XZXinxin Zhou

Xinxin Zhou

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Xinxin Zhou has a total of 57 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Oblique drone panoramic imagery for fine-scale human-induced land-cover change detection

    by Xinxin Zhou, Yulin Xiao, Runtian Wang, Yuanpei Gou, Hongjun Ke, Zhenjun Yan, Yuanyuan Guo

    High-Resolution Spatiotemporal Monitoring of Seasonal Land-Use and Land-Cover (LULC) Changes (HRST-MSLC) holds broad application prospects. However, existing unmanned aerial vehicle (UAV) change-detection benchmarks are predominantly based on nadir-view, bi-temporal imagery, providing limited lateral structural information and seasonal observations, while data-acquisition costs and multi-temporal registration constrain dataset construction. To address this, this paper introduces the Unmanned Aerial Vehicle Oblique Multi-View Panoramic Change Detection Dataset (UAV-OMVCD) for fine-scale human-induced LULC change monitoring. It contains 864 panoramic observations acquired in four study areas across Nanjing, Hefei, and Tangshan using DJI Matrice 4D/4TD and Matrice 4E platforms. Each panorama is exported at 1448 × 7200 pixels, with nominal nadir-equivalent source-frame ground sampling distances (GSDs) of approximately 3.7–8.5 cm, yielding 2,181 paired 512 × 512 samples. UAV-OMVCD documents seasonal changes in impervious surfaces, engineering machinery, buildings, greenhouses, and piled soil, including structural additions/removals and morphological evolution. Its oblique views capture lateral facades and occluded areas while reducing seasonal pseudo-change interference. Five benchmark models achieved overall accuracies of 0.9429–0.9492, macro-averaged F1-scores of 0.8861–0.9007, and macro-averaged Intersection over Union (IoU) values of 0.8049–0.8266. UAV-OMVCD provides a standardized benchmark for fine-scale human-induced land-cover change detection using oblique drone panoramic imagery.

  • A Location Model for Air–Ground Intermodal VTOL Vertiports Network Integrating Crowd Origin–Destination Flows: A Case Study in Nanjing

    by Xinxin Zhou, Dongle Zhao, Runtian Wang, Yong Tian, Jianwu Xu, Changbin Wu

    The configuration of urban low‐altitude vertiports directly influences the efficiency and convenience of urban air mobility (UAM) transportation. This study aims to optimize the urban low‐altitude VTOL vector network (VVN) by integrating urban commuting and air–ground intermodal transportation. This study develops a data‐driven framework for city‐scale eVTOL aircraft vector network planning by integrating multi‐source geospatial data, mobile signaling‐based crowd OD flows, and an air–ground intermodal location model. Building on existing network design formulations, this study uses crowd OD flows rather than individual trips as the unit of analysis. More importantly, it proposes a pruning‐and‐reduction strategy that makes the resulting mixed‐integer optimization computationally tractable with massive amounts of observed OD data. The empirical analysis takes Nanjing, a first‐tier city in China, as the research area, integrating multi‐source large‐scale geospatial data, especially mobile signaling origin–destination (OD) flow data. The experimental results show that the pruning strategy reduces the variable scale by 99.79% while preserving the near‐equivalent solution quality. The optimized VVN layout significantly improves long‐distance trip service accessibility and overall benefits. With respect to air–ground intermodal trips, the flight segment accounts for approximately 16% of the total travel time, whereas the first and last‐mile connection times account for 84%. Therefore, it is necessary to optimize accessibility between residential communities and vertiport stations, improve interconnections for ground transportation, and enhance the construction of parking facilities at vertiports. This study implements data‐driven modeling by fusing multi‐source spatiotemporal data and coupling air–ground intermodal trip transfer/connection decisions, providing methodological references for optimizing VVN location selection in multi‐modal transportation coordination.

  • Collaborative conservation zoning with human footprint index integration: A case study of Haixi Prefecture, Tibetan Plateau

    by Xinxin Zhou, Zhenbang Wang, Yulin Xiao, Changzheng Liu, Jian Shen, Lvhua Wang

    With the rapid development of the Tibetan Plateau, various human activities have increasingly extended into ecologically vulnerable areas with low‐intensity human disturbance, leading to aggravated habitat fragmentation. This poses severe challenges to ecological conservation efforts on the plateau. Therefore, incorporating diverse human activity factors into the planning of plateau ecological reserves to formulate proactive and quantitative conservation plans has become an issue that warrants urgent investigation. 'This study integrates land use/land cover, climate, GDP and human footprint data to quantify ecosystem services, model biodiversity distributions and optimize spatial conservation zones using a combined InVEST‐MaxEnt‐Marxan model. It calculates grid‐based carbon stocks for planning units, simulates biodiversity distribution, identifies key environmental variables, constructs four distinct cost objective functions and optimizes diverse conservation planning scenarios. The study established four cost‐based scenario cases using Haixi Prefecture as the experimental area: Case 1 (single‐dimensional natural factors), Case 2 (single‐dimensional socioeconomic factors), Case 3 (multi‐dimensional GDP and human footprint weighting) and Case 4 (integrated natural, socioeconomic and human footprint weighting). Based on comparative analysis of the experimental results, the ‘ecological‐economic’ synergistic development scheme demonstrated significant advantages. Integrating human activity footprint data into the ecological conservation planning of the Tibetan Plateau enables more precise identification of the spatial distribution of areas with high‐intensity human disturbance. By avoiding such areas in conservation planning, originally isolated ecological patches can be reconnected, thereby mitigating the ‘ecological island’ effect—referring to fragmented, isolated habitats severed by human activities. Meanwhile, human footprint data can quantify the degree of anthropogenic disturbance in each grid cell, allowing grids with low disturbance and high ecological value to be designated as priority conservation zones, thus enhancing the precision of planning. Synthesis and applications . This study confirms that integrating human footprint data enhances the practical applicability and ecological connectivity of plateau conservation planning. The ‘ecological‐economic’ synergistic scheme, with fewer units, high irreplaceability overlap and good connectivity, offers direct references for Haixi Prefecture and similar plateau regions. The combined InVEST‐MaxEnt‐Marxan model also provides a replicable tool for spatiotemporal data‐driven conservation planning in ecologically sensitive areas.

  • Open Access

    Integrating Scenario Forecasting with SPNN-AtGNNWR for China’s Carbon Peak Pathway Projection

    by Lizhi Miao, Heng Xu, Xinkai Feng, Jvmin Wang, Sheng Tang, Xinxin Zhou, Xiying Sun, Gang Lu, Mei-Po Kwan

    As the world’s leading carbon emitter, China’s ability to reach its pledged carbon peak by 2030 is pivotal for its own green transition and global climate governance. This research proposes a novel integration of spatial proximity neural networks with attention-enhanced geographically weighted neural network regression. This new model integrates spatial dependencies and an attention mechanism into the traditional geographically weighted neural network regression framework. The model demonstrates good performance in forecasting carbon emissions (coefficient determination = 0.904, root mean square error = 48.927). Using this model, alongside population, GDP, total energy consumption, and other influencing factors, the research integrated scenario forecasting to project China’s total carbon emissions from 2023 to 2040. Three policy-relevant scenarios—baseline, low-carbon, and extensive development—were set to forecast and analyze various potential outcomes under uncertain conditions. Under the baseline scenario, China’s emissions peak in 2029 at 9926.26 Mt; the low-carbon scenario advances the peak to 2027 at 9688.88 Mt; whereas an extensive growth path delays the peak to 2032 at 10,347.70 Mt. These findings underscore the urgency of optimizing energy structure, curbing fossil fuel dependence, and balancing economic growth with the deep decoupling of emissions. This research offers policymakers a robust, spatially explicit tool for evaluating future trajectories under diverse development pathways.

  • Open Access

    CAU2DNet: A Dual-Branch Deep Learning Network and a Dataset for Slum Recognition with Multi-Source Remote Sensing Data

    by Xi Lyu, Chenyu Zhang, Lizhi Miao, Xiying Sun, Xinxin Zhou, Xinyi Yue, Zhongchang Sun, Yueyong Pang

    The efficient and precise identification of urban slums is a significant challenge for urban planning and sustainable development, as their morphological diversity and complex spatial distribution make it difficult to use traditional remote sensing inversion methods. Current deep learning (DL) methods mainly face challenges such as limited receptive fields and insufficient sensitivity to spatial locations when integrating multi-source remote sensing data, and high-quality datasets that integrate multi-spectral and geoscientific indicators to support them are scarce. In response to these issues, this study proposes a DL model (coordinate-attentive U2-DeepLab network [CAU2DNet]) that integrates multi-source remote sensing data. The model integrates the multi-scale feature extraction capability of U2-Net with the global receptive field advantage of DeepLabV3+ through a dual-branch architecture. Thereafter, the spatial semantic perception capability is enhanced by introducing the CoordAttention mechanism, and ConvNextV2 is adopted to optimize the backbone network of the DeepLabV3+ branch, thereby improving the modeling capability of low-resolution geoscientific features. The two branches adopt a decision-level fusion mechanism for feature fusion, which means that the results of each are weighted and summed using learnable weights to obtain the final output feature map. Furthermore, this study constructs the São Paulo slums dataset for model training due to the lack of a multi-spectral slum dataset. This dataset covers 7978 samples of 512 × 512 pixels, integrating high-resolution RGB images, Normalized Difference Vegetation Index (NDVI)/Modified Normalized Difference Water Index (MNDWI) geoscientific indicators, and POI infrastructure data, which can significantly enrich multi-source slum remote sensing data. Experiments have shown that CAU2DNet achieves an intersection over union (IoU) of 0.6372 and an F1 score of 77.97% on the São Paulo slums dataset, indicating a significant improvement in accuracy over the baseline model. The ablation experiments verify that the improvements made in this study have resulted in a 16.12% increase in precision. Moreover, CAU2DNet also achieved the best results in all metrics during the cross-domain testing on the WHU building dataset, further confirming the model’s generalizability.

  • Open Access

    Ecological Zoning Based on Suitability Evaluation of Ecological Product Development from the Value-Risk-Cost-Demand Perspective

    by Ming Gao, Pei Du, Xinxin Zhou, Zhenxia Liu, Wen Luo, Zhaoyuan Yu, Linwang Yuan

    Current ecological value assessment models predominantly emphasize the potential value of ecological resources, neglecting the crucial aspect of value realization processes. Analyzing the value of ecological resources from the perspective of ecological products (EPs) is more instructive in realizing ecological values. The key factors controlling the realization of ecological product value are potential value, ecological risk, development costs, and human demand. Previous research has rarely integrated these four factors within the ecological zoning framework. This study proposes a suitability evaluation and zoning framework for ecological product development based on the “value-risk-cost-demand” perspective. First, an evaluation index system for the potential value of ecological products was developed, dividing EPs into ecological agriculture (EA), ecological industry (EI), and ecological tourism (ET), and assessing them using 13 indicators. Ecological risks were modeled using multi-scale patch analysis (MSPA) and other models. Development costs were estimated using cost entropy. The impact of population dynamics on EP demand was quantified using population density, night-time light data, and average land GDP, along with stacked buffer analysis. Next, an improved TOPSIS method was applied to integrate these four dimensions, producing a comprehensive suitability assessment for EP development. Finally, EP zoning was determined by overlaying the comprehensive evaluation results. This framework was used to identify the dominant mode zones of EPs within the region of Jintan District, Jiangsu Province, China. The findings suggest that the integrated assessment model proposed in this study has produced more reasonable outcomes in terms of spatial layout, land use area, reduction of fragmentation and ecological risk. This conclusion is supported by spatial distribution comparisons, optimal area deviation analyses, landscape index calculations and multi-model driven future simulations. This model effectively resolves the spatial mismatch present in the traditional approach, which solely focuses on the potential value of EPs. This study can be applied to other regions with developed economies and rich ecological resources, providing an effective reference for the choice of paths to realize the value of EPs.


  • Open Access

    Investigating spatiotemporal patterns for comprehensive accessibility of service facilities by location-based service data in Nanjing (2016–2022)

    by Yi Huang, Dongle Zhao, Can Ma, Kaidong Gu, Xinxin Zhou, Yuting Jiang

    The annual spatiotemporal variation for the comprehensive accessibility of service facilities (CASF), expressing the dynamic development of urban service, has not been routinely monitored adequately (such as multi-type facility interactions, policy-driven dynamics, and persistent inequities). To address this gap, we propose three key innovations: (1) Utility-improved accessibility model: A novel framework integrating 15 facility types through entropy-weighted utility indices, overcoming the single-type limitation of traditional models. This model dynamically couples facility diversity (point of interest density), population demand (mobile big data), and time-decayed travel impedance (route planning application programming interface), enabling multi-dimensional CASF evaluation. (2) Integrated spatiotemporal methodology: We synergistically combine four methods, standard deviational ellipse (directional trends), space-time cube (3D continuity), emerging hotspot analysis (cluster evolution), and time series clustering (trajectory phases) to capture multi-year interactions between policy-driven urban expansion and accessibility dynamics. (3) Policy-ready findings: Our analysis of Nanjing (2016–2022) reveals that CASF exhibits multi-center agglomeration (centroid shift toward Jiangbei New District) with divergent trajectories persistent hotspots in historic cores (Gulou District) vs stagnant peripheries (northern Qixia). These patterns directly align with Nanjing’s 2021–2035 master plan. The framework provides actionable insights for policymakers to prioritize investments in intensifying hotspots (e.g., Jiangbei’s commercial hubs), address systemic deprivation in persistent cold spots (e.g., Qixia’s underserved areas), and monitor phased impacts of transportation projects through CASF trajectory clusters.

  • Integrating geospatial data and street‐view imagery to reconstruct large‐scale 3D urban building models

    by Changbin Wu, Xinyang Yu, Can Ma, Rongkai Zhong, Xinxin Zhou

    3D urban building modeling is a vital foundational step for building Digital Twins and Smart Cities. In response to existing challenges, such as high time costs, complex production processes, and low consistency with real‐world textures in large‐scale 3D urban building modeling methods, this research proposes a reconstructing 3D urban building models (3DUBM) approach that integrates geospatial data and street view. The approach achieves an enhanced generation of large‐scale 3DUBMs. Based on open geospatial data and street‐view imagery (SVI), the approach was tested in modeling experiments conducted in Shanghai, Hongkong, and Nanjing. Furthermore, a dataset covering unique blocks of 30 cities in China was constructed to demonstrate the approach's characteristics of large coverage, high time efficiency, high model quality and low economic cost. The accuracy of texture mapping from SVI to 3DUBM reached 85%. This achievement has significant economic value in bridging the gap in the production of large‐scale and low‐cost 3DUBM data, promoting the construction of Digital Twins, Smart Cities, and Real‐world 3D modeling.


  • Open Access

    Spatiotemporal Data-Driven Multiperiod Relocation Optimization of Emergency Medical Services: Maximum Equality Objective

    by Xinxin Zhou, Xinxin Zhou, Yujie Chen, Yujie Chen, Yingying Li, Yingying Li, Bingjie Liu, Bingjie Liu, Zhaoyuan Yu, Zhaoyuan Yu

    As a kind of first aid healthcare service, emergency medical services (EMSs) present high spatiotemporal sensitivity due to significant changes in the time-dependent urban environment. Taking full advantage of big spatiotemporal data to realize multiperiod relocation optimization of EMSs can reduce idle resources and improve service utilization efficiency and fairness. First, we established the dynamic time-dependent accessibility and equality model to formulate the multiperiod maximization objective of global equality. Second, we proposed a capacitated integer evolution algorithm that relocates emergency medical vehicles to optimize the scheduling scheme. Based on multiperiod mobile phone records and multiperiod online route planner data, the equality of EMSs in the research metropolis, Nanjing, China, rose by 41.5% on average, which has an incentivizing effect on alleviating the tension of prehospital service and minimizes accessibility disparities without constructing more infrastructure. We also created maps to visualize the changes in equality patterns over time. This relocation optimization approach can be regarded as a trade-off approach to dispatch time-dependent sensitive services and provide a practical tool for healthcare decision-makers to evaluate public healthcare systems and improve strategic urban service planning.