Analysis Of Remote Sensing Data Research Articles

Multi-Scale Fusion Siamese Network Based on Three-Branch Attention Mechanism for High-Resolution Remote Sensing Image Change Detection

Remote sensing image change detection (CD) is an important means in remote sensing data analysis tasks, which can help us understand the surface changes in high-resolution (HR) remote sensing images. Traditional pixel-based and object-based methods are only suitable for low- and medium-resolution images, and are still challenging for complex texture features and detailed image detail processing in HR images. At present, the method based on deep learning has problems such as inconsistent fusion and difficult model training in the combination of the difference feature information of the deep and shallow layers and the attention mechanism, which leads to errors in the distinction between the changing region and the invariant region, edge detection and small target detection. In order to solve the above problems of inconsistent fusions of feature information aggregation and attention mechanisms, and indistinguishable change areas, we propose a multi-scale feature fusion Siamese network based on attention mechanism (ABMFNet). To tackle the issues of inconsistent fusion and alignment difficulties when integrating multi-scale fusion and attention mechanisms, we introduce the attention-based multi-scale feature fusion module (AMFFM). This module not only addresses insufficient feature fusion and connection between different-scale feature layers, but also enables the model to automatically learn and prioritize important features or regions in the image. Additionally, we design the cross-scale fusion module (CFM) and the difference feature enhancement pyramid structure (DEFPN) to assist the AMFFM module in integrating differential information effectively. These modules bridge the spatial disparity between low-level and high-level features, ensuring efficient connection and fusion of spatial difference information. Furthermore, we enhance the representation and inference speed of the feature pyramid by incorporating a feature enhancement module (FEM) into DEFPN. Finally, the BICD dataset proposed by the laboratory and public datasets LEVIR-CD and BCDD are compared and tested. We use F1 score and MIoU values as evaluation metrics. For AMBMFNet, the F1 scores on the three datasets are 77.69%, 81.57%, and 77.91%, respectively, while the MIoU values are 84.65%, 85.84%, and 84.54%, respectively. The experimental results show that ABMFNet has better effectiveness and robustness.

Storage and Analysis of Remote Sensing Data

Storage and Analysis of Remote Sensing Data

Quantifying the Influence of Climate Variables on Vegetation Through Remote Sensing and Multi-dimensional Data Analysis

Quantifying the Influence of Climate Variables on Vegetation Through Remote Sensing and Multi-dimensional Data Analysis

S3L: Spectrum Transformer for Self-Supervised Learning in Hyperspectral Image Classification

In the realm of Earth observation and remote sensing data analysis, the advancement of hyperspectral imaging (HSI) classification technology is of paramount importance. Nevertheless, the intricate nature of hyperspectral data, coupled with the scarcity of labeled data, presents significant challenges in this domain. To mitigate these issues, we introduce a self-supervised learning algorithm predicated on a spectral transformer for HSI classification under conditions of limited labeled data, with the objective of enhancing the efficacy of HSI classification. The S3L algorithm operates in two distinct phases: pretraining and fine-tuning. During the pretraining phase, the algorithm learns the spatial representation of HSI from unlabeled data, utilizing a masking mechanism and a spectral transformer, thereby augmenting the sequence dependence of spectral features. Subsequently, in the fine-tuning phase, labeled data is employed to refine the pretrained weights, thereby improving the precision of HSI classification. Within the comprehensive encoder–decoder framework, we propose a novel spectral transformer module specifically engineered to synergize spatial feature extraction with spectral domain analysis. This innovative module adeptly navigates the complex interplay among various spectral bands, capturing both global and sequential spectral dependencies. Uniquely, it incorporates a gated recurrent unit (GRU) layer within the encoder to enhance its ability to process spectral sequences. Our experimental evaluations across several public datasets reveal that our proposed method, distinguished by its spectral transformer, achieves superior classification performance, particularly in scenarios with limited labeled samples, outperforming existing state-of-the-art approaches.

A fast and robust cirrus removal method for Landsat 8/9 images

High-quality cirrus removal plays a crucial role in remote sensing data analysis. Cirrus parallaxes are commonly observed within the vicinity of cirrus clouds in the visible and near-infrared (VNIR) bands of Landsat 8/9 images. Cirrus parallaxes have a nonnegligible effect on cirrus removal, but the existing methods do not account for the correction of parallaxes. Meanwhile, large-scale image processing involves intensive computation that requires extensive computing time. To address the effect of cirrus parallaxes and the low processing efficiency, we propose a fast and robust cirrus removal (FRCR) method. FRCR has achieved the first realization of the statistics law of cirrus parallax between the VNIR and cirrus bands, thus realizing the cirrus parallax correction. In addition, FRCR introduces an automatic sampling method to obtain the regression samples for practicality. Then, a Compute Unified Device Architecture (CUDA) based Newton method with constraints is introduced to parallelize the computation, to improve the computational performance. Experiment results of various scenarios demonstrate that the FRCR method can achieve high-quality cirrus removal by eliminating cirrus parallaxes, and significantly improving computational performance.

Analysis of Surface Runoff and Remote Sensing Data to Identify Flood Potential in Simbang Sub-Watershed

Increased water runoff due to land use change phenomena has triggered flooding events. The objective is to identify flood potential in Simbang Sub-Watershed in Maros Regency using rational methods and remote sensing data. Potential flood hazards were analyzed using the weighting method with the parameters NDVI, MNDWI, NDSI, rainfall, and annual flow coefficient. The highest debit of runoff occurred in 2017, with a value of 113.36 m3/s, while the lowest occurred in 2019, with a value of 63.91 m3/s. The NDVI value is 0.37–1 with high vegetation covering an area of 3,089 ha, while the low-very low vegetation has value -0.03–0.25 with an area of 1,668 ha. The MNDWI value ranges from 0-0.33 with a moderate wetness level covering an area of 741 ha and an NDSI value ranging from -0.06–0.43 for bare land surrounding an area of 738 ha, which has an impact on reducing water catchment areas which can trigger an increase in surface water runoff discharge. The average rainfall is 2,965 mm/year, the area with low potential for flooding is 3,705 ha, and the area prone (moderate) to flooding is 1,450 ha. The rainfall factor is the main priority trigger for flood events with weight of 0.266, and the soil index is the lowest priority factor with weight of 0.145. Surface water runoff in the Simbang Sub-Watershed area makes a small contribution to the flood events that occurred in Maros Regency with an annual flow coefficient value of 0.23.

Automation of digital elevation model analyzing when solving problems of economic development of karst territories

The combination of geographic information systems (GIS) and remote sensing data analysis tools will make it possible
 to assess in a short time and with high accuracy the suitability of an area for the construction of land or linear structures in the
 areas where dangerous geological processes are developed. In this article we use the example of karst area in Kisherstky district
 of Perm region as an example to carry out the karts hazard zoning of the territory according to geomorphological indicators by
 analyzing ALOS-DEM digital elevation model (DEM). Patterns in the distribution of karst forms in the study area were evaluated
 depending on the dissection of the relief, its slope and distance from the watercourses. Based on the patterns, an integral map
 was constructed. The approach allows to do zoning the territory with respect to the studied danger geological process in a short
 time and with high accuracy, which is due to the use of patterns of forms distribution in the analysis.

ِA new occurrence of rift-related damtjernite (ultramafic) lamprophyre, Gebel Anweiyib area, Arabian Nubian shield: Insights from bulk rock geochemistry and remote sensing data analysis

The occurrence of lamprophyre dikes is being reported first time in the Gebel Anweiyib area, south Eastern Desert, Egypt. The investigation of the ultramafic lamprophyre (damtjernite) dikes intruded into the crustal younger granites and metasediments based on their spectral characteristics, mineralogical and geochemical compositions. Three multispectral sensors of Landsat-8, ASTER and S2 have been used to detect and characterize the lithologic units and the lamprophyre dikes. The reflectance and absorption signatures of the granitic rocks and lamprophyre dikes were the basis for detecting them from the remotely sensed data using the application of advanced image processing algorithms including FCC, BR, PCA, MNF, and DS. Geochemical studies inferred that the damtjernite dikes are under-saturated with SiO2 (<38.89 wt%), alkaline with total alkalis (avg. 4 wt%), metaluminous with A/CNK (0.22–0.51), and ultramafic with MgO (13.88–19.23 wt%). They exhibit unique geochemical characteristics of mantle (with high MgO, Ni, Cr, and Mg#) coupled with crustal (high Sr, Ba, Zr, Ga and REEs) sources, similar to global ultramafic lamprophyres. High TiO2 and highly fractionated REEs with low HREEs (avg. 23 ppm) relative to LREEs (avg. 354 ppm) contents, yielding residual garnet in melts. Low partial melting of primitive garnet lherzolite can account the obtained REEs in the ultramafic lamprophyres. These features signify their derivation from a deeper enriched asthenospheric (plume-type) magma like OIB, which is attested by the high Nb/La (avg. 1.32), Nb/U (avg. 102), La/Yb (avg. 43), and low Y/Nb < 0.52, Rb/Sr < 1, Rb/Ba (<0.14), and Sr/Ba (<0.9), suggesting their derivation from a mantle-derived source rather than crustal rocks. High concentrations of TiO2 and Fe2O3 in the ultramafic lamprophyres, reflecting their formation from a fertile peridotite by mantle plume-derived melts with K-residual phlogopite rather than amphibole as obtained by low Ba/Rb and high Rb/Sr ratios. The Anweiyib lamprophyres are anorogenic and confined to being derived from rifting regions, and their magma is not contaminated with crustal sources during emplacement.

Multi-spectral remote sensing for current irrigated area mapping of the Rift Valley Lakes Basin in Ethiopia

Abstract The study was conducted in the Rift Valley Basin in 2020–2022, with the objective of evaluating the newly developed techniques for irrigated area mapping of spatially large areas and assessing the current irrigated area in the basin. Methods used for irrigated area mapping are imagery analysis using the NDVI and EVI methods and land use classification. All results were verified with ground-truthing data using the sample locations selected. A 10 m × 10 m spatial resolution image derived from the European Space Agency Sentinel-2 satellite was used to create a time series spanning 2020, 2021, and 2022. Irrigated area maps from all three techniques were obtained and evaluated. Metric indicators were used to evaluate the performance of the irrigated area mapping techniques, the mean overall accuracy was 0.82, with a kappa coefficient of 0.76 and an F1-score of 0.86 with the highest and lowest overall accuracy observed in 2020 (0.86) and 2022 (0.76), respectively. Coefficient of determination was used to quantify the correlation of geospatial information and the multispectral remote sensing data analysis results maps agreed with irrigated area mapping with respect to ground truthing with R2 mean value of 0.87 which suggests a strong agreement.

A Generic Self-Supervised Learning (SSL) Framework for Representation Learning from Spectral–Spatial Features of Unlabeled Remote Sensing Imagery

Remote sensing data has been widely used for various Earth Observation (EO) missions such as land use and cover classification, weather forecasting, agricultural management, and environmental monitoring. Most existing remote-sensing-data-based models are based on supervised learning that requires large and representative human-labeled data for model training, which is costly and time-consuming. The recent introduction of self-supervised learning (SSL) enables models to learn a representation from orders of magnitude more unlabeled data. The success of SSL is heavily dependent on a pre-designed pretext task, which introduces an inductive bias into the model from a large amount of unlabeled data. Since remote sensing imagery has rich spectral information beyond the standard RGB color space, it may not be straightforward to extend to the multi/hyperspectral domain the pretext tasks established in computer vision based on RGB images. To address this challenge, this work proposed a generic self-supervised learning framework based on remote sensing data at both the object and pixel levels. The method contains two novel pretext tasks, one for object-based and one for pixel-based remote sensing data analysis methods. One pretext task is used to reconstruct the spectral profile from the masked data, which can be used to extract a representation of pixel information and improve the performance of downstream tasks associated with pixel-based analysis. The second pretext task is used to identify objects from multiple views of the same object in multispectral data, which can be used to extract a representation and improve the performance of downstream tasks associated with object-based analysis. The results of two typical downstream task evaluation exercises (a multilabel land cover classification task on Sentinel-2 multispectral datasets and a ground soil parameter retrieval task on hyperspectral datasets) demonstrate that the proposed SSL method learns a target representation that covers both spatial and spectral information from massive unlabeled data. A comparison with currently available SSL methods shows that the proposed method, which emphasizes both spectral and spatial features, outperforms existing SSL methods on multi- and hyperspectral remote sensing datasets. We believe that this approach has the potential to be effective in a wider range of remote sensing applications and we will explore its utility in more remote sensing applications in the future.

Analysis of radar images of the Sentinel-1 satellite using the EO Browser web resource

The article's purpose is to review the capabilities of the EO Browser web resource and tools for remote sensing data analysis and visualization. Examples of satellite images obtained using a synthetic aperture radar are given. The presented results were obtained during the processing of radar images of the Sentinel-1 satellite, which are freely available thanks to the implementation of the Copernicus program concept regarding the public availability of Earth remote sensing data. The results of satellite radar acquisition are used to prevent, preserve, and forecast natural or technogenic threats and risks, considering the high frequency of acquisition and the promptness of obtained data. The effectiveness of using the web resource in solving the environmental monitoring tasks in seawater areas, identifying accident sites, determining the area of oil spills, planning measures to eliminate the consequences and reduce environmental risks, and collecting the evidence base about the event is shown. The mechanism of reflection of the radar signal from various natural surfaces (including oil spills into the sea) and artificial objects (for detecting the position of sea vessels) is considered. A comparison of the effectiveness of using radar images of the Sentinel-1 satellite with optical images taken by the Sentinel-2 satellite for monitoring the condition of forest areas was performed. Satellite radar images are a reliable data source for continuously monitoring unregulated woodland destruction. The results' practical significance lies in using EO Browser to solve a wide range of scientific, practical, and educational problems using radar images. Further research using satellite radar images and the EO Browser platform is related to environmental monitoring in marine water areas. This applies to issues related to the detection of accumulations of garbage and plastic on water surfaces. It is promising to use the potential of the web resource for position designing and rapid determination of the effectiveness of deploying ground corner reflectors during geodynamic monitoring of the earth's surface. Keywords: synthetic aperture radar, remote sensing, environmental monitoring

Topological defects law for migrating banded vegetation patterns in arid climates.

Self-organization and pattern formation are ubiquitous processes in nature. We study the properties of migrating banded vegetation patterns in arid landscapes, usually presenting dislocation topological defects. Vegetation patterns with dislocations are investigated in three different ecosystems. We show through remote sensing data analysis and theoretical modeling that the number of dislocations N(x) decreases in space according to the law N ∼ log(x/B)/x, where x is the coordinate in the opposite direction to the water flow and B is a suitable constant. A sloped topography explains the origin of banded vegetation patterns with permanent dislocations. Theoretically, we considered well-established approaches to describe vegetation patterns. All the models support the law. This contrasts with the common belief that the dynamics of dislocations are transient. In addition, regimes with a constant distribution of defects in space are predicted. We analyze the different regimes depending on the aridity level and water flow speed. The reported decay law of defects can warn of imminent ecosystem collapse.

An attention-based multiscale transformer network for remote sensing image change detection

The bi-temporal change detection (CD) is still challenging for high-resolution optical remote sensing data analysis due to various factors such as complex textures, seasonal variations, climate changes, and new requirements. We propose an attention-based multiscale transformer network (AMTNet) that utilizes a CNN-transformer structure to address this issue. Our Siamese network based on the CNN-transformer architecture uses ConvNets as the backbone to extract multiscale features from the raw input image pair. We then employ attention and transformer modules to model contextual information in bi-temporal images effectively. Additionally, we use feature exchange to bridge the domain gap between different temporal image domains by partially exchanging features between the two Siamese branches of our AMTNet. Experimental results on four commonly used CD datasets – CLCD, HRSCD, WHU-CD, and LEVIR-CD – demonstrate the effectiveness and efficiency of our proposed AMTNet approach. The code for this work will be available on GitHub.11https://github.com/linyiyuan11/AMT_Net.

Coexistence of a Marginal Mountain Community with Large-Scale and Low Kinematic Landslide: The Intensive Monitoring Approach

Mountain territories affected by natural hazards are vulnerable areas for settlements and inhabitants. Additionally, those areas are characterized by socio-economic marginality, further favoring their abandonment. The study area is located in Liguria (Italy), and a large, slow-moving phenomenon endangers the settlements in the region. Monitoring such phenomena requires the use of instruments capable of detecting yearly, millimetric displacements and, due to their size, the use of remote techniques which can provide deformation measurement of the entire extent of the phenomenon. The methodology proposed here couples long-term interferometric remote sensing data analysis with intensive in situ monitoring (inclinometer, piezometers and global navigation satellite systems). Furthermore, the inclinometric measurements were carried out with an experimental, robotized inclinometer. The aim is to frame the overall context of ground deformation, assure information for inhabitants, stakeholders and land-planners, and secure coexistence with the phenomenon. Remote sensing provided a time series of 28 years of deformation measurements while in situ instrumentations allowed, in the last years, a better understanding of the surficial and deep behavior of the phenomenon, confirming the satellite data. Additionally, the high-frequency monitoring allowed us to record acceleration after precipitation peaks. The proposed approach, including the experimental instruments, proved its viability and can be replicated in similar mountain contexts.

Using geospatial data to identify land grabbing. Detecting spatial reconfigurations during the implementation of the Nacala Development Corridor in Mozambique with remote sensing and land conflicts databases

ABSTRACT The contemporary food system pushes agriculture to a globalized value-chain, affecting landscapes, resource access, and institutional arrangements. Institutions operating in Africa adopt development corridors to integrate multisector investments and induce export-driven primary sector, leading to massive land deals, also known as land-grabbing. Organizations struggle to monitor land deals accurately, lacking spatial precision and contextual information for affected communities. This research examines Mozambique's Nacala Corridor, using geospatial data as a tool to detect spatial (re)configurations due to exported-oriented policies and infrastructure. Data from land conflicts databases (Land Matrix and Environmental Justice) were analyzed with remote sensing Landsat and MODIS imagery using multiple indexes, an EVI time series, and the application of the LandTrendr algorithm. The results show that the temporal and spatial analysis of remote-sensing data is in line with the major political and economic dynamics of the region. Hotspots of land cover changes were detected in the same areas where land grabbing were reported; however, reported and detected land areas did not coincide. Temporal analysis showed that institutional changes played a greater role in triggering land use changes than infrastructure implementation. We conclude that land cover modifications, conflicts, and spatial development initiatives follows policies and institutional arrangements targeting international investments.

Zonation of The Waters of Spermonde Archipelago Base on Nutrient Concentration and Its Impact to Coral Reef Ecosystem

The Spermonde Archipelago is divided into four zones, the division of zones is based on the distribution of coral reefs, distance from the mainland, and water depth, but this condition tends to change as a result of high anthropogenic activity in the Spermonde Islands. Based on this, it is very necessary to re-zoning the Spermonde Archipelago Waters based on the condition of nutrient distribution. The research method uses the integration of remote sensing data analysis, field surveys, and secondary data which are then analyzed using the Boleean method for re-zoning. The results showed that the nutrient, in this case, chlorophyll-a, was concentrated in the deep zone (near the mainland) and the concentration was higher during the rainy season, this condition affected the coral reef cover in the Spermonde Islands. Based on these conditions, the Spermonde Islands waters are divided into three zones, namely the inner zone with high nutrient conditions with a benthic cover dominated by fleshy macroalgae, the middle zone with low nutrient conditions with crustose coralline dominance, and the outer zone with low nutrient conditions with algal turf dominance turf alga/corals.

Pemetaan Sebaran Habitat Dasar Perairan Laut Dangkal Menggunakan Citra Sentinel-2A di Teluk Gilimanuk

West Bali National Park is a conservation area with such a high biodiversity. That includes mangrove forest, coastal, and shallow water ecosystem. As a conservation area, a distribution map of the shallow water habitat is necessary for decision making in the conservation’s benthic habitat developing program. One of the mapping method is remote sensing and requires Sentinel-2A Level 1C satelitte image. Sentinel-2A Level 1C is used due to its high resolution which is 10x10 meters for a single pixel. The aim for this research is for mapping the distribution of the shallow water habitat in Gilimanuk Bay while also assesses the accuracy of Sentinel-2A image. This research uses the Quantum GIS and SAGA GIS software to analyze the Sentinel-2A image. The image is transformed using Depth Invariant Index to reduce the scattering effect on the reflectance. After the image is being transformed by the Depth Invariant Index, it is classified using maximum likelihood method. The classification showed, the area of each classes are 56.59 hectares of corals, 18.01 hectares of seagrass, 128.97 hectares of waters, 45.76 hectares of mud sediment, 78.95 hectares of macro algae, and 26.67 hectares of sands. Overall accuracy assessment for this research calculated using confusion matrix and the result is 65.1% and considered good enough due to Coral Ecosystem Remote Sensing Data Analysis Guide by National Institute of Aeronautics and Space (LAPAN), in which the tolerance of the accuracy is 60%. Accuracy levels are influenced by number of classes, image resolution, class resemblance, turbidity, and seafloor material.

UNMANNED AIRCRAFT SYSTEMS AND SATELLITE TECHNOLOGIES FOR TOPSOIL MAPPING IN PRECISION AGRICULTURE

Abstract. The use of satellite/UAS technologies in Precision Agriculture (PA) is increasing significantly and the continuous need of such kind of technologies is generating a great economic impact. This study carried out results by using Unmanned Aircraft Systems (UAS) and Very Hight Resolution (VHR) satellite imagery for the estimation of topsoil variability. The purpose of this contribution concerns the data analysis over two Production Units by evaluating radiometric response along the surface (e.g., spectral index). Our results concern the integration of UAS and satellite, improving the estimation accuracy of soil homogeneous areas variability, over which to manage agricultural practices. The remote sensing data covered by bare soil were isolated from the vegetative matrix along the study area. Thus, by using a statistical approach through the correlation between remote imagery and specific indices, it could be possible to determine variability in each pedological context, determining the zoning over land plots. According to this, in absence of high-resolution proximal soil sensing instrumentation, remote sensing data analysis can provide preliminary information suitable to establish a weighted topsoil/subsoil sampling campaign.

Evolution of County Socio–Ecological Systems in Nature Reserves in Western China Over the Past 30 Years

The evolution process of socio–ecological systems from 1990 to 2020 in the Tianzhu Tibetan Autonomous County, located in the National Key Ecological Function Area of western China, was analyzed quantitatively based on resilience theory and methodology combined with catastrophe theory and adaptive cycle theory, through field investigation, questionnaires and interviews, social data collection and remote sensing data analysis. The results show that the coordinated development of socio–ecological systems has made great progress in recent decades in the study area, and the coordinated development of social systems and ecosystems has a high degree of coupling and a strong connection. Changes in ecosystem resilience regularly surpassed changes in the social system, indicating the significant impact and success of ecological protection policies and projects in recent decades. In future, improvements in the social sub-system will be the key to developing the socio–ecological system in the study area. Enhancing social sub-system resiliency, implementing transformational development and green industry development, and transforming and realizing ecological product values are important topics for further investigation in the study area. Substantial changes in policy, production, population and climate change are needed to promote the evolution of socio–ecological systems. Stable national policies are crucial for improving people's livelihoods and providing ecological protections.

Cloud removal using scattering model and evaluation via semi-realistic simulation

ABSTRACT Cloud removal is an essential task in remote sensing data analysis. As the image sensors are distant from the earth ground, it is likely that part of the area of interests is covered by cloud. Moreover, the atmosphere in between creates a constant haze layer upon the acquired images. To recover the ground image, we propose to use scattering model for temporal sequence of images of any scene in the framework of low rank and sparse models. We further develop its variant, which is much faster and yet more accurate. To measure the performance of different methods objectively, we develop a semi-realistic simulation method to produce cloud cover so that various methods can be quantitatively analysed, which enables detailed study of many aspects of cloud removal algorithms, including verifying the effectiveness of proposed models in comparison with the state-of-the-arts, including deep learning models, and addressing the long standing problem of the determination of regularization parameters. Theoretic analysis on the range of the sparsity regularization parameter is provided and verified numerically.