Challenge Title
Remote Sensing-Based Dynamic Monitoring of Urban Expansion and Analysis of Driving Forces
Challenge Description：Cities concentrate large populations, industries, and resources, and their spatiotemporal development patterns exert a profound influence on regional ecological security and socioeconomic development. This challenge requires participating teams to select a representative city within a region as the study area. Using long-term terrestrial remote-sensing image series, teams shall quantitatively extract information on urban impervious surfaces and construct a standardized long-term urban-development dataset. On this basis, teams shall analyze the spatiotemporal evolution of the built-up area, including the speed, scale, and direction of urban expansion, and are encouraged to use AI-assisted analytical methods to identify patterns of spatiotemporal differentiation. Teams shall further select multidimensional driving factors relating to the economy, society, policy, and the natural environment, and use methods such as correlation analysis, geographical detectors, and machine-learning attribution models to quantify the influence of each factor on urban expansion and the coupled interactions among those factors. The final results should provide scientific data support and decision-making references for regional ecological protection and high-quality development, coordinated planning among coastal cities, and optimized territorial spatial layouts.
Core Requirements：
1. Data Processing: Complete the preprocessing of long-term terrestrial remote-sensing image series, including radiometric calibration, atmospheric correction, image clipping, and mosaicking. Accurately extract key remote-sensing parameters such as urban impervious surfaces and construct a standardized long-term urban-development dataset.
2. Spatiotemporal Analysis and Pattern Discovery: Quantitatively reveal the evolutionary characteristics of built-up-area expansion in coastal cities, including expansion speed, scale, and direction. Use AI-assisted analytical methods to examine patterns of spatiotemporal differentiation and spatial diffusion.
3. Intelligent Analysis of Driving Factors: Select multidimensional driving factors relating to the economy, society, policy, and the natural environment. Apply correlation analysis, geographical detectors, machine-learning attribution models, and other methods to quantify the degree to which each factor affects urban expansion.
Submission Requirements：
1. A solution document in PDF format (approximately 5,000 Chinese characters), including an analysis of current conditions and problems in the study area; data sources and preprocessing; core technical methods, including long-term remote-sensing image processing, impervious-surface extraction, and AI-assisted analysis; the spatiotemporal evolution of urban expansion and identification of shortcomings; quantitative analysis of driving factors; coordinated planning measures and optimization recommendations; and an assessment of implementation value.
2. A results-demonstration video of no more than eight minutes, or a PowerPoint presentation, clearly showing the remote-sensing data-processing workflow, urban-expansion monitoring results, visualization of spatiotemporal evolution, the driving-factor analysis process, and the effects of the proposed optimization plan.
3. Optional materials: spatial-analysis outputs such as maps of the spatiotemporal evolution of urban expansion, impervious-surface distribution maps, and driving-factor analysis maps; and an implementation plan including recommendations for optimizing territorial space and a coordinated planning proposal.
Challenge Title
Urban Land Subsidence Monitoring and Multidimensional Driving-Factor Analysis
Challenge Description：With rapid urban development, cities commonly experience land subsidence to varying degrees as a result of the combined effects of geological structures, excessive groundwater extraction, engineering construction, urban expansion, soil erosion, and other factors. As a typical slow-onset geological hazard, land subsidence can cause ground collapse, wall cracking, infrastructure damage, and reduced flood-control capacity, thereby posing a serious threat to the safety of urban territorial space, ecological stability, and high-quality development. This challenge requires participating teams to select a representative city as the study area. Using long-term SAR satellite data such as Sentinel-1, together with multi-source auxiliary data relating to geological surveys, hydrological monitoring, socioeconomic conditions, urban construction, and policy regulation, teams shall apply time-series InSAR techniques to accurately retrieve land-subsidence information. Teams are also encouraged to integrate AI to improve the accuracy of deformation identification and the analysis of driving mechanisms, and to develop a forward-looking, practical, and scalable solution for urban land-subsidence monitoring and risk assessment.
Core Requirements：
1. SAR Data Processing and AI Integration: Complete the preprocessing and interferometric processing of long-term SAR image series. Apply time-series InSAR techniques, including SBAS-InSAR, PS-InSAR, or DS-InSAR, to extract parameters such as subsidence rates and cumulative subsidence. The introduction of AI algorithms, such as time-series forecasting and anomaly detection, is encouraged to improve the accuracy of deformation identification. Construct a high-precision long-term subsidence-monitoring dataset and complete accuracy validation.
2. Detailed Analysis of the Spatiotemporal Evolution of Subsidence: Quantitatively reveal the spatial distribution, severity levels, concentration ranges, expansion directions, and temporal evolution of subsidence. Examine changes across different periods, including increases or decreases in subsidence rates, expansion or contraction of affected areas, and migration of the subsidence center. Accurately classify stable areas and areas of slight, moderate, and severe subsidence. Summarize the spatiotemporal differentiation, evolutionary stages, and diffusion patterns of land subsidence in the selected city, and clarify intra-urban variations in subsidence.
3. Multidimensional Intelligent Analysis of Driving Factors: Select natural factors, including geology, topography, groundwater, and precipitation, as well as anthropogenic factors, including groundwater extraction, urban construction, transportation, mining, population, and economic activity. Use correlation analysis, geographical detectors, machine-learning attribution models, and other methods to quantify the independent and interactive effects of these factors.
Submission Requirements：
1. A solution document in PDF format (approximately 5,000 Chinese characters), including an analysis of current conditions and problems in the study area; data sources and preprocessing; core technical methods, including time-series InSAR and AI-assisted analysis; the spatiotemporal evolution of land subsidence and identification of shortcomings; quantitative analysis of driving factors; prevention and control measures and optimization recommendations; and an assessment of implementation value.
2. A results-demonstration video of no more than eight minutes, or a PowerPoint presentation, clearly showing the SAR data-processing workflow, subsidence-monitoring results, visualization of spatiotemporal evolution, the driving-factor analysis process, and the effects of the proposed optimization plan.
3. Optional materials: spatial-analysis outputs such as subsidence-rate distribution maps, spatiotemporal evolution maps, and driving-factor analysis maps; and an implementation plan including a monitoring and early-warning scheme and recommended prevention and control measures.
03
Challenge Title
“AI+”-Empowered Collaborative Space-Air-Ground Intelligent Urban Health Assessment and Applications
Challenge Description：Urban health assessment is an important means of advancing new smart-city development and achieving refined urban management. It is intended to support routine monitoring, diagnosis, and evaluation of the urban living environment, infrastructure, ecological environment, building safety, transportation operations, and other aspects of city performance. This challenge explores the use of multi-source space-air-ground data to conduct intelligent sensing, diagnostic assessment, and decision support around the core indicators of urban health assessment.
Core Requirements：
1. Scenario Innovation: Address key urban-governance challenges in urban health assessment and diagnostic needs relating to building safety, energy consumption, the urban heat-island effect, public-space quality, infrastructure operations, and other areas. Highlight the distinctive role of remote-sensing technology in dynamic spatiotemporal monitoring and spatial analysis, and propose innovative application scenarios for intelligent identification and assessment based on space-air-ground information.
2. Technology Integration: Deeply integrate AI with space-air-ground information. The use of at least one type of remote-sensing data source, such as optical imagery, SAR, hyperspectral imagery, or thermal infrared data, is encouraged to enable automated extraction and intelligent analysis of multidimensional urban health-assessment indicators.
3. Solution Design: Develop a complete technical solution covering multi-source data acquisition and alignment, feature extraction and intelligent analysis, and interpretable outputs, thereby forming a closed-loop system for urban health assessment.
4. Value Demonstration: Validate the solution empirically in a representative urban area. Quantitatively analyze its potential benefits in improving diagnostic efficiency, reducing labor costs, and enhancing the accuracy of decision support, and assess its scalability and socioeconomic benefits.
5. Optional Directions: Possible themes include, but are not limited to, AI-based urban health assessment for building diagnosis, environmental monitoring, transportation evaluation, infrastructure safety, and carbon measurement.
Submission Requirements：
1. A technical solution document in PDF format (approximately 5,000 Chinese characters), including the problem definition, data sources, methodological framework, experimental design, and analysis of results.
2. An evaluation report and dataset description presenting quantitative evaluation results obtained on a public or self-constructed dataset.
3. A demonstration video of three to five minutes showing the system workflow, core functions, and comparative results.
04
Challenge Title
“AI+”-Empowered Intelligent Modeling and Scene Reconstruction for Urban Digital Twins
Challenge Description：Urban digital twins are a core platform for smart-city development. Their effectiveness depends on high-precision, updateable, and multiscale intelligent modeling of urban entities, together with dynamic scene reconstruction. This challenge explores the use of multi-source space-air-ground data, including remote-sensing imagery, street-view imagery, LiDAR, IoT data, GIS attributes, and engineering specifications, to develop forward-looking, practical, and scalable AI-based solutions.
Core Requirements：
1. Scenario Innovation: Address the needs for urban-entity representation and dynamic updating in urban digital-twin development. For typical scenarios such as urban planning, construction management, emergency simulation, renewal decision-making, and intelligent operation and maintenance, highlight the foundational role of remote-sensing data in large-scale, multi-temporal urban modeling, and propose innovative applications for intelligent modeling and scene reconstruction based on AI and space-air-ground data.
2. Technology Integration: Deeply integrate AI with space-air-ground information. Fuse at least one remote-sensing data source, such as satellite imagery, UAV imagery, or LiDAR, to achieve high-precision intelligent representation and dynamic reconstruction of urban entities and scenes.
3. Solution Design: Develop a complete technical pipeline covering multi-source space-air-ground data acquisition and fusion, entity-feature extraction and semantic understanding, geometric and attribute modeling, dynamic scene reconstruction, visual presentation, and integration with digital-twin applications. The result shall be an end-to-end intelligent modeling and scene-reconstruction solution for urban digital twins.
4. Value Demonstration: Validate the solution empirically in a representative urban area or a real application scenario. Quantitatively analyze its potential benefits in improving modeling accuracy and update efficiency, reducing labor costs, and strengthening decision-support effectiveness, and assess its scalability and socioeconomic value.
5. Optional Directions: Possible themes include, but are not limited to, AI-based digital twins for intelligent building modeling, dynamic reconstruction of urban elements, semantic scene editing and generation, digital-twin data updating, and multiscale scene fusion.
Submission Requirements：
1. A technical solution document in PDF format (approximately 5,000 Chinese characters), including the problem definition, data sources, methodological framework, experimental design, and analysis of results.
2. An evaluation report and dataset description presenting quantitative evaluation results obtained on a public or self-constructed dataset.
3. A demonstration video of three to five minutes showing the system workflow, core functions, and comparative results.