ISPRS Journal of Photogrammetry and Remote Sensing最新文献

{"title":"Self-attention and frequency-augmentation for unsupervised domain adaptation in satellite image-based time series classification","authors":"David Gackstetter ,&nbsp;Kang Yu ,&nbsp;Marco Körner","doi":"10.1016/j.isprsjprs.2025.03.024","DOIUrl":"10.1016/j.isprsjprs.2025.03.024","url":null,"abstract":"<div><div>With the increasing availability of Earth observation data in recent years, the development of deep learning algorithms for the classification of satellite image time series (SITS) has substantially progressed. Yet, when encountering settings of lacking target labels and distinct feature variations, even the latest classification algorithms may perform poorly in transferring knowledge from a trained dataset to an unknown target dataset, despite similar or even identical label sets. The research field of unsupervised domain adaptation (UDA) focuses on methods to overcome these challenges by providing algorithms that explicitly learn domain shifts between different data domains in the absence of target-labeled data. Building upon recent advances on generic UDA research in time series settings, we propose RAINCOAT-SRS, an enhancement of the frequency-augmented UDA-algorithm RAINCOAT specifically for the SITS domain. To evaluate the default and adjusted model variants, we designed several closed-label set, cross-regional and multi-temporal crop type mapping experiments, which represent common sub-problems of UDA in SITS. We first benchmark RAINCOAT against TimeMatch as a leading algorithm in this application context. Subsequently, we explored different encoder-to-decoder constellations as architectural enhancements. These analyses revealed that a combination of an self-attention-based encoder with the default decoder yields a performance increase to the standard algorithm of up to 6 % in average f1-score, and to TimeMatch by up to 24 %. Beyond, we assessed the impact of the frequency feature and SITS-specific feature extensions by integrating weather data, which both showed to improve classification accuracy only in individual sub-experiments however not consistently across the entire scope of scenarios. Finally, we outline key factors influencing the transferability, thereby emphasizing the major importance of domain-overarching stability of class-relative, structural patterns rather than of collective, linear shifts between domains. Through this research, we introduce RAINCOAT-SRS, a novel model for UDA in SITS, designed to advance generalization in remote sensing by enabling more comprehensive cross-regional and multi-temporal SITS experiments in face of lacking target-labeled data.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"224 ","pages":"Pages 113-132"},"PeriodicalIF":10.6,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143808096","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Simultaneous planning of standpoints and routing for laser scanning of buildings with network redundancy","authors":"Julius Knechtel ,&nbsp;Youness Dehbi ,&nbsp;Lasse Klingbeil ,&nbsp;Jan-Henrik Haunert","doi":"10.1016/j.isprsjprs.2025.03.017","DOIUrl":"10.1016/j.isprsjprs.2025.03.017","url":null,"abstract":"<div><div>Stop-and-go laser scanning is becoming increasingly prevalent in a variety of applications, e.g., the survey of the built environment. For this, a surveyor needs to select a set of standpoints as well as the route between them. This choice, however, has a high impact on both the economic efficiency of the respective survey as well as the completeness, accuracy, and subsequent registrability of the resulting point cloud.</div><div>Assuming a set of building footprints as input, this article proposes a one-step optimization method to find the minimal number of selected standpoints based on scanner-related constraints. At the same time, we incorporate the length of the shortest route connecting the standpoints in the objective function. A local search method to speed up the time for solving the corresponding Mixed-Integer Linear Program (MILP) is additionally presented. The results for different scenarios show constantly shorter routes in comparison to existing approaches while still maintaining the minimal number of standpoints.</div><div>Moreover, in our formulation we aim to minimize the effects of inaccuracies in the software-based registration. Inspired by the ideas of network survivability, we to this end propose a novel definition of connectivity tailored for laser scanning networks. On this basis, we enforce redundancy for the registration network of the survey. To prove the applicability of our formulation, we applied it to a large real-world scenario.</div><div>This paves the way for the future use of fully automatic autonomous systems to provide a complete and high-quality model of the underlying building scenery.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"224 ","pages":"Pages 59-74"},"PeriodicalIF":10.6,"publicationDate":"2025-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143777365","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Improving visual grounding in remote sensing images with adaptive modality guidance","authors":"Shabnam Choudhury ,&nbsp;Pratham Kurkure ,&nbsp;Biplab Banerjee","doi":"10.1016/j.isprsjprs.2025.02.031","DOIUrl":"10.1016/j.isprsjprs.2025.02.031","url":null,"abstract":"<div><div>This study explores the visual grounding (VG) task using data from remote sensing sensors, focusing on grounding phrases or prompts. We achieve this by employing masked feature encoding to create enriched, target-specific, and semantically aware visual representations. We aim to develop an efficient grounding architecture that uses language expressions as prompts for aerial images. Despite significant progress in phrase grounding within natural scenes, the vision community has yet to effectively translate these advancements to aerial imagery. This gap is mainly due to the lack of finely annotated remote sensing datasets and the substantial variation in the dimensional attributes of instances observed on Earth’s surface. We propose a novel framework, <span>AMVG</span>, for visual grounding, designed to enhance object localisation by leveraging multi-modal deformable attention encoding and adaptive decoding guided by language. In contrast to traditional methods that utilise region proposals or anchor-box-based predictions, our approach dynamically adapts attention to emphasise key spatial details and contextually relevant areas, enabling more accurate alignment of visual and textual features. Additionally, we propose a new training objective, Attention Alignment Loss (AAL), to address inconsistencies in attention allocation that can lead to suboptimal grounding accuracy. Experiments demonstrate that our method, Adaptive Modality-guided Visual Grounding <span>(AMVG)</span>, significantly improves grounding accuracy compared to baseline methods, especially in challenging scenarios involving multiple textual and instance descriptions or ambiguous visual contexts. Our methodology outperforms recent methods, including MGVLF, CrossVG, VSMR, LPVA, and LQVG, achieving marginal improvements of 7.14%, 4.06%, 5.74%, 2.27%, and 3.35% in mean IoU, respectively. Similarly, for cumulative IoU, our approach surpasses MGVLF, VSMR, LPVA, and LQVG by 11.79%, 10.5%, 5.31%, and 3.14%, respectively. Similar trends are observed on the RSVG-HR and OPT-RSVG datasets, where <span>AMVG</span> demonstrates superior grounding accuracy in challenging scenarios involving complex textual prompts, ambiguous visual contexts, and diverse object attributes. These results reaffirm the effectiveness of our proposed method and its potential to address critical challenges in remote sensing VG tasks, setting a new benchmark in this domain. The full implementation of <span>AMVG</span> is publicly available <span><span>https://github.com/Shabnamchoudhury/AMVG.git</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"224 ","pages":"Pages 42-58"},"PeriodicalIF":10.6,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143739126","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Efficient metric-resolution land cover mapping using open-access low resolution annotations with prototype learning and modified Segment Anything model","authors":"W.Y. Shi,&nbsp;H.G. Sui,&nbsp;C.L. Zhang,&nbsp;N. Zhou,&nbsp;M.T. Zhou,&nbsp;J.X. Wang,&nbsp;Z.T. Du","doi":"10.1016/j.isprsjprs.2025.03.021","DOIUrl":"10.1016/j.isprsjprs.2025.03.021","url":null,"abstract":"<div><div>Large-scale metric-resolution land cover mapping is crucial for a detailed understanding of large areas of the Earth surface, supporting fine-scale decision-making in sectors such as agriculture, forestry, and conservation. Despite advances in remote sensing technology, this type of mapping is often limited by a lack of high-quality manually labeled data. To address this issue, here we propose a novel framework called Segment Anything Model with Prototype Learning and Enhanced Refinement (SAMPLER), using open-access and lower-resolution land cover products (LCPs) as labels for metric-resolution land cover mapping. Specifically, SAMPLER is built upon Segment Anything Model (SAM), leveraging its advanced feature extraction capability for precise and stable segmentation across diverse scenarios. To bridge the resolution gap between LCPs and metric-resolution imagery, we designed the class prototype-based object-oriented decoder (CPO-Decoder) to accurately classify small-scale features and maintain spectral consistency, effectively managing complex intra- and inter-class variations by dynamically aligning object-level features with semantic prototypes. Additionally, a label refinement strategy iteratively updates noisy LCP labels by replacing low-confidence annotations with high-precision model predictions, thereby mitigating label ambiguity caused by coarse resolution and enabling the model to progressively adapt to fine-scale features while improving classification accuracy throughout the training process. This framework exhibits exceptional adaptability and performance across various sensor types and diverse geographic regions, achieving an overall accuracy (OA) of 85.3% and a frequency-weighted intersection over union (FWIoU) of 74.7% across four study areas, with an OA increase from 5.6% to 9.5% and an FWIoU enhancement from 5.2% to 11.8% relative to the original LCP labels. Comparative experiments demonstrate that the proposed SAMPLER outperforms other deep learning models by up to 6.9% in OA and 9.8% in FWIoU. Ablation experiments further prove the effectiveness of CPO-Decoder, prototype learning and label-iterative-refined strategy. The overall results highlight the potential of SAMPLER for efficient metric-resolution land cover mapping at a large scale without manually labeled data, providing a valuable tool for fine-scale applications from environmental conservation to urban planning.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"224 ","pages":"Pages 19-41"},"PeriodicalIF":10.6,"publicationDate":"2025-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143735268","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"DiffSARShipInst: Diffusion model for ship instance segmentation from synthetic aperture radar imagery","authors":"Xiaowo Xu,&nbsp;Xiaoling Zhang,&nbsp;Shunjun Wei,&nbsp;Jun Shi,&nbsp;Wensi Zhang,&nbsp;Tianwen Zhang,&nbsp;Xu Zhan,&nbsp;Yanqin Xu,&nbsp;Tianjiao Zeng","doi":"10.1016/j.isprsjprs.2025.02.030","DOIUrl":"10.1016/j.isprsjprs.2025.02.030","url":null,"abstract":"<div><div>Recently, deep learning (DL) methods, particularly convolutional neural networks (CNNs)-based ones, have significantly advanced the development of synthetic aperture radar (SAR) ship instance segmentation. However, existing instance segmentation algorithms typically rely on preset candidate boxes, which are challenging to perfectly match to ships from a regression optimization perspective, limiting segmentation accuracy. Therefore, we propose a novel diffusion model, DiffSARShipInst, for SAR ship instance segmentation. This model represents ship instance segmentation as a denoising process from noisy boxes to target boxes and a reconstruction process from target boxes to ship instances. It innovatively handles the ship instance segmentation task from a generative perspective, treating random boxes as object candidates to overcome the limitations of existing methods that require target priors. To achieve superior SAR ship instance segmentation accuracy, DiffSARShipInst also offers: 1) a spatial-contextual joint enhanced feature pyramid network (SCJE-FPN) to improve the multi-scale ship feature extraction ability for the subsequent denoising and reconstruction processes; 2) a focused intersection-over-union (FIoU) loss to suppress redundant noisy samples during the denoising process; and 3) an instance-aware mask representation (IAMR) to adaptively generate ship instances from denoised target boxes during the reconstruction process. Extensive experiments on the SAR ship detection dataset (SSDD) and the high-resolution SAR image dataset (HRSID) demonstrate its superior performance. Specifically, DiffSARShipInst achieves up to 70.6 %/70.9 % mask average precision (AP) in offshore scenes of SSDD/HRSID, and 56.2 %/42.6 % mask AP in inshore scenes of SSDD/HRSID.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"223 ","pages":"Pages 440-455"},"PeriodicalIF":10.6,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143724708","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An nD-histogram technique for querying non-uniformly distributed point cloud data","authors":"Haicheng Liu ,&nbsp;Zhiwei Li ,&nbsp;Peter van Oosterom ,&nbsp;Martijn Meijers ,&nbsp;Chuqi Zhang","doi":"10.1016/j.isprsjprs.2025.03.014","DOIUrl":"10.1016/j.isprsjprs.2025.03.014","url":null,"abstract":"<div><div>Point cloud data contains abundant information besides XYZ, such as Level of Importance (LoI) and intensity. These non-spatial dimensions are also frequently used and queried. Therefore, developing an efficient nD solution for managing and querying point clouds is imperative. Previous researchers have developed PlainSFC that maps both nD points and queries into a one-dimensional Space Filling Curve (SFC) space and uses B+-tree for indexing. However, when computing SFC ranges for selection, PlainSFC subdivides the nD space mechanically to approach the query window without considering the point distribution. Then, excessive ranges are generated in vacant areas, and ranges generated in dense point areas are coarse. Consequently, a large number of false positives are selected, slowing down the whole querying process.</div><div>This paper develops a new solution called HistSFC to resolve the issue. HistSFC builds an nD-histogram which records point data distribution, and uses it to compute ranges for selecting data. Also, this paper discovers a novel statistical metric, Cumulative Hypercubic Coverage (CHC), to measure the uniformity of the point cloud data. Theory is established and it indicates that the nD-histogram is more beneficial when CHC is smaller. Thus, CHC can be used to guide the building of HistSFC. In addition, the paper conducts simulations and benchmark tests to examine the improvement on PlainSFC. It turns out that using the nD-histogram can decrease the false positive rate by orders of magnitude. HistSFC is also evaluated against state-of-the-art solutions. The result shows that HistSFC leads the performance in nearly all the tests.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"224 ","pages":"Pages 1-18"},"PeriodicalIF":10.6,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143725166","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ICESat-2 bathymetry algorithms: A review of the current state-of-the-art and future outlook","authors":"Jaehoon Jung ,&nbsp;Christopher E. Parrish ,&nbsp;Lori A. Magruder ,&nbsp;Joan Herrmann ,&nbsp;Suhong Yoo ,&nbsp;Jeffrey S. Perry","doi":"10.1016/j.isprsjprs.2025.03.016","DOIUrl":"10.1016/j.isprsjprs.2025.03.016","url":null,"abstract":"<div><div>Over six years of on-orbit operations, the NASA Ice, Cloud and Land Elevation Satellite-2 (ICESat-2) has proven the value of space-based laser altimetry for monitoring the response of Earth’s surfaces to a changing climate with continuous elevation measurements. Although bathymetry is not an official science requirement for the mission, ICESat-2 has had a transformative impact on understanding nearshore bathymetry. Despite its successful, proven contributions to nearshore and coastal studies, there is currently no standardized, fully-automated algorithm for the extraction of ICESat-2 bathymetry signal. However, there are many published algorithms in the literature that present novel and innovative approaches for all aspects of the bathymetric workflow, as well as processes for conflating the ICESat-2 measurements with optical data to support bathymetric mapping. In this work, we provide a comprehensive review and synthesis of the existing algorithms and procedures comprising the main steps in an end-to-end ICESat-2 bathymetry workflow, including water surface extraction, classification of bathymetric bottom returns, refraction correction, accuracy assessment, integration with optical imagery, and ancillary steps. The review is intended to inform the development of a new Level 3 along-track data product for the ICESat-2 mission (ATL24) as a global resource for nearshore bathymetry and to aid other researchers in developing and testing their own algorithms for ICESat-2 bathymetry. We provide an assessment of the current-state-of-the-art algorithms, challenges and limitations, as well as recommended next steps for the community of researchers working on and with ICESat-2 bathymetry.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"223 ","pages":"Pages 413-439"},"PeriodicalIF":10.6,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143724707","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Global patterns and determinants of year-to-year variations in surface urban heat islands","authors":"Xuanqing Guo ,&nbsp;Huilin Du ,&nbsp;Wenfeng Zhan ,&nbsp;Yingying Ji ,&nbsp;Chenguang Wang ,&nbsp;Chunli Wang ,&nbsp;Shuang Ge ,&nbsp;Shasha Wang ,&nbsp;Jiufeng Li ,&nbsp;Sida Jiang ,&nbsp;Dazhong Wang ,&nbsp;Zihan Liu ,&nbsp;Yusen Chen ,&nbsp;Jiarui Li","doi":"10.1016/j.isprsjprs.2025.03.019","DOIUrl":"10.1016/j.isprsjprs.2025.03.019","url":null,"abstract":"<div><div>Investigations on year-to-year variations in surface urban heat island intensity (Δ<em>I</em>_s, the change in urban heat island intensity between consecutive years) are crucial for capturing the dynamics of urban climates at mid-term scales. While the patterns and underlying drivers of <em>I</em>_s have been extensively studied, their year-to-year variability remains poorly understood, especially across global cities. Using MODIS land surface temperature observations from March 2003 to February 2024, here we examined the spatiotemporal patterns of Δ<em>I</em>_s across 1,642 cities worldwide, by removing the interannual component from yearly <em>I</em>_s observations. We also analyzed the impacts from various background climate and urban surface property factors on these patterns. Additionally, we simulated the Δ<em>I</em>_s by integrating the advanced Light Gradient Boosting Machine (LightGBM) model with various controlling factors. Our analysis yielded three key findings: (1) The global mean absolute Δ<em>I</em>_s (i.e., Δ<em>I</em>_{s_mean}) was 0.30 ± 0.02 K (mean ± S.D.) during the day and 0.18 ± 0.01 K at night, accounting for approximately 19.40 % and 13.57 % of overall <em>I</em>_s observations. Spatially, both daytime and nighttime Δ<em>I</em>_{s_mean} were notably higher in snow climates compared to equatorial, arid, and warm climates. (2) In terms of controlling factors, global daytime Δ<em>I</em>_{s_mean} showed strong negative correlations with year-to-year variations in both urban–rural EVI contrast (<em>r</em> = −0.69, <em>p</em> &lt; 0.01) and background surface air temperature (<em>r</em> = −0.62, <em>p</em> &lt; 0.01). By comparison, these correlations became less significant at night. (3) The LightGBM model demonstrated high accuracy in estimating the Δ<em>I</em>_s across global cities, with <em>r</em> values exceeding 0.96 and MAE values below 0.09 K for both daytime and nighttime. These findings are critical for enriching our understanding of urban heat island patterns at multiple temporal scales. They also provide an efficient approach for identifying abrupt urban climate changes due to extreme climate events or anthropogenic activities.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"223 ","pages":"Pages 399-412"},"PeriodicalIF":10.6,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143724307","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Paved or unpaved? A deep learning derived road surface global dataset from mapillary street-view imagery","authors":"Sukanya Randhawa ,&nbsp;Eren Aygün ,&nbsp;Guntaj Randhawa ,&nbsp;Benjamin Herfort ,&nbsp;Sven Lautenbach ,&nbsp;Alexander Zipf","doi":"10.1016/j.isprsjprs.2025.02.020","DOIUrl":"10.1016/j.isprsjprs.2025.02.020","url":null,"abstract":"<div><div>Road surface information is essential for applications in urban planning, disaster routing or logistics optimization and helps to address various Sustainable Development Goals (SDGS): especially SDGs 1 (No poverty), 3 (Good health and well-being), 8 (Decent work and economic growth), 9 (Industry, Innovation and Infrastructure), 11 (Sustainable cities and communities), 12 (Responsible consumption and production), and 13 (Climate action). We have released an open dataset with global coverage that provides road surface characteristics (paved or unpaved). The data was derived by a GeoAI approach that utilized 105 million images from the world’s largest crowdsourcing-based street-view platform, Mapillary. We propose a hybrid deep learning approach which combines SWIN-Transformer based road surface prediction and CLIP-and-DL segmentation based thresholding for filtering of bad quality images. The road surface prediction results have been matched and integrated with OpenStreetMap (OSM) road geometries. Model validation against OSM surface data achieved strong performance, with F1 scores for paved roads varying between 91%–97% across continents. The dataset expands the availability of global road surface information by nearly four million kilometers compared to currently available information in OSM — now representing approximately 36% of the total length of the global road network. Most regions showed moderate to high paved road coverage (60%–80%), but significant gaps were noted in specific areas of Africa and Asia. Urban areas tend to have near-complete paved coverage, while rural regions displayed more variability. This information has the potential to derive more reliable estimations for indicators such as rural accessibility or regional economic development potential and to assist e.g. humanitarian actors in emergency logistic planning.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"223 ","pages":"Pages 362-374"},"PeriodicalIF":10.6,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143697149","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Towards building floor-level nighttime light exposure assessment using SDGSAT-1 GLI data","authors":"Hailan Huang ,&nbsp;Bin Wu ,&nbsp;Yu Wang ,&nbsp;Bailang Yu ,&nbsp;Huabing Huang ,&nbsp;Wuming Zhang","doi":"10.1016/j.isprsjprs.2025.03.018","DOIUrl":"10.1016/j.isprsjprs.2025.03.018","url":null,"abstract":"<div><div>The profound impact of light pollution on both natural and human systems is well-recognized. Particularly, light pollution at the building scale is inextricably intertwined with human living and has garnered increasing attention in recent years. However, the coarse spatial resolution of nighttime light data, coupled with the inadequacy of existing methods, have precluded detailed investigation into the light pollution at building scale. The high-resolution Glimmer Imager (GLI) sensor onboard the SDGSAT-1 satellite provides nighttime light data with a 40-meter resolution, offering new opportunities for precise assessment of light pollution at the building scale. To this end, this study introduces a novel approach for calculating light exposure at the building floor-level using SDGSAT-1 GLI data. Two measures, Floor Light Exposure Index (FLEI) and Building Light Exposure Index (BLEI), are proposed to quantify the cumulative nighttime light radiation received at each floor and building, respectively, thereby facilitating the analysis of variances in light exposure across different buildings and floors. Utilizing this approach, we computed the floor-level light exposure for 57,221 buildings within three core districts—Yuexiu, Haizhu, and Tianhe—of Guangzhou city, China. The results, perhaps for the first time, quantified the level of light exposure at the building scale, revealing substantial differences in light exposure both inter-building and intra-building across various floors. Comparative analysis with field-collected data confirmed the robustness of our method and the reliability of the calculation results. We found that the light exposure is generally lower on lower floors, with a significant increase in light exposure above the 50th floor. Buildings in proximity to light sources and roads are more susceptible to light pollution, with light exposure in residential areas intensifying from the center to the periphery, and light exposure in commercial outskirts decreasing with increasing distance from the commercial center. The average FLEI in commercial zones is approximately 550 nW cm⁻² sr^-1 higher than that in residential areas. The approach and resultant building floor-level light exposure map generated by this study hold substantial promise in aiding the evaluation of various targets and indicators associated with multiple Sustainable Development Goals (SDGs) targets and indicators, including SDG 3 (Good Health and Well-being), SDG 11 (Sustainable Cities and Communities), and SDG 13 (Climate Action).</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"223 ","pages":"Pages 375-397"},"PeriodicalIF":10.6,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143697150","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}