Search Results (28)

Earth observation data serve as a fundamental resource in Earth system science. The rapid advancement of remote sensing and in situ measurement technologies has led to the generation of massive volumes of data, accompanied by a growing body of geographic textual information. Efficient and accurate classification and management of these geographic texts has become a critical challenge in the field. However, the effectiveness of traditional classification approaches is hindered by several issues, including data sparsity, class imbalance, semantic ambiguity, and the prevalence of domain-specific terminology. To address these limitations and enable the intelligent management of geographic information, this study proposes an efficient geographic text classification framework based on large language models (LLMs), tailored to the unique semantic and structural characteristics of geographic data. Specifically, LLM-based data augmentation strategies are employed to mitigate the scarcity of labeled data and class imbalance. A semantic vector database is utilized to filter the label space prior to inference, enhancing the model’s adaptability to diverse geographic terms. Furthermore, few-shot prompt learning guides LLMs in understanding domain-specific language, while an output alignment mechanism improves classification stability for complex descriptions. This approach offers a scalable solution for the automated semantic classification of geographic text for unlocking the potential of ever-expanding geospatial big data, thereby advancing intelligent information processing and knowledge discovery in the geospatial domain. Full article

Remote sensing images contain a wealth of Earth-observation information. Efficient extraction and application of hidden knowledge from these images will greatly promote the development of resource and environment monitoring, urban planning and other related fields. Remote sensing image caption (RSIC) involves obtaining textual descriptions from remote sensing images through accurately capturing and describing the semantic-level relationships between objects and attributes in the images. However, there is currently no comprehensive review summarizing the progress in RSIC based on deep learning. After defining the scope of the papers to be discussed and summarizing them all, the paper begins by providing a comprehensive review of the recent advancements in RSIC, covering six key aspects: encoder–decoder framework, attention mechanism, reinforcement learning, learning with auxiliary task, large visual language models and few-shot learning. Subsequently a brief explanation on the datasets and evaluation metrics for RSIC is given. Furthermore, we compare and analyze the results of the latest models and the pros and cons of different deep learning methods. Lastly, future directions of RSIC are suggested. The primary objective of this review is to offer researchers a more profound understanding of RSIC. Full article

Inter-satellite interferometric missions are critical in the ongoing monitoring of climate change. Next-generation Earth geodesy missions are opportunities to improve on mission cost and measurement sensitivity through revised design. To be considered feasible, mission architectures must meet an optical power requirement that factors in both shot noise and laser frequency noise. Reference-transponder mission configurations, like the Gravity Recovery and Climate Experiment-Follow On (GRACE-FO) mission, are designed for measurement down to a received carrier-to-noise density ratio of 70 dB-Hz—1.9 picowatts in shot-noise-limited detection. This work shows, through modeling and simulation, that the optical power level required to perform robust measurement varies significantly between mission configurations. Alternate configurations, such as retro-reflector-based schemes, can operate robustly down to much lower carrier-to-noise density ratios, with the example parameters considered here: down to 29 dB-Hz—150 attowatts in shot-noise-limited detection. These results motivate exploration of alternate missions configurations with revised optical power requirements, increasing the feasibility of new designs. Full article

In few-shot image classification (FSIC), the feature extraction module of the traditional convolutional neural networks is often constrained by the local nature of the convolutional kernel. As a result, it becomes challenging to handle global information and long-distance dependencies effectively. In order to address this problem, an innovative FSIC method is proposed in this paper, which is the integration of Swin Transformer and CSAM and Earth Mover’s Distance (EMD) technology (STCE). We utilize the Swin Transformer network for image feature extraction, and perform CSAM attention mechanism feature weighting on the output feature map, while we adopt the EMD algorithm to generate the optimal matching flow between the structural units, minimizing the matching cost. This approach allows for a more precise representation of the classification distance between images. We have conducted numerous experiments to validate the effectiveness of our algorithm. On three commonly used few-shot datasets, namely mini-ImageNet, tiered-ImageNet, and FC100, the accuracy of one-shot and five-shot has reached the state of the art (SOTA) in the FSIC; the mini-ImageNet achieves an accuracy of 98.65 ± 0.1% for one-shot and 99.6 ± 0.2% for five-shot tasks, while tiered ImageNet has an accuracy of 91.6 ± 0.1% for one-shot tasks and 96.55 ± 0.27% for five-shot tasks. For FC100, the accuracy is 64.1 ± 0.3% for one-shot tasks and 79.8 ± 0.69% for five-shot tasks. On two commonly used few-shot datasets, namely CUB, CIFAR-FS, CUB achieves an accuracy of 83.1 ± 0.4% for one-shot and 92.88 ± 0.4% for five-shot tasks, while CIFAR-FS achieves an accuracy of 86.95 ± 0.2% for one-shot and 94 ± 0.4% for five-shot tasks. Full article

Soil has long been one of the most widely used building materials globally. The evolution of soil-based construction materials has seen steady improvement over the centuries, even as traditional materials have given way to newer options like reinforced concrete. Nonetheless, soil-based construction has maintained its relevance and, in recent decades, has garnered increased attention due to sustainability concerns and renewed research interest. Among the innovative earth-based materials, shot-earth (SE) stands out as one of the most advanced. Research on SE has facilitated efficient handling of soil variability in mix design and provided structural engineers with relevant models for dimensioning and detailing reinforced SE constructions. This paper focuses on studying the durability characteristics of various types of SE to ascertain their ability to withstand environmental degradation over their intended lifespan. The tests conducted indicate that SE can serve as a viable construction material in numerous real-life scenarios, offering a sustainable alternative to existing materials. Full article

Change detection (CD) in remote sensing (RS) imagery is a pivotal method for detecting changes in the Earth’s surface, finding wide applications in urban planning, disaster management, and national security. Recently, deep learning (DL) has experienced explosive growth and, with its superior capabilities in feature learning and pattern recognition, it has introduced innovative approaches to CD. This review explores the latest techniques, applications, and challenges in DL-based CD, examining them through the lens of various learning paradigms, including fully supervised, semi-supervised, weakly supervised, and unsupervised. Initially, the review introduces the basic network architectures for CD methods using DL. Then, it provides a comprehensive analysis of CD methods under different learning paradigms, summarizing commonly used frameworks. Additionally, an overview of publicly available datasets for CD is offered. Finally, the review addresses the opportunities and challenges in the field, including: (a) incomplete supervised CD, encompassing semi-supervised and weakly supervised methods, which is still in its infancy and requires further in-depth investigation; (b) the potential of self-supervised learning, offering significant opportunities for Few-shot and One-shot Learning of CD; (c) the development of Foundation Models, with their multi-task adaptability, providing new perspectives and tools for CD; and (d) the expansion of data sources, presenting both opportunities and challenges for multimodal CD. These areas suggest promising directions for future research in CD. In conclusion, this review aims to assist researchers in gaining a comprehensive understanding of the CD field. Full article

The high performance of deep learning networks relies mainly on massive data. However, collecting enough samples of crop disease is impractical, which significantly limits the intelligent diagnosis of diseases. In this study, we propose Heterogeneous Metric Fusion Network-based Few-Shot Learning (HMFN-FSL), which aims to recognize crop diseases with unseen categories using only a small number of labeled samples. Specifically, CBAM (Convolutional Block Attention Module) was embedded in the feature encoders to improve the feature representation capability. Second, an improved few-shot learning network, namely HMFN-FSL, was built by fusing three metric networks (Prototypical Network, Matching Network, and DeepEMD (Differentiable Earth Mover’s Distance)) under the framework of meta-learning, which solves the problem of the insufficient accuracy of a single metric model. Finally, pre-training and meta-training strategies were optimized to improve the ability to generalize to new tasks in meta-testing. In this study, two datasets named Plantvillage and Field-PV (covering 38 categories of 14 crops and containing 50,403 and 665 images, respectively) are used for extensive comparison and ablation experiments. The results show that the HMFN-FSL proposed in this study outperforms the original metric networks and other state-of-the-art FSL methods. HMFN-FSL achieves 91.21% and 98.29% accuracy for crop disease recognition on 5way-1shot, 5way-5shot tasks on the Plantvillage dataset. The accuracy is improved by 14.86% and 3.96%, respectively, compared to the state-of-the-art method (DeepEMD) in past work. Furthermore, HMFN-FSL was still robust on the field scenes dataset (Field-PV), with average recognition accuracies of 73.80% and 85.86% on 5way-1shot, 5way-5shot tasks, respectively. In addition, domain variation and fine granularity directly affect the performance of the model. In conclusion, the few-shot method proposed in this study for crop disease recognition not only has superior performance in laboratory scenes but is also still effective in field scenes. Our results outperform the existing related works. This study provided technical references for subsequent few-shot disease recognition in complex environments in field environments. Full article

Environmental pollution exists not only within our atmosphere but also in space. Space debris is a critical problem of modern and future space infrastructure. Congested orbits raise the question of spacecraft disposal. Therefore, state-of-the-art satellites come with a deorbit system in cases of low Earth orbit (LEO) and with thrusters for transferring into the graveyard orbit for geostationary and geosynchronous orbits. No practical solution is available for debris objects that stem from fragmentation events. The present study focuses on objects in LEO orbits with dimensions in the dangerous class of 1 to 10 cm. Our assumed method for the change of trajectories of space debris is laser ablation for collision avoidance or complete removal by ground-based laser systems. Thus, we executed an experimental feasibility study with focus on thermal and impulse coupling between laser and sample. Free-fall experiments with a 10 ns laser pulse at nominally 60 J and 1064 nm were conducted with GSI Darmstadt’s nhelix laser on various sample materials with different surfaces. Ablated mass, heating, and trajectory were recorded. Furthermore, we investigated the influence of the sample surface roughness on the laser-object interaction. We measured impulse coupling coefficients between 7 and 40 µNs/J and thermal coupling coefficients between 2% and 12.5% both depending on target fluence, surface roughness, and material. Ablated mass and changes in surface roughness were considered via simulation to discriminate their relevance for a multiple shot concept. Full article

The rapid development of Earth observation technology has promoted the continuous accumulation of images in the field of remote sensing. However, a large number of remote sensing images still lack manual annotations of objects, which makes the strongly supervised deep learning object detection method not widely used, as it lacks generalization ability for unseen object categories. Considering the above problems, this study proposes a few-shot remote sensing image object detection method that integrates context dependencies and global features. The method can be used to fine-tune the model with a small number of sample annotations based on the model trained in the base class, as a way to enhance the detection capability of new object classes. The method proposed in this study consists of three main modules, namely, the meta-feature extractor (ME), reweighting module (RM), and feature fusion module (FFM). These three modules are respectively used to enhance the context dependencies of the query set features, improve the global features of the support set that contains annotations, and finally fuse the query set features and support set features. The baseline of the meta-feature extractor of the entire framework is based on the optimized YOLOv5 framework. The reweighting module of the support set feature extraction is based on a simple convolutional neural network (CNN) framework, and the foreground feature enhancement of the support sets was made in the preprocessing stage. This study achieved beneficial results in the two benchmark datasets NWPU VHR-10 and DIOR. Compared with the comparison methods, the proposed method achieved the best performance in the object detection of the base class and the novel class. Full article

Clouds are a significant factor in regional climates and play a crucial role in regulating the Earth’s water cycle through the interaction of sunlight and wind. Meteorological agencies around the world must regularly observe and record cloud data. Unfortunately, the current methods for collecting cloud data mainly rely on manual observation. This paper presents a novel approach to identifying ground-based cloud images to aid in the collection of cloud data. However, there is currently no publicly available dataset that is suitable for this research. To solve this, we built a dataset of surface-shot images of clouds called the SSC, which was overseen by the Macao Meteorological Society. Compared to previous datasets, the SSC dataset offers a more balanced distribution of data samples across various cloud genera and provides a more precise classification of cloud genera. This paper presents a method for identifying cloud genera based on cloud texture, using convolutional neural networks. To extract cloud texture effectively, we apply Gamma Correction to the images. The experiments were conducted on the SSC dataset. The results show that the proposed model performs well in identifying 10 cloud genera, achieving an accuracy rate of 80% for the top three possibilities. Full article

The $GW$ method is a standard method to calculate the electronic band structure from first principles. It has been applied to a large variety of semiconductors and insulators but less often to metallic systems, in particular, with respect to a self-consistent employment of the method. In this work, we take a look at all-electron quasiparticle self-consistent $GW$ (QS$GW$) calculations for simple metals (alkali and alkaline earth metals) based on the full-potential linearized augmented-plane-wave approach and compare the results to single-shot (i.e., non-selfconsistent) $G_0{W}_0$ calculations, density-functional theory (DFT) calculations in the local-density approximation, and experimental measurements. We show that, while DFT overestimates the bandwidth of most of the materials, the $GW$ quasiparticle renormalization corrects the bandwidths in the right direction, but a full self-consistent calculation is needed to consistently achieve good agreement with photoemission data. The results mainly confirm the common belief that simple metals can be regarded as nearly free electron gases with weak electronic correlation. The finding is particularly important in light of a recent debate in which this seemingly established view has been contested. Full article

Using deep learning-based object detection algorithms for landslide hazards detection is very popular and effective. However, most existing algorithms are designed for landslides in a specific geographical range. This paper constructs a set of landslide detection models YOLOX-Pro, based on the improved YOLOX (You Only Look Once) target detection model to address the poor detection of complex mixed landslides. Wherein the VariFocal is used to replace the binary cross entropy in the original classification loss function to solve the uneven distribution of landslide samples and improve the detection recall; the coordinate attention (CA) mechanism is added to enhance the detection accuracy. Firstly, 1200 historical landslide optical remote sensing images in thirty-eight areas of China were extracted from Google Earth to create a mixed sample set for landslide detection. Next, the three attention mechanisms were compared to form the YOLOX-Pro model. Then, we tested the performance of YOLOX-Pro by comparing it with four models: YOLOX, YOLOv5, Faster R-CNN, and Single Shot MultiBox Detector (SSD). The results show that the YOLOX-Pro(m) has significantly improved the detection accuracy of complex and small landslides than the other models, with an average precision (AP0.75) of 51.5%, APsmall of 36.50%, and ARsmall of 49.50%. In addition, optical remote sensing images of a 12.32 ${\mathrm{km}}^2$ group-occurring landslides area located in Mibei village, Longchuan County, Guangdong, China, and 750 Unmanned Aerial Vehicle (UAV) images collected from the Internet were also used for landslide detection. The research results proved that the proposed method has strong generalization and good detection performance for many types of landslides, which provide a technical reference for the broad application of landslide detection using UAV. Full article

The Global Ecosystem Dynamics Investigation (GEDI), specifically designed to measure vertical forest structures, has acquired, since April 2019, more than 35 billion waveforms of Earth’s surface on a nearly global scale. GEDI is equipped with 3 identical 1064 nm lasers with a power of 10 mJ per shot, where 1 laser is split into 2 lasers, resulting in two 5 mJ coverage lasers and two 10 mJ full-power lasers. In this study, we evaluate the potential of GEDI’s four lasers to penetrate through canopies and detect the ground, and their capabilities to detect the top of the canopies over a tropical forest (in French Guiana) characterized by a dense canopy cover and tall trees. The accurate detection of both of these surfaces is the first step in characterizing vertical forest structures. The SRTM Digital Elevation Model (DEM) is used as a reference point for elevations while a canopy height model (CHM), derived from airborne and spaceborne LiDAR data, is used as a reference for canopy heights. In addition, the ground and canopy-top elevations estimated from NASA’s Land, Vegetation, and Ice Sensor (LVIS, 1064 nm full-waveform LiDAR, 5 mJ per shot, ~8 km altitude) are used as a benchmark for comparison with GEDI’s lasers. Results indicate that GEDI’s coverage and full-power lasers, even after the application of a preliminary filter that removes around 50% of acquisitions, tend to underestimate tree heights in densely vegetated, tall forests. Moreover, GEDI’s coverage lasers also exhibited a lower level of performance in comparison to both the full-power lasers and LVIS. Overall, the average estimated maximum canopy heights (RH₁₀₀) for a CHM greater than 30 m was 24.4 m with the coverage lasers, 32.1 m with the full-power lasers, and 36.7 m with LVIS. The analysis of shots with high-beam sensitivity (sensitivity ≥ 98%) showed that they tend to have a better probability of reaching the ground and have better detection of canopy tops for both GEDI laser types. Nonetheless, GEDI’s coverage lasers still showed an underestimation of canopy heights with an average RH₁₀₀ of 29.8 m, while for GEDI’s full-power lasers and LVIS, the average RH₁₀₀ was 35.2 m and 37.7 m, respectively. Finally, the assessment of the acquisition time on the detection of the ground return and the top of the canopies showed that, for the coverage lasers, solar noise could affect the detection of the ground return as acquisitions made during early mornings or late afternoons have more penetration than shots acquired between 8 a.m. and 4 p.m. The effect of acquisition time on the detection of the tops of canopies showed that solar noise slightly affected the coverage lasers. Regarding the full-power lasers, the acquisition time of the shots seem to affect neither the penetration of the lasers, nor the detection of the tops of canopies. Full article

The Global Ecosystem Dynamics Investigation (GEDI) LiDAR on the International Space Station has acquired more than 35 billion shots globally in the period between April 2019 and August 2021. The acquired shots could offer a significant database for the measure and monitoring of inland water levels over the Earth’s surface. Nonetheless, previous and current studies have shown that the provided GEDI elevation estimates are significantly less accurate than any available radar or LiDAR altimeter. Indeed, our analysis of GEDI’s altimetric capabilities to retrieve water levels over the five North American Great Lakes presented estimates with a bias that ranged between 0.26 and 0.35 m and a root mean squared error (RMSE) ranging between 0.54 and 0.68 m. Therefore, our objective in this study is to post-process the original GEDI water level estimates from an error model taking instrumental, atmospheric, and lakes surface state factors as proxies, which affect the physical shape of the waveforms, hence introducing uncertainties on the elevation estimates. The first tested model, namely a random forest regressor ($R{F}_{ICW}$) with the instrumental, atmospheric, and water surface state factors as inputs, was validated temporally (trained on a given year and validated on another) and spatially (trained on a given lake and validated on the remaining four). The results showed a significant decrease in elevation estimation errors both temporally and spatially. The temporally validated models showed an RMSE on the corrected elevation estimates of 0.18 m. Concerning the spatially validated model, the results varied based on the lake data used for training. Indeed, the most accurate spatially validated model showed an RMSE of 0.17 m, while the least accurate model showed an RMSE of 0.26 m. Finally, given that an elevation correction model using all the factors (instrumental, atmospheric, and water surface state factors) presents a best-case scenario, as water surface state factors are only available over a selected number of lakes globally, three additional models based on random forest were tested. The first,$R{F}_I$, uses only instrumental factors as correction factors, $R{F}_{IC}$ uses both instrumental and atmospheric factors, while the third, $R{F}_{IW}$, uses instrumental and water surface state factors. The temporal validation of these models showed that the model using instrumental factors, while less accurate than the remaining two models, was capable of correcting the original GEDI elevation estimates by a factor of two across the five lakes. On the other hand, the $R{F}_{IC}$ model was the most accurate between the three, with a slight degradation in comparison to the full model. Indeed, the $R{F}_{IC}$ model showed an RMSE on the estimation of water levels of 0.21 m. Full article

The ice, cloud, and land elevation satellite-2 (ICESat-2) is providing global elevation measurements to the science community. ICESat-2 measures the height of the Earth’s surface using a photon counting laser altimeter, ATLAS (advanced topographic laser altimetry system). As a photon counting system, the number of reflected photons per shot, or radiometry, is a function primarily of the transmitted laser energy, solar elevation, surface reflectance, and atmospheric scattering and attenuation. In this paper, we explore the relationship between detected scattering and attenuation in the atmosphere against the observed radiometry for three general forest types, as well as the radiometry as a function of day versus night. Through this analysis, we found that ATLAS strong beam radiometry exceeds the pre-launch design cases for boreal and tropical forests but underestimates the predicted radiometry over temperate forests by approximately half a photon. The weak beams, in contrast, exceed all pre-launch conditions by a factor of two to six over all forest types. We also observe that the signal radiometry from day acquisitions is lower than night acquisitions by 10% and 40% for the strong and weak beams, respectively. This research also found that the detection ratio between each beam-pair was lower than the predicted 4:1 values. This research also presents the concept of ICESat-2 radiometric profiles; these profiles provide a path for calculating vegetation structure. The results from this study are intended to be informative and perhaps serve as a benchmark for filtering or analysis of the ATL08 data products over vegetated surfaces. Full article