Search Results (131)

Cross-view image geolocalization aims to achieve accurate localization of geo-tagged images without geo-tagging by matching ground-view images with satellite images. However, there are huge imaging differences between ground and satellite viewpoints, and existing methods usually rely on a large number of accurately labeled cross-view image pairs. Therefore, to address issues such as significant perspective differences, high annotation costs, and low utilization of unpaired data, this paper proposes a cross-view generation model that integrates multi-scale contrastive learning and dynamic optimization, designs a multi-scale contrast loss function to strengthen the semantic consistency between the generated images and the target domain, adaptively balances the quality and quantity of pseudo-labels according to a dynamic threshold screening mechanism, and introduces a hard-sample triplet loss to enhance the model discriminative ability. Ablation experiments on the CVUSA and CVACT datasets show that the BEV-CycleGAN+CL (Bird’s-Eye View Cycle-Consistent Generative Adversarial Network with Contrastive Learning) model proposed in this paper significantly outperforms the comparative models in PSNR, SSIM, and RMSE metrics. Specifically, on the CVACT dataset, compared with the BEV-CycleGAN, BEV, and CycleGAN baselines, PSNR increased by 2.83%, 16.02%, and 42.30%, SSIM increased by 6.12%, 8.00%, and 18.48%, and RMSE decreased by 9.28%, 15.51%, and 25.35%, respectively. Similar advantages are observed on the CVUSA dataset. Compared with current state-of-the-art models, the dynamic threshold pseudo-label localization method in this paper demonstrates overall superiority in recall metrics such as R@1, R@5, R@10, and R@1%, for example achieving an R@1 of 98.94% on CVUSA, outperforming the best comparative model, Sample4G^†, which reached 98.68%. This study provides innovative methodological support for disaster emergency response, high-precision map construction for autonomous driving, military reconnaissance, and other applications. Full article

This paper exposes the latent but potent role of seemingly hidden spatial autocorrelation (SA) in all geographic theories, highlighting that it is everywhere, matters, and is a fundamental property of geotagged phenomena. This narrative examines and extends the literature about the inescapable nature of the SA paradigm and the near-universal mixing of positive and negative SA. This study summary transcends the widespread but often implicit treatment of SA within geographic theories that their assumptions help achieve when they embed spatial processes, shape geospatial expectations, and define independent areal units so that these theory-delineating constraints largely absorb SA, reducing residual spatial dependence/correlation and improving conjectural validity, masking its presence for decades if not centuries. This paper explores selected prominent human geography theories (spatial optimization, agricultural location, gravity-model-based spatial interaction, central place systems), cultural and humanistic geography, geohumanities abstractions, physical geography theories (plate tectonics, climatology, uniformitarianism, soil formation), cartographic theories (geometric projections, semiotic/communication, cognitive/perceptual, geographic information systems anchored spatial analysis), and basic geospatial data gathering methodologies (qualitative and quantitative spatial sampling). It demonstrates that across the discipline of geography, exposing masquerading SA deepens theoretical coherence and strengthens methodological integrity, encouraging integrated spatial reasoning that bridges interpretive and analytical traditions. This article concludes by providing exemplifications of bringing scholastically unrealized SA in geographic theories out of obscurity, together with certain salient benefits from doing so, affirming the magnitude of fulfilling its major objective: SA is poised for discovery in all geospatial theories, from those for human and humanistic geography, through physical geography, to those for cartography as well as methodologies concerning all georeferenced data collection missions. Full article

Plastic pollution poses severe threats to ecosystems, human health, and economies as plastics fragment into macro- and microplastics that accumulate across marine and terrestrial environments. Conventional monitoring is constrained by scale, cost, and resources, particularly in under-resourced regions, whereas citizen science provides an inclusive, community-driven alternative for data collection, analysis, and remediation to support evidence-based policy. This systematic review advances the field through three novel contributions: a refined participatory typology that explicitly prioritizes co-creative models for equitable engagement in the Global South; the first comprehensive synthesis of direct citizen involvement in plastic bioremediation, including community microbial isolation, household biodegradation trials, and real-world testing of biodegradable materials; and a new conceptual framework positioning citizen science as the central nexus linking upstream prevention, technological innovation, bioremediation, and global governance. Findings highlight large-scale geotagged datasets, behavioral change, and policy influence, while persistent challenges include data standardization, digital exclusion, and Global North bias. We therefore advocate institutional mainstreaming through dedicated policy offices, decolonial integration of indigenous knowledge, and hybrid citizen–lab validation pipelines, especially in underrepresented regions such as Africa, establishing citizen science as a transformative mechanism for participatory and equitable responses to escalating plastic pollution. Full article

Urban parks are fundamental to building sustainable and inclusive cities, yet the mechanisms linking their spatial configuration to human activities and social functions remain insufficiently understood. A scalable multimodal framework is developed to quantify how spatial visibility is associated with patterns of park use and the provision of social ecosystem services. A total of 94,635 geo-tagged user-generated images from 148 parks in Vancouver, Canada, were analyzed using the Contrastive Language-Image Pretraining (CLIP) model to classify user activities into six behavioral categories. Concurrently, airborne LiDAR data and space syntax analysis were used to derive three visibility metrics—Mean Isovist Area (MIA), reflecting internal openness; Mean Visual Integration (MVI), indicating visual connectivity within the park interior; and External Isovist Ratio (EIR), representing edge openness and boundary visibility. The results indicate that EIR exhibits the strongest and most consistent relationships with user activity patterns, positively associated with family recreation, social vibrancy, and physical activity, while negatively linked to nature immersion and quiet relaxation. MIA shows moderate associations with socially interactive and child-oriented activities, whereas MVI contributes little explanatory power compared to localized visibility conditions. These findings highlight spatial visibility as a critical design attribute that is closely associated with human–forest interactions. By illustrating that moderate visual openness and edge permeability are associated with more inclusive and multifunctional patterns of park use, actionable insights are provided for urban park planning and design, and the promotion of social sustainability. Full article

A few countries have requested open IP locations of posters in order to combat rumors and strengthen management. Such policies intensify information surveillance of users, which may in turn influence their online behavior. In the context of multiple governments considering the implementation of this policy, it is essential to assess its impact. We examine the impact of IP location openness on posters’ behavior and patterns based on the empirical data of Sina Weibo, and analyze the heterogeneous impact on users of different genders. Regression discontinuity and short-run panel data regression results show that IP location openness reduces the frequency of users’ social media participation behavior; specifically, the frequency of reposting microblogs and posting geo-tagged microblogs is remarkably diminished, while the frequency of posting photos is not discernibly changed. Long-run panel data regression results indicate that the overall inhibitory effect on the frequency of social media participation behavior disappears, and it only has a negative effect on posting geo-tagged microblogs. The results of heterogeneity analysis suggest that the short-run negative impact of IP location openness on female users’ social media participation behavior is more remarkable than that of male users. Full article

Invasive alien species (IASs) pose escalating threats to global ecosystems, biodiversity, and human well-being. Public participation in IAS monitoring is often limited by taxonomic expertise gaps. To address this, we established a multi-taxa image dataset covering 54 key IAS in China, benchmarked nine deep learning models, and quantified impacts of varying scenarios and target scales. EfficientNetV2 achieved superior accuracy, with F1-scores of 83.66% (original dataset) and 93.32% (hybrid dataset). Recognition accuracy peaked when targets occupied 60% of the frame against simple backgrounds. Leveraging these findings, we developed EyeInvaS, an AI-powered system integrating image acquisition, recognition, geotagging, and data sharing to democratize IAS surveillance. Crucially, in a large-scale public deployment in Huai’an, China, 1683 user submissions via EyeInvaS enabled mapping of Solidago canadensis, revealing strong associations with riverbanks and roads. Our results validate the feasibility of deep learning in empowering citizens in IAS surveillance and biodiversity governance. Full article

This study examines the statistical and spatial alignment between urban place perceptions and the census-based evidence of socio-spatial segregation. We process a large dataset of geotagged images from Mapillary and KartaView with ZenSVI to score six place perception dimensions (safety, liveliness, wealth, beauty, boredom, depression) for the metropolitan area of Thessaloniki, Greece. The socio-economic structure is derived from census indicators and property values using Location Quotients and principal component analysis. We assess alignment through Pearson’s correlation (r) to capture statistical association, and bivariate Moran’s I to test spatial correspondence while accounting for spatial dependence. Results reveal a robust northwest–southeast divide: southeastern and central districts are perceived as safer, livelier, wealthier, and more beautiful, while northwestern and industrial zones score higher on boredom and depression. The historic city center emerges as vibrant and affluent, acting as a key interface between social groups, especially students, the elderly, and migrants. Perceptual dimensions vary in spatial form: safety, beauty, and depression cluster locally, whereas wealth and vibrancy extend over broader sectors. The study demonstrates the combined use of perceptual and socio-economic data for urban analysis and provides a replicable framework for monitoring inequalities and guiding participatory and inclusive planning. Full article

In the era of big data, the rapid proliferation of user-generated content enriched with geolocations offers new perspectives and datasets for probing the spatiotemporal dynamics of tourist mobility. Mining large-scale geospatial traces has become central to tourism geography: it reveals preferences for attractions and routes to enable intelligent recommendation, enhance visitor experience, and advance smart tourism, while also informing spatial planning, crowd management, and sustainable destination development. Using Mount Huangshan—a UNESCO World Cultural and Natural Heritage site—as a case study, we integrate GPS trajectories and geo-tagged photographs from 2017–2023. We apply a Density-Field Hotspot Detector (DF-HD), a Space–Time Cube (STC), and spatial gridding to analyze behavior from temporal, spatial, and fully spatiotemporal perspectives. Results show a characteristic “double-peak, double-trough” seasonal pattern in the number of GPS tracks, cumulative track length, and geo-tagged photos. Tourist behavior exhibits pronounced elevation dependence, with clear vertical differentiation. DF-HD efficiently delineates hierarchical hotspot areas and visitor interest zones, providing actionable evidence for demand-responsive crowd diversion. By integrating sequential time slices with geography in a 3D framework, the STC exposes dynamic spatiotemporal associations and evolutionary regularities in visitor flows, supporting real-time crowd diagnosis and optimized spatial resource allocation. Comparative findings further confirm that Huangshan’s seasonal intensity is significantly lower than previously reported, while the high agreement between trajectory density and gridded photos clarifies the multi-tier clustering of route popularity. These insights furnish a scientific basis for designing secondary tour loops, alleviating pressure on core areas, and charting an effective pathway toward internal structural optimization and sustainable development of the Mount Huangshan Scenic Area. Full article

In high-density built environments, perceived density (PD)—shaped by physical, socio-cultural, and perceptual factors—often induces sensations of crowding, stress, and spatial oppression. Although green spaces are recognised for their stress-reducing effects, the influence of built-environment characteristics on public sentiment under stringent mobility restrictions remains inadequately explored. This study takes Chongqing, a representative mountainous metropolis in China, as a case to examine how natural and built environmental elements modulate emotional valence across varying PD levels. Using housing data (n = 4865) and geotagged Weibo posts (n = 120,319) collected during the 2022 lockdown, we constructed a PD-sensitive sentiment dictionary and applied Python’s Jieba package and natural language processing (NLP) techniques to analyse emotional scores related to PD. Spatial and bivariate autocorrelation analyses revealed clustered patterns of sentiment distribution and their association with physical density. Using entropy weighting, building density and floor area ratio were integrated to classify residential built environments (RBEs) into five tiers based on natural breaks. Key factors influencing positive sentiment across PD groups were identified through Pearson correlation heatmaps and OLS regression. Three main findings emerged: (1) Although higher-PD areas yielded a greater volume of positive sentiment expressions, they exhibited lower diversity and intensity compared to low-PD areas, suggesting inferior emotional quality; (2) Environmental and socio-cultural factors showed limited effects on sentiment in low-PD areas, whereas medium- and high-PD areas benefited from a significantly enhanced cumulative effect through the integration of socio-cultural amenities and transportation facilities—however, this positive correlation reversed at the highest level (RBE 5); (3) The model explained 20.3% of the variance in positive sentiment, with spatial autocorrelation effectively controlled. These findings offer nuanced insights into the nonlinear mechanisms linking urban form and emotional well-being in high-density mountainous settings, providing theoretical and practical guidance for emotion-sensitive urban planning. Full article

Amid growing demands for heritage conservation and precision urban governance, this study proposes a multimodal framework to analyze tourist perception and behavior in Macau’s Historic Centre. We integrate geotagged social media images and text, ultra-wideband (UWB) pedestrian trajectories, and a LiDAR-derived 3D digital twin to examine the interplay among spatial configuration, movement, and affect. Visual content in tourist photos is classified with You Only Look Once (YOLOv8), and sentiment polarity in Weibo posts is estimated with a fine-tuned Bidirectional Encoder Representations from Transformers (BERT) model. UWB data provide fine-grained trajectories, and all modalities are georeferenced within the digital twin. Results indicate that iconic landmarks concentrate visual attention, pedestrian density, and positive sentiment, whereas peripheral sites show lower footfall yet strong emotional resonance. We further identify three coupling typologies that differentiate tourist experiences across spatial contexts. The study advances multimodal research on historic urban centers by delivering a reproducible framework that aligns image, text, and trajectory data to extract microscale patterns. Theoretically, it elucidates how spatial configuration, movement intensity, and affective expression co-produce experiential quality. Using Macau’s Historic Centre as an empirical testbed, the findings inform heritage revitalization, wayfinding, and crowd-management strategies. Full article

Social media platforms like Instagram significantly shape destination images and influence tourist behavior. Understanding how different cities are represented and perceived on these platforms is crucial for effective tourism marketing. This study provides a comparative analysis of Instagram content and engagement patterns in Seoul and Tokyo, two major Asian metropolises, to derive actionable marketing insights. We collected and analyzed 59,944 public Instagram posts geotagged or location-tagged within Seoul (n = 29,985) and Tokyo (n = 29,959). We employed a mixed-methods approach involving content categorization using a fine-tuned convolutional neural network (CNN) model, engagement metric analysis (likes, comments), Valence Aware Dictionary and sEntiment Reasoner (VADER) sentiment analysis and thematic classification of comments, geospatial analysis (Kernel Density Estimation [KDE], Moran’s I), and predictive modeling (Gradient Boosting with SHapley Additive exPlanations [SHAP] value analysis). A validation analysis using balanced samples (n = 2000 each) was conducted to address Tokyo’s lower geotagged data proportion. While both cities showed ‘Person’ as the dominant content category, notable differences emerged. Tokyo exhibited higher like-based engagement across categories, particularly for ‘Animal’ and ‘Food’ content, while Seoul generated slightly more comments, often expressing stronger sentiment. Qualitative comment analysis revealed Seoul comments focused more on emotional reactions, whereas Tokyo comments were often shorter, appreciative remarks. Geospatial analysis identified distinct hotspots. The validation analysis confirmed these spatial patterns despite Tokyo’s data limitations. Predictive modeling highlighted hashtag counts as the key engagement driver in Seoul and the presence of people in Tokyo. Seoul and Tokyo project distinct visual narratives and elicit different engagement patterns on Instagram. These findings offer practical implications for destination marketers, suggesting tailored content strategies and location-based campaigns targeting identified hotspots and specific content themes. This study underscores the value of integrating quantitative and qualitative analyses of social media data for nuanced destination marketing insights. Full article

The task of visual geo-localization based on street-view images estimates the geographical location of a query image by recognizing the nearest reference image in a geo-tagged database. This task holds considerable practical significance in domains such as autonomous driving and outdoor navigation. Current approaches typically use perspective street-view images as reference images. However, the lack of scene content resulting from the restricted field of view (FOV) in such images is the main cause of inaccuracies in matching and localizing the query and reference images with the same global positioning system (GPS) labels. To address this issue, we propose a perspective-to-panoramic image visual geo-localization framework. This framework employs 360° panoramic images as references, thereby eliminating the issue of scene content mismatch due to the restricted FOV. Moreover, we propose the structural feature enhancement (SFE) module and integrate it into LskNet to enhance the ability of the feature extraction network to capture and extract long-term stable structural features. Furthermore, we propose the adaptive scene alignment (ASA) strategy to address the issue of data capacity and information content asymmetry between perspective and panoramic images, thereby facilitating initial scene alignment. In addition, a lightweight feature aggregation module, MixVPR, which considers spatial structure relationships, is introduced to aggregate the scene-aligned region features into robust global feature descriptors for matching and localization. Experimental results demonstrate that the proposed model outperforms current state-of-the-art methods and achieves R@1 scores of 72.5% on the Pitts250k-P2E dataset and 58.4% on the YQ360 dataset, indicating the efficacy of this approach in practical visual geo-localization applications. Full article

Accurate crop yield modeling (CYM) is inherently challenging due to the complex, nonlinear, and temporally dynamic interactions of biotic and abiotic factors. Crop traits, which historically capture the cumulative effect of these factors, exhibit functional relationships critical for optimizing productivity. This underscores the necessity of multi-trait-based CYM approaches. Crop growth models enable trait dynamics with reflectance data and spectral indices as proxies for crop health and traits, respectively, to have real-time, spatially explicit monitoring. The Agricultural Production Systems sIMulator was calibrated to simulate multiple traits across the growth season based on geo-tagged wheat field ground information. Reflectance and spectral indices were processed for the geo-tagged fields across temporal observations to enable real-time, spatially explicit monitoring. Based on these parameters, this study addresses a critical gap in existing CYM frameworks by proposing a machine learning-based model that synergized multiple crop traits with reflectance and spectral indices to generate site-specific yield estimates. The performance evaluation revealed that the Long Short-Term Memory (LSTM) model achieved superior accuracy for the integrated parameters (RMSE = 250.68 kg/ha, MAE = 193.76 kg/ha, and R² = 0.84), followed by traits alone. The Random Forest model followed the LSTM model, with an RMSE = 293.56 kg/ha, MAE = 230.68 kg/ha, and R² = 0.78 for integrated parameters, and an RMSE = 291.73 kg/ha, MAE = 223.17 kg/ha, and R² = 0.78 for crop traits. The superior prediction demonstrated the dominant role of multiple crop traits with satellite-derived reflectance metrics to develop robust CYM frameworks capable of capturing intra- and inter-field yield variability. Full article

Urban diversity is essential for promoting urban vitality and achieving sustainable urban development. However, existing studies rely on static and non-visual data and seldom incorporate human perception aspects in the diversity estimation. Together with the modifiable areal unit problem (MAUP) in the traditional entropy-based approach, urban diversity is prone to be biased or underestimated. In this study, we use urban function (from POI) and visual semantics (from geo-tagged photos) to estimate what we call “perceived urban diversity”. More importantly, we propose to improve the traditional entropy-based diversity measures by addressing the MAUP issue using area- and accessibility-based extensions. Empirical analysis using Shenzhen, China, as a case study reveals that our “perceived diversity” indicators display stronger correlations to urban vitality. Furthermore, combining different data sources (e.g., geo-tagged photos) provides a more comprehensive portrayal of urban diversity. Finally, our results suggest that neighborhoods dominated by residential or commercial land uses would benefit the most from enhanced diversity. These findings are useful for a refined assessment of urban diversity and offer valuable insights for urban planning and community design. Full article

Timely, region-based geo-maps like choropleths are essential for smart city applications like traffic monitoring and urban planning because they can reveal statistical patterns in geotagged data. However, because data overloading is brought on by the quick inflow of massive geospatial data, creating these visualizations in real time presents serious difficulties. This paper introduces ApproxGeoMap, a novel system designed to efficiently generate approximate geo-maps from fast-arriving georeferenced data streams. ApproxGeoMap employs a stratified spatial sampling method, leveraging geohash tessellation and Earth Mover’s Distance (EMD) to maintain both accuracy and processing speed. We developed a prototype system and tested it on real-world smart city datasets, demonstrating that ApproxGeoMap meets time-based and accuracy-based quality of service (QoS) constraints. Results indicate that ApproxGeoMap significantly enhances efficiency in both running time and map accuracy, offering a reliable solution for high-speed data environments where traditional methods fall short. Full article