Remote Sensing Applications-Society and Environment最新文献_第6页

Modeling PM2.5 concentration in tehran using satellite-based Aerosol optical depth (AOD) and machine learning: Assessing input contributions and prediction accuracy 利用基于卫星的气溶胶光学深度（AOD）和机器学习模拟德黑兰PM2.5浓度：评估输入贡献和预测精度

IF 3.8

Remote Sensing Applications-Society and Environment Pub Date : 2025-04-01 DOI: 10.1016/j.rsase.2025.101549

Zahra Amiri , Maryam Zare Shahne

{"title":"Modeling PM2.5 concentration in tehran using satellite-based Aerosol optical depth (AOD) and machine learning: Assessing input contributions and prediction accuracy","authors":"Zahra Amiri , Maryam Zare Shahne","doi":"10.1016/j.rsase.2025.101549","DOIUrl":"10.1016/j.rsase.2025.101549","url":null,"abstract":"<div><div>The adverse effects of PM<sub>2.5</sub> on human health and the environment necessitate precise and continuous monitoring of this pollutant. Satellite remote sensing technology provides an effective and cost-efficient alternative to ground-based measurements. However, accurately estimating ground-based PM<sub>2.5</sub> concentrations using Aerosol Optical Depth (AOD) is challenging due to the influence of various parameters and atmospheric conditions on the AOD-PM<sub>2.5</sub> relationship. In this study, Aerosol Optical Depth (AOD) data were retrieved from the Moderate Resolution Imaging Spectroradiometer (MODIS) sensor at a spatial resolution of 1 km, covering a ten-year period from 2012 to 2023. The objective was to estimate PM<sub>2</sub>.<sub>5</sub> concentrations for six ground-based monitoring stations in Tehran, Iran. These estimated concentrations were compared with daily measurements from the Tehran Air Quality Control Company air pollution monitoring stations for the same period. To determine the most significant conditioning factors in the modeling process and their impacts, the genetic algorithm optimization method and the Recursive Feature Elimination (RFE) technique were employed. The results indicated that, in addition to the AOD parameter, meteorological parameters such as wind speed, wind direction, temperature, precipitation, normalized difference vegetation index (NDVI), and visibility (VIS) could enhance model accuracy. Predictions were made using three machine learning algorithms: Random Forest (RF), Support Vector Machine (SVM), and <span>Gaussian</span> Process Regression (GPR). The findings revealed that the RF method was the most accurate, achieving accuracy in the range of 94–98 % for predicting PM<sub>2.5</sub> concentrations for all the studied stations. This study's results can significantly aid policymakers and researchers in utilizing satellite data for air pollution monitoring and management.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"38 ","pages":"Article 101549"},"PeriodicalIF":3.8,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143868625","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Deep learning-based detection and tracking of fin whales using high-resolution space-borne remote sensing data 利用高分辨率空间遥感数据对长须鲸进行基于深度学习的探测和跟踪

IF 3.8

Remote Sensing Applications-Society and Environment Pub Date : 2025-04-01 DOI: 10.1016/j.rsase.2025.101580

Vasavi Sanikommu, Akshaya Sura, Pranavi Chimirala

{"title":"Deep learning-based detection and tracking of fin whales using high-resolution space-borne remote sensing data","authors":"Vasavi Sanikommu, Akshaya Sura, Pranavi Chimirala","doi":"10.1016/j.rsase.2025.101580","DOIUrl":"10.1016/j.rsase.2025.101580","url":null,"abstract":"<div><div>Despite the ban on commercial whaling, the conservation of fin whale populations remains a significant challenge due to human-induced threats such as ship collisions, fishing gear entanglements, and underwater noise pollution. Traditional monitoring methods are logistically challenging and expensive, especially in remote and inaccessible regions. Recent advancements in high-resolution satellite imagery have demonstrated potential for automated marine species monitoring; several research gaps remain, including limited spectral band utilization, suboptimal deep-learning model adaptations, and lack of real-time tracking capabilities. This study presents an advanced deep-learning framework integrating U-Net for semantic segmentation, an enhanced YOLO model for object detection, and ResNet101 for classification to automate the detection and tracking of fin whales in satellite and infrared imagery. A key contribution is the integration of specific spectral bands optimized for underwater visibility, improving detection accuracy. The proposed system is deployed on edge devices, enabling real-time fin whale tracking with geospatial mapping of their locations. Experimental results demonstrate high performance across multiple datasets. U-Net achieves a segmentation accuracy of 92.21 %, the enhanced YOLO model attains a mean average precision (mAP) of 82 %, and ResNet101 reaches a classification accuracy of 99 %. Comparative analysis against existing methodologies highlights the improved detection precision and robustness of the proposed approach. By addressing key research gaps in spectral band selection, deep learning adaptation, and real-time deployment, this work contributes significantly to automated marine species monitoring and conservation. This study integrates drone-based surveys, hyperspectral imaging, thermal imagery, and Google Earth data with satellite imagery to enhance tracking capabilities.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"38 ","pages":"Article 101580"},"PeriodicalIF":3.8,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143923611","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

SISCNet: A novel Siamese inception-based network with spatial and channel attention for flood detection in Sentinel-1 imagery SISCNet：基于Siamese inception的新型网络，具有空间和通道关注，用于Sentinel-1图像的洪水检测

IF 3.8

Remote Sensing Applications-Society and Environment Pub Date : 2025-04-01 DOI: 10.1016/j.rsase.2025.101571

Sahand Tahermanesh , Ali Mohammadzadeh , Amin Mohsenifar , Armin Moghimi

{"title":"SISCNet: A novel Siamese inception-based network with spatial and channel attention for flood detection in Sentinel-1 imagery","authors":"Sahand Tahermanesh , Ali Mohammadzadeh , Amin Mohsenifar , Armin Moghimi","doi":"10.1016/j.rsase.2025.101571","DOIUrl":"10.1016/j.rsase.2025.101571","url":null,"abstract":"<div><div>The increasing frequency and severity of floods, driven by climate and socio-economic changes, underscore the necessity of quickly and accurately identifying flooded areas to minimize damage and support recovery efforts. Synthetic Aperture Radar (SAR) sensors are invaluable for this task, as they operate effectively in all weather conditions, both day and night, providing timely and precise data for flood mapping. However, processing SAR images can be challenging due to issues such as speckle noise and the scarcity of labeled training data. To address these challenges, our study introduces SISCNet, a novel deep learning-based framework specifically designed to detect flooded areas using SAR imagery. SISCNet employs a shared-weight dual-branch architecture that processes pre- and post-flood satellite images, enabling accurate and efficient flood detection. Compared to existing methods like FloodNet, SISCNet offers several advantages, including fewer training parameters, faster processing times, and improved accuracy in detecting flood events. The model's effectiveness was validated across seven different flood scenarios, consistently outperforming other techniques and demonstrating its robustness, even with smaller datasets. Integrating attention mechanisms further enhances SISCNet's ability to focus on critical features, resulting in superior overall performance in flood mapping tasks.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"38 ","pages":"Article 101571"},"PeriodicalIF":3.8,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143942473","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Modelling current and future suitable cultivation areas of cashew trees in Benin (West Africa) based on the major parasite and its parasitoid distribution under global climate warming 基于全球气候变暖背景下贝宁腰果主要寄生虫及其拟寄生虫分布的现状和未来适宜种植区域模型

IF 3.8

Remote Sensing Applications-Society and Environment Pub Date : 2025-04-01 DOI: 10.1016/j.rsase.2025.101589

Coffi Fulgence Gbèwommindéa Dotonhoué , Adigla Appolinaire Wédjangnon , Gafarou Agounde , Christine A.I. Nougbodé Ouinsavi

{"title":"Modelling current and future suitable cultivation areas of cashew trees in Benin (West Africa) based on the major parasite and its parasitoid distribution under global climate warming","authors":"Coffi Fulgence Gbèwommindéa Dotonhoué , Adigla Appolinaire Wédjangnon , Gafarou Agounde , Christine A.I. Nougbodé Ouinsavi","doi":"10.1016/j.rsase.2025.101589","DOIUrl":"10.1016/j.rsase.2025.101589","url":null,"abstract":"<div><div>The cashew tree is an essential source of income in West African households, especially in Benin. However, it faces declining productivity due to parasites and climate change. The insect <em>Oecophylla longinoda</em> (Latreille) is commonly used to control the cashew pest <em>Helopeltis schoutedeni</em> Reuter.; however, how climate change affects their distribution and how this can be used to identify suitable cashew cultivation areas remains a challenge. For this purpose, we used machine learning to identify suitable areas for cashew cultivation in Benin, considering occurrence points and environmental factors that limit the distribution of cashew trees, the pest, and the beneficial insect. Globally, models performed well, with mean values of the area under the curve ranging from 0.76 to 0.97 and mean values of the true skill statistics ranging from 0.44 to 0.85. Both precipitation seasonality and isothermality influenced the spatial distribution of cashew trees in Benin; while the mean temperature of the warmest months and annual precipitation determined the distribution of <em>H. schoutedeni</em>. As for <em>O</em>. <em>longinoda</em>, the precipitation of the driest quarter and wind speed determined its distribution. Suitable areas for cashew cultivation in current conditions were mainly concentrated in the agricultural development pole 4 (ATDA 4), encompassing the municipalities of Savalou, Bassila, Bantè, Glazoué, Tchaourou, Ouèssè, Savè, Dassa, and Parakou. These suitable areas are expected to decrease by 15.16–28.47 % by 2070, with a shift towards the south, especially in agricultural development pole 5 (ATDA 5) and 7 (ATDA 7) under ssp245 and ssp585. These findings are relevant for decision-makers in the medium and long-term targeting of suitable cultivation areas of cashew trees in Benin.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"38 ","pages":"Article 101589"},"PeriodicalIF":3.8,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143948688","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Integrated assessment of hydroclimatic extremes, land cover exposure, and vegetation responses in sub-humid and semi-arid regions in Brazil 巴西半湿润和半干旱地区极端水文气候、土地覆盖暴露和植被响应的综合评估

IF 3.8

Remote Sensing Applications-Society and Environment Pub Date : 2025-04-01 DOI: 10.1016/j.rsase.2025.101595

Beatriz M. Funatsu , Pedro R. Mutti , Vincent Dubreuil , Damien Arvor , Hugo Romero , Pablo Sarricolea

{"title":"Integrated assessment of hydroclimatic extremes, land cover exposure, and vegetation responses in sub-humid and semi-arid regions in Brazil","authors":"Beatriz M. Funatsu , Pedro R. Mutti , Vincent Dubreuil , Damien Arvor , Hugo Romero , Pablo Sarricolea","doi":"10.1016/j.rsase.2025.101595","DOIUrl":"10.1016/j.rsase.2025.101595","url":null,"abstract":"<div><div>The São Francisco (SF) River crosses large swaths of the Brazilian semi-arid territory, and has a strategic role in the socio-economic development of northeastern Brazil. Hydroclimatic extremes (HCE) have the potential to inflict considerable social and ecological perturbations, thus a fine-scale spatial analysis of HCE and their trends is essential for decision-making concerning adaptive measures in the region. Trend analysis indicate that rainfall intensity increased/decreased by 10 % in 28 %/14 % of the basin during the period of 1981–2021. Parts of northern SF shows both rainfall intensification and longer dry periods, but dry conditions prevail in most of the basin. In particular, 52 % of the basin presents extended consecutive dry days. Vegetation responses to HCEs were assessed with the Normalized Difference Vegetation Index (NDVI) for the period 2001–2021, considering two contrasting land cover types: “natural vegetation” (NatVeg) and “croplands” (Crops). Correlations between NDVI and HCE indices revealed positive and significant values maximized with a lag between −1 and −3 months for both NatVeg and Crops. However, vegetation responses in singular dry and wet years showed contrasts only for the lower range of NDVI. Most of the basin show negative mean annual NDVI trends even in areas with rainfall intensification. Since NDVI showed a stronger decrease in dry episodes than increase in wet ones, it is inferred that even highly adapted vegetation such as those in the semi-arid portions of the basin face mounting challenges to recover from extended droughts exacerbated by drier year-to-year observed conditions.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"38 ","pages":"Article 101595"},"PeriodicalIF":3.8,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144099067","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Analysis of spatial patterns of light pollution risk in urbanization areas and influencing factors based on Luojia 1-01 nighttime imagery and illuminance measurements 基于洛家1-01夜间影像和照度测量的城市化地区光污染风险空间格局及影响因素分析

IF 3.8

Remote Sensing Applications-Society and Environment Pub Date : 2025-04-01 DOI: 10.1016/j.rsase.2025.101587

Lujie Lin , Yiming Liu , Hui Zeng

{"title":"Analysis of spatial patterns of light pollution risk in urbanization areas and influencing factors based on Luojia 1-01 nighttime imagery and illuminance measurements","authors":"Lujie Lin , Yiming Liu , Hui Zeng","doi":"10.1016/j.rsase.2025.101587","DOIUrl":"10.1016/j.rsase.2025.101587","url":null,"abstract":"<div><div>Quantitative analyses of the spatial characteristics and drivers of risk levels in urban nighttime light pollution remain underexplored in research. Using Shenzhen as a case study, this paper combines nighttime remote sensing data with vertical illuminance measurements and applies five empirical models to estimate overall nighttime environmental illuminance. This approach supports spatial pattern analysis and light pollution risk assessment. Eight socio-economic and natural factors were selected as independent variables. Random forest, support vector machine, and back-propagation neural network models were employed to predict light pollution risk levels. Key findings include: (1) High-precision nighttime remote sensing data, coupled with ground survey results, effectively estimates surface illuminance levels; (2) the univariate linear model was optimal for surface illuminance inversion, indicating that Shenzhen's surface illuminance ranges from 0 to 70 lx, with higher values in the west and lower values in the east; (3) among machine learning models, the random forest (RF) model best identified influencing factors, with commercial area density, road network density, and residential area density as the main drivers of spatial differentiation in light pollution risk, while topography had no impact. (4) It is recommended that in urbanized areas with significant differences in topography, sampling and analysis should be conducted in illuminated areas when analyzing the causes of light pollution. This study offers systematic insights into urban light pollution patterns and drivers, providing scientific support for urban light environment planning and management.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"38 ","pages":"Article 101587"},"PeriodicalIF":3.8,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143934981","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Estimation of soil particle fractions in Korea based on spectral imaging and ANN for UAV-based field applications 基于光谱成像和人工神经网络的韩国土壤颗粒组分估算在无人机领域的应用

IF 3.8

Remote Sensing Applications-Society and Environment Pub Date : 2025-04-01 DOI: 10.1016/j.rsase.2025.101572

Hwan-Hui Lim , Enok Cheon , Woojae Jang , Tae-Hyuk Kwon , Jin-Woo Cho , Seung-Rae Lee

{"title":"Estimation of soil particle fractions in Korea based on spectral imaging and ANN for UAV-based field applications","authors":"Hwan-Hui Lim , Enok Cheon , Woojae Jang , Tae-Hyuk Kwon , Jin-Woo Cho , Seung-Rae Lee","doi":"10.1016/j.rsase.2025.101572","DOIUrl":"10.1016/j.rsase.2025.101572","url":null,"abstract":"<div><div>This study aimed to develop a straightforward method for predicting soil particle fractions, a key factor for landslide prediction and land use management, using hyperspectral imaging. A total of 114 soil samples were collected across Korea and categorized into four color groups (brown, red, yellow, and gray) based on their red, green, and blue values. Spectral images of each sample were captured through hyperspectral experiments, with spectral data analyzed on a pixel-by-pixel basis. Next, we developed an index for classifying soil particle fractions from these spectral data and created an artificial neural network (ANN) model that uses this index to predict soil particle fractions. The ANN model demonstrated a prediction accuracy with a root mean square error (RMSE) of 3.04 for sandy soil and 1.16 for fine particles. The ANN model was field-tested in the Pyeongchang region, where spectral images were obtained using an unmanned aerial vehicle (UAV) equipped with a multispectral camera. The images were preprocessed into a single orthophoto, and the spectral data were extracted through band-merging. The field RMSEs were 5.48 for sandy soil and 0.107 for fine particles. Therefore, the ANN model shows potential for field application in identifying soil characteristics across various regions.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"38 ","pages":"Article 101572"},"PeriodicalIF":3.8,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143903870","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Drone remote sensing and machine learning for green stormwater infrastructure condition assessment 基于无人机遥感和机器学习的绿色雨水基础设施状况评估

IF 3.8

Remote Sensing Applications-Society and Environment Pub Date : 2025-04-01 DOI: 10.1016/j.rsase.2025.101590

Matthew Dupasquier, Walter McDonald

{"title":"Drone remote sensing and machine learning for green stormwater infrastructure condition assessment","authors":"Matthew Dupasquier, Walter McDonald","doi":"10.1016/j.rsase.2025.101590","DOIUrl":"10.1016/j.rsase.2025.101590","url":null,"abstract":"<div><div>Maintenance and operations of green stormwater infrastructure is critical to preserve the functionality of urban stormwater infrastructure. However, doing so is a challenge due to the disperse locations of green stormwater infrastructure that may be difficult to access, which results in limited and inconsistent inspections that are also human and resource intensive. The objective of this study is to overcome this limitation through a novel approach to green stormwater infrastructure inspection that applies machine learning models to remote sensing data from an unmanned aerial system to assess green stormwater infrastructure landcover. To do so, machine learning models were applied to categorize land cover of green stormwater infrastructure into 4 condition-related classes: healthy plants, unhealthy plants, dead plants and organic material, and inorganic material. Models were trained and tested via multitemporal analysis at 12 unique locations encompassing various green stormwater infrastructure types (e.g., bioswale, green roof, rain garden, native planting area). The landcover classification accuracy assessments showed that supervised object-based and pixel-based methods exhibited similar overall accuracy (87 % and 88 %, respectively) during training and testing. Notably, Random Trees and Support Vector Machine algorithms outperformed Maximum Likelihood and k-Nearest Neighbors by an average of (+4 %). Overall, these methods can be used to obtain informative data that can enhance green stormwater infrastructure monitoring and maintenance efforts.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"38 ","pages":"Article 101590"},"PeriodicalIF":3.8,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144069310","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Detection of Oil Spilled at Sea with Entropy and Alpha polarimetric in Sentinel-1 C-band SAR data: Lima-Peru case 利用Sentinel-1 c波段SAR数据的熵和α极化检测海上溢油：利马-秘鲁案例

IF 3.8

Remote Sensing Applications-Society and Environment Pub Date : 2025-04-01 DOI: 10.1016/j.rsase.2025.101585

Wilmer Moncada , Joel Rojas Acuña , Cristhian Aldana , Jaime Lloret

{"title":"Detection of Oil Spilled at Sea with Entropy and Alpha polarimetric in Sentinel-1 C-band SAR data: Lima-Peru case","authors":"Wilmer Moncada , Joel Rojas Acuña , Cristhian Aldana , Jaime Lloret","doi":"10.1016/j.rsase.2025.101585","DOIUrl":"10.1016/j.rsase.2025.101585","url":null,"abstract":"<div><div>The Sentinel-1 (S1) SLC-IW Synthetic Aperture Radar (SAR) contributes to ensuring global sustainability and environmental impact mitigation with early detection of Oil Spills at Sea (OSS), such as the one that occurred on January 15, 2022, Pampilla refinery, Ventanilla-Lima. This event was catalogued as the worst ecological disaster in recent times in Lima due to its high proportions causing serious environmental effects. The objective is to detect OSS with entropy (H) and polarimetric alpha angle (α) in S1A, SLC-IW, dual-pol data in VV + VH channels, Lima-Peru case. From the SLC product, the values of the covariance matrix components <span><math><mrow><msub><mi>C</mi><mn>11</mn></msub></mrow></math></span>, <span><math><mrow><msub><mi>C</mi><mn>12</mn></msub></mrow></math></span>, <span><math><mrow><msub><mi>C</mi><mn>21</mn></msub></mrow></math></span> and <span><math><mrow><msub><mi>C</mi><mn>22</mn></msub></mrow></math></span>, to which the refined Speckle Lee polarimetric filter has been applied, are evaluated for their decomposition and determination of the polarimetric properties, such as H and α in the identification of areas with OSS. The results of S1A, January 25, 2022, from the threshold values of H (0.5–0.89) and α (10°–30°), are validated with the sampled points (SP) on that date. The Kappa index (κ) shows moderate agreement for H (κ = 0.49) and good agreement for α (κ = 0.61). For S1A, February 02, 2022, the threshold values of H (0.5–0.72) and α (10°–20°) are validated with the SP on that date, indicating very good agreement for H(κ = 0.85) and good agreement for α (κ = 0.76).</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"38 ","pages":"Article 101585"},"PeriodicalIF":3.8,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144099068","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

LOD1 3D city model from LiDAR: The impact of segmentation accuracy on quality of urban 3D modeling and morphology extraction 基于LiDAR的LOD1三维城市模型：分割精度对城市三维建模和形态提取质量的影响

IF 3.8

Remote Sensing Applications-Society and Environment Pub Date : 2025-04-01 DOI: 10.1016/j.rsase.2025.101534

Fatemeh Chajaei, Hossein Bagheri

{"title":"LOD1 3D city model from LiDAR: The impact of segmentation accuracy on quality of urban 3D modeling and morphology extraction","authors":"Fatemeh Chajaei, Hossein Bagheri","doi":"10.1016/j.rsase.2025.101534","DOIUrl":"10.1016/j.rsase.2025.101534","url":null,"abstract":"<div><div>Three-dimensional reconstruction of buildings, particularly at Level of Detail 1 (LOD1), plays a crucial role in various applications such as urban planning, urban environmental studies, and designing optimized transportation networks. This study focuses on assessing the potential of LiDAR data for accurate 3D building reconstruction at LOD1 and extracting morphological features from these models. Four deep semantic segmentation models — U-Net, Attention U-Net, U-Net3+, and DeepLabV3+ — were used, applying transfer learning to extract building footprints from LiDAR data. The results showed that U-Net3+ and Attention U-Net outperformed the others, achieving IoU scores of 0.833 and 0.814, respectively. Various statistical measures, including maximum, range, mode, median, and the 90th percentile, were used to estimate building heights, resulting in the generation of 3D models at LOD1. As the main contribution of the research, the impact of segmentation accuracy on the quality of 3D building modeling and the accuracy of morphological features like building area and external wall surface area was investigated. The results showed that the accuracy of building identification (segmentation performance) significantly affects the 3D model quality and the estimation of morphological features, depending on the height calculation method. Overall, the UNet3+ method, utilizing the 90th percentile and median measures, leads to accurate height estimation of buildings and the extraction of morphological features.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"38 ","pages":"Article 101534"},"PeriodicalIF":3.8,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143785925","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0