaderboard. The code will be available at https://github.com/Xjh-UCAS/YoloOW.

Tiled remote sensing data is essential for web-based remote sensing applications, widely applied in online map services, cloud-based remote sensing processing, etc. However, most existing tiling algorithms focus on tiling single remote sensing images with stand-alone machines. As the volume of remote sensing data increases, the demand for tiling high-resolution and large-scale remote sensing datasets increases dramatically. In this research, we propose a cloud-based parallel tiling algorithm for large-scale remote sensing datasets. A three-step processing flow is designed to implement the tiling of datasets composed of a set of images. Furthermore, three types of cloud-based storage are adopted to improve the efficiency of data extraction, namely, cloud storage, block storage, and NoSQL. We experimented with the proposed algorithm for tiling a national-scale 2 m resolution remote sensing dataset. The whole process took about 25.7 hours with up to 180 cores.

Today, remote sensing (RS) data are already regarded as “big data.” Developments in computer science have made it possible to explore the potential treasure within remote sensing big data, but only limited remote sensing research has made use of big data technology due to gaps in techniques between big data and remote sensing. In this research, we analyzed the full processing flow of remote sensing big data from the perspective of both computer science and remote sensing science and proposed a modular framework. Computation ready data (CRD), a dynamic data type for computation based on analysis ready data (ARD), is proposed to connect the two main modules of the framework, the data module and computation module. Compared with existing research, the proposed framework classifies and abstracts the key technical and research points of the processing of remote sensing big data as replaceable modules and bridges them through an open organization. Subsequently, we built a prototype platform with open-source technologies and carried out three experiments to validate the feasibility and advantages of the framework, namely normalized difference vegetation index (NDVI) production, water body change detection, and land use classification. Results indicate that this framework can greatly reduce experimental costs for remote sensing researchers. While the proposed framework has proven flexible and practical, further research is needed for the technical implementation of certain modules to achieve the original intention of the framework.