[Objective] This research addresses the persistent challenges of controllability and interpretability in generative algorithms for landscape architecture, aiming to bridge the gap between theoretical model development and practical design application. Current end-to-end generative frameworks often lack semantic transparency and controllable intermediate mechanisms, limiting their adaptability to complex, real-world design contexts. To overcome these limitations, this research proposes a multi-model, multi-path generative framework that integrates semantic and process-level interpretability with designer interaction. The framework is designed to enhance both the rationality and professional logic of generative outcomes, providing a controllable and transparent computational pathway for intelligent park and open space design.
[Methods] The proposed framework is built upon a “functional − semantic − spatial” integrated node classification system, in which each landscape node is abstracted into a multi-attribute characteristic vector. These node representations function as the semantic mediation layer throughout the generative process, supporting the implementation of the following three core tasks. Further, a nodal functional relationship refers to the structured spatial and functional connections among nodes within a landscape system, describing how individual functional units (e.g., plazas, paths, vegetated zones, and waterfronts) interact, overlap, or depend on each other within a coherent spatial organization, thereby capturing both the topological (positional and connectivity) and semantic (functional purpose and hierarchy) dimensions of landscape composition, and serving as the mediating mechanism that bridges abstract design intent and concrete spatial generation. 1) Inference of nodal functional relationship: Image-to-structure translation is adopted to predict spatial and functional linkages among landscape elements. 2) Generation of complete layout scheme: Nodal functional relationships are translated into coherent site layouts under varying spatial and functional constraints. 3) Terrain-aware adaptive generation: Elevation and slope data are integrated to enable topographically responsive design outcomes. To ensure flexibility across different design needs, the framework supports two application paths, namely the rapid generation path and the directed generation path, which are detailed as follows. 1) The rapid generation path, automatically infers nodal functional relationships via Model A and generates complete layouts through Models B-1 and B-2, requiring only minimal designer adjustments. This path is suitable for conceptual and early-stage design, where efficiency and iterative exploration are prioritized. 2) The directed generation path entails designers to manually define or modify nodal relationships based on site conditions, design intent, or functional strategies. This path allows targeted intervention and stronger alignment with specific planning objectives, supporting semi-automatic, user-guided generation. Depending on the inclusion of topographic data, two complementary sub-models are employed: Model B-1 handles general sites without terrain information using a CycleGAN-based end-to-end architecture, while Model B-2 integrates topographic characteristics into a multi-channel input (B for spatial structure, G for semantic function, and R for terrain information), enhancing adaptive learning for complex terrains. A multi-dimensional evaluation system is incorporated to assess and optimize generative outcomes. The evaluation framework encompasses indicators of structural coherence, spatial configuration, land use ratio, and road attributes, enabling both quantitative comparison with existing reference plans and independent evaluation under unconstrained conditions. In the latter case, a norm-constrained multi-objective optimization process is applied to ensure design practicality, guiding the generative system beyond visual similarity toward functional and implementable outcomes.
[Results] Empirical validation is conducted using two representative cases — Beixiaohe Park in Beijing and Fanchuan Park in Xi’an — demonstrating the framework’s robustness and adaptability across site typologies and environmental constraints. In the Beixiaohe Park case, where terrain variation is minimal, the system effectively produces spatially coherent and functionally reasonable layouts through the rapid generation path, achieving fast scheme generation and iterative refinement with minimal manual intervention. The integration of the evaluation framework enables automatic selection of optimized results based on multi-dimensional indicators, verifying the system’s operational flexibility and practicality in early-stage design contexts. In contrast, in the Fanchuan Park case characterized by complex terrain and larger spatial heterogeneity, the advantages of the directed generation path and the terrain-aware Model B-2 are highlighted. By incorporating multi-channel inputs that encoded spatial, functional, and topographic data, the system significantly improves road network continuity, spatial organization, and topographic adaptability, producing results closely aligned with expert-designed schemes. Quantitative evaluation confirms that the system boasts higher structural coherence and spatial rationality compared with baseline generative approaches, while maintaining a balance between diversity and functionality. The integrated multi-dimensional evaluation system proves applicable under both comparative (compared with real reference plans) and non-comparative conditions, offering an objective, transparent mechanism for solution screening and optimization. This capability effectively compensates for the prevailing “similarity-oriented but less practical” tendency in current generative design research, demonstrating the framework’s value as a decision-support tool for real-world landscape planning.
[Conclusion] This research establishes a multi-model, multi-path generative framework incorporating a semantic mediation mechanism providing a systematic and process-level interpretable technical route for intelligent spatial layout in complex landscape sites. By integrating node semantics, terrain constraints, and quantitative evaluation into a unified workflow, the framework advances both the theoretical and applied dimensions of generative design in landscape architecture. The research contributes three major insights: 1) Introducing nodal functional relationships as mediating mechanism effectively decomposes the opaque end-to-end generative process, enhancing structural clarity and interpretability; 2) the dual-path strategy allows seamless transition between rapid automated generation and designer-directed customization, achieving a practical balance between efficiency and control; 3) the multi-dimensional evaluation system establishes a standardized, data-driven basis for assessing design rationality and spatial performance, promoting the shift of generative design from similarity-based learning to utility-oriented application. Future work will focus on developing goal & function−driven dynamic tuning mechanisms and multi-source data integration to strengthen model generalization and applicability. Efforts will also be made to explore the coupling of behavioral, functional, and spatial dimensions, as well as real-time human − AI co-design interfaces, so as to further enhance collaboration, adaptability, and practical impact in intelligent landscape and urban design.

[Objective] As urban design faces increasing demands for contextual responsiveness, iterative optimization, and data-informed reasoning, integrating artificial intelligence into the design process has gained renewed relevance. Among emerging technologies, generative artificial intelligence (GAI) shows strong potential for automating content creation and simulating spatial configurations. This research provides a comprehensive review of recent developments in the application of GAI to urban design. The research identifies representative technical pathways, their respective intervention stages, and the functional mechanisms by which generative models are reshaping the design workflow. This research presents a structured, theory-informed synthesis of how different generative models contribute to tasks such as intention modeling, spatial reasoning, and performance-driven design. Building on design thinking and a descriptive lens informed by the technology acceptance model (TAM), the research examines how model type, data modality, and task characteristics affect GAI’s functional role, usability, and acceptance. Particular attention is given to mapping deployment forms, from isolated tools to coordinated multi-model workflows, and to characterizing cross-cutting challenges of controllability, transparency, and contextual adaptability in urban design settings.
[Methods] Following the PRISMA protocol, the research conducts a multi-stage literature review combining automated search and expert screening. A total of 125 peer-reviewed articles and high-impact preprints are selected from Web of Science, CNKI, arXiv, and selected industry sources, covering the period from 2014 to July 2025. Search terms such as “generative AI”, “AIGC”, “GAN”, “diffusion model”, “variational autoencoder”, “autoregressive model”, “large language model”, and urban-related keywords are used in various combinations. Based on the collected literature, four types of generative models are summarized as image-driven, language-driven, structure-driven, and feedback-optimized models, according to their application characteristics in urban design tasks. These types are aligned with four stages of the design process: preliminary analysis, scheme generation, evaluation and decision-making, and outcome expression. On this basis, a two-dimensional framework to examine how different GAI pathways intervene across tasks is formed. To refine the mapping, each design stage is further broken down into three representative sub-tasks. Preliminary analysis includes public demand analysis, urban data enhancement, case/task framing, and spatial element recognition. Scheme generation covers design intention modeling, spatial layout generation, and 3D form construction. The evaluation and decision-making stage includes multi-objective optimization, scheme evaluation, and scenario prediction. The final expression stage involves textual documentation, 2D representation, and visual rendering. A quantitative analysis is also conducted to show the distribution of model types over design stages, identify common combinations, and trace the evolution of research focus over time. TAM informs a descriptive synthesis of perceived usefulness (PU) and perceived ease of use (PEU) across model types to illuminate adoption patterns.
[Results] The findings reveal that GAI models are increasingly integrated into urban design workflows but exhibit uneven adoption across task types and modalities. Image-driven models dominate in both early-stage analysis and final visual representation due to their high interpretability, usability, and compatibility with existing design practices. Language-driven models are commonly used in public demand analysis, participatory planning, and scenario scripting, enabled by the rise of large language models (LLMs) such as ChatGPT and DeepSeek. Structure-driven models, though less prevalent, show promise in generating street networks, land-parcel layouts, and spatial typologies using graph-based logic. Feedback-optimized models, which rely on reinforcement learning, evolutionary algorithms, and performance simulation are the least adopted, but demonstrate strong potential in multi-objective optimization and iterative decision-making. Recent research indicates an increasing use of multi-model workflows, such as text-to-image pipelines integrated with urban simulation or feedback loops. While GAI applications increasingly support design iteration, their adoption is heavily influenced by the controllability, explainability, and contextual adaptability of models. PU and PEU vary significantly by model type, with image-driven models rated highest and structure-driven and feedback-optimized models facing usability challenges due to complexity and low transparency.
[Conclusion] Although GAI has demonstrated broad applicability across the urban design process, current implementations are largely procedural and auxiliary in nature. Most models recombine existing inputs rather than construct original logic, and few possess autonomous reasoning or normative awareness. This limits their role to content augmentation rather than conceptual guidance in design development. Moreover, issues such as opaque decision logic, lack of domain-specific knowledge embedding, and poor adaptability to local planning norms hinder practical adoption. Addressing these challenges requires multi-level efforts: 1) Construct structured, regionally grounded urban design datasets; 2) improve model interpretability, controllability, and responsiveness to professional input; and 3) develop modular, multi-model systems that support seamless interaction across design stages. Human – AI collaboration mechanisms — especially those based on iterative prompts and semantic feedback-must be enhanced to enable AI not just as a tool, but as an active design partner. This review offers a comprehensive reference for scholars and practitioners seeking to understand how GAI is reshaping the logic, structure, and agency of urban design.

[Objective] As one of the most scarce landscape resources in cities, the coastal shoreline endows urban coastal spaces with unique ecological and landscape conditions. Meanwhile, with the development of the social economy, the expansion of urban fringes, and the improvement of living standards, coastal spaces are playing an increasingly important role in urban development. However, due to the influence of location, form, and landscape, problems such as insufficient human presence, seasonal differences, and limited radiation range in the vitality of coastal spaces have gradually emerged during the urbanization process. How to improve the quality of coastal spaces and enhance their vitality has become a widely concerned issue in both academic and industrial circles.
[Methods] To address this issue, this research constructs a digital portrait of different crowds in coastal spaces based on the analysis of the evaluation dimensions of the vitality of coastal spaces. The digital portrait is created through four dimensions: basic attributes, socio-economic status, travel purposes, and lifestyle. By analyzing the spatial distribution of stay points and the spatio-temporal patterns of travel trajectories of each typical crowd at different times, the vitality of coastal spaces for each crowd is further analyzed, including the attractiveness of coastal spaces to various crowds, as well as the capacity and transportation accessibility of coastal spaces. Then, through field research, questionnaire interviews, and spatial simulation analysis, the root causes of relevant problems are identified. Finally, based on the behavioral trajectories and spatio-temporal vitality differences of different crowds, optimization strategies for the spatial layout of coastal spaces are proposed. In contrast to traditional approaches, the analysis of coastal space vitality grounded in the digital profiling of crowds enables the screening of key subjects from a vast and intricate crowd. It can also pinpoint the core issues in a targeted manner, thereby effectively enhancing the vitality and quality of coastal spaces. Ultimately, by taking into account the behavioral trajectories of each crowd and their vitality variances across different time intervals, optimization strategies for traffic guidance and spatial layout within coastal spaces are put forward. Additionally, integrating the spatial distribution of crowd stop points and crowd categories, the morphological structure and spatial nodes are optimized and upgraded.
[Results] Taking the coastal space of Jiulong Bay in Weihai as an example, this research selects 18 typical crowds with the largest proportion to construct a crowd digital portrait. The research finds that the crux of the vitality issue of coastal spaces lies in three aspects: First, the closed layout and spatial fragmentation prevent people from traveling to coastal spaces; second, the lagging infrastructure construction makes it difficult for people to stay in coastal spaces; third, the long and narrow transportation system makes it difficult to disperse people in coastal spaces. In response to these three problems, this research proposes design strategies such as attracting recreation, inhibiting pass-through, promoting the integration and sharing of diverse crowds, and optimizing the form of green space nodes.
[Conclusion] The crowd digital portrait proposed in this research as a digital means of analyzing crowd activities, has the characteristic of deeply depicting and classifying the age and gender composition, behavioral purposes, activity methods, and trajectory patterns of the crowds in coastal spaces based on their geographical location, transportation environment, and landscape characteristics. It can help understand the vitality characteristics and spatio-temporal patterns of coastal spaces, and then, in combination with the form of coastal spaces, identify the crux of problems such as insufficient human presence and uneven vitality, and propose corresponding strategies.

[Objective] Urban parks play a vital role in enhancing residents’ physical and mental well-being and offering leisure opportunities. Their vitality has become a crucial indicator of urban spatial quality and public welfare. Rapid urbanization has further intensified the imbalance in the allocation of public service resources. Existing research, which primarily relies on heat maps, mobile signaling data, or ground-based camera monitoring, can reveal macroscopic trends but fail to capture the dynamic spatiotemporal characteristics of crowd distribution at the micro scale. Meanwhile, aerial photography obtained through unmanned aerial vehicle (UAV) offers high spatial resolution and flexible data acquisition capabilities, while the advancement of object detection algorithms based on deep learning presents new technological opportunities for crowd recognition in complex urban environments. This research aims to develop and validate a micro-scale vitality measurement method for urban parks based on aerial time-series imagery and an improved object detection model. The method seeks to reveal the spatiotemporal patterns of crowd distribution, identify high-frequency vitality nodes and their driving mechanisms, and provide data support and strategic insights for optimizing the spatial layout, facility allocation, and refined management of parks. Taking Xi’an Xingfu Linear Park as an example, the research focuses on analyzing vitality intensity, fluctuation, and spatial balance at a fine spatiotemporal scale.
[Methods] Between March 27 and 30, 2025, continuous UAV-based aerial photography was conducted at a fixed altitude of 75 m during six standard time periods (08:00, 10:00, 12:00, 14:00, 16:00, 18:00), yielding over 2,300 high-resolution images. A manually annotated dataset of 2,000 sub-images with 12,340 pedestrian instances is constructed for model training. To address challenges of small-scale targets and complex occlusions in aerial imagery, an enhanced YOLO11m-CBAM model is developed by embedding a convolutional block attention module (CBAM) into YOLO11m. The improved model achieves notable performance gains: mAP50 increases from 77.1% to 81.3%, mAP50–95 from 45.6% to 51.7%, with precision and recall reaching 86.4% and 72.0% respectively, demonstrating enhanced robustness under medium and low occlusion conditions. Detection outputs are orthorectified to geographic coordinates to construct a structured spatiotemporal dataset. Spatial analysis employs kernel density estimation, coefficient of variation (CV), spatial Gini coefficient, and the “latitude-population” curve to characterize multidimensional vitality patterns.
[Results] The temporal analysis results indicate that the overall utilization of Xingfu Forest Belt exhibits a distinct “dual-peak” pattern. On rest days, the number of visitors reaches 2,112 at 10:00 and 3,641 at 16:00, reflecting typical peaks of family and leisure activities. The daily coefficient of variation (CV = 38.67%) is relatively low, suggesting stable visiting patterns with activity concentrated in leisure hours. In contrast, on working days, vitality peaks occur at 10:00 and 18:00, corresponding to post-commuting and after-work relaxation periods, respectively. The higher daily visiting (CV = 55.34%) indicates a more uneven temporal distribution of activities. Notably, 12:00 represents the lowest point of visiting (the minimum number of visitors is only 595, and the average number is 883), implying underutilization of space during midday and suggesting potential opportunities for future facility optimization or time-specific programming. The spatial equilibrium analysis further reveals that during peak hours (14:00 and 16:00), the spatial Gini coefficient reaches 0.44 – 0.48, indicating a strong concentration of vitality in specific functional zones and a pronounced spatial polarization effect. In contrast, the Gini coefficient drops to 0.24 during off-peak periods (08:00 and 12:00), reflecting a more dispersed and evenly distributed use of space. At 18:00, the Gini coefficient remains between 0.38 and 0.41, suggesting a moderate level of aggregation in the evening. Overall, the vitality of Xingfu Forest Belt demonstrates a dynamic pattern of “daytime polarization with evening recovery”. In terms of spatial distribution, vitality hotspots are primarily concentrated along the central and northern segments of the belt, forming localized peaks. The emergence of these core areas is driven by two main factors: 1) the attraction of fixed functional facilities such as children’s play areas, fitness zones, and square-dancing spaces; and 2) the temporal aggregation generated by periodic activities, including weekend family events and morning exercise. At the macro scale, the concentration of residential and educational land uses, high accessibility to bus stops, and the scarcity of comparable recreational facilities jointly reinforce the sustained vitality of the central children’s play area. Maintaining consistently high footfall and strong spatial spillover effects across multiple time periods, this area serves as a key vitality hub within the overall spatial structure of Xingfu Forest Belt.
[Conclusion] The research demonstrates that the proposed UAV-based and YOLO-based vitality measurement framework provides high spatiotemporal resolution at the micro-park scale, enabling accurate identification of vitality hotspots, temporal fluctuations, and spatial imbalances. This approach offers an operational, quantitative basis for optimizing facility layouts, designing flexible spaces, and implementing differentiated management strategies. Methodological limitations are also discussed: The approach performs reliably in spring, autumn, and winter with low to moderate vegetation coverage, but may encounter partial omissions under dense canopy or multi-layer pergola structures in summer. To enhance applicability, future improvements include multi-drone and multi-view data acquisition, infrared thermal imaging to mitigate occlusion, air-ground data fusion, inter-frame trajectory matching to distinguish stay/pass behaviors, and fine-grained activity recognition. Overall, the proposed method provides a replicable technical pathway and empirical reference for refined park governance and smart park development. The findings contribute to advancing quantitative urban vitality assessment and provide methodological insights for integrating AI and spatial analysis in urban landscape research.

[Objective] The intensification of climate change has led to a significant escalation in flood risk within shallow mountainous areas, posing a severe threat to human life, health, and ecological security. These transitional areas, often situated at the interface between mountainous terrain and urbanized plains, are uniquely vulnerable to the hydrological impacts of extreme precipitation. Existing research has established that green infrastructure (GI), through its influence on fundamental hydrological processes such as the rainfall – runoff and runoff – sediment relationships, can play a pivotal role in stormwater management. However, the current body of literature predominantly focuses on two main scales: the effectiveness of individual GI elements at the localized plot level and the impact of the broader green space matrix at the large basin scale. Consequently, a critical knowledge gap persists concerning the influence of the spatial configuration of GI patches — such as their shape, size, and degree of fragmentation — on hydrological responses at the finer, sub-basin scale, which is the most relevant scale for understanding flood generation. Clarifying the mechanisms through which GI spatial patterns affect mountainous stormwater runoff and subsequently optimizing these patterns are crucial steps toward enhancing the flood prevention and control capabilities of shallow mountainous areas. This research aims to bridge the knowledge gap by elucidating these mechanisms and developing an optimization framework to mitigate the adverse effects of extreme rainfall in the sensitive shallow mountainous areas.
[Methods] This research adopts a two-stage research framework, comprising the two stages of mechanism exploration and pattern optimization. In the stage of exploration of hydrological mechanisms, two sample basins are selected within the shallow mountainous area of Beijing and, based on historical meteorological data and land cover data, the SWAT (soil and water assessment tool) model is used to simulate runoff generation in mountainous sub-basins with high spatiotemporal resolution. Meanwhile, machine learning methods, specifically an XGBoost-based model, are applied to the sample data to construct a high-accuracy predictive model for stormwater runoff generation, with a focus on GI spatial pattern characteristics as predictor variables. To interpret the machine learning results, the SHAP (SHapley Additive exPlanations) framework is employed to quantitatively elucidate the impact mechanisms of various GI spatial pattern metrics on mountainous stormwater runoff. In the pattern optimization stage, key GI spatial metrics are identified as optimization variables based on their hydrological influence. Under a dual-objective framework emphasizing both cost-effectiveness and flood mitigation efficacy, the NSGA-Ⅱ (nondominated sorting genetic algorithm Ⅱ) is used to optimize GI configuration for a representative shallow mountainous area. The effectiveness of these optimizations in reducing flood risks is validated through extreme historical rainfall scenarios.
[Results] The resulting predictive model for mountainous runoff generation demonstrates excellent simulation and forecasting capabilities, especially in modeling the influence of GI spatial pattern changes on runoff processes in complex mountainous terrains. The interpretive analysis using SHAP on the trained model provides crucial insights into the underlying mechanisms. Among the numerous GI landscape metrics evaluated, two features emerge as the most critical drivers positively correlated with increased mountainous stormwater runoff: the patch density (PD) of closed-canopy deciduous broad-leaved forests and the percent of landscape (PLAND) occupied by grasslands. The analysis reveals that an increase in either of the aforesaid two metrics consistently contributes to higher predicted runoff volumes. In contrast, the spatial pattern characteristics of other vegetation types, such as closed-canopy evergreen coniferous forests and closed-canopy deciduous coniferous forests, are found with a comparatively weak and less significant influence on the hydrological response. During the multi-objective pattern optimization process, using the two most influential metrics (PD and PLAND) as adjustable variables for a typical area, the optimized spatial pattern is able to reduce flood risk by 13.5% under the scenario of once-in-a-century extreme rainfall.
[Conclusion] The XGBoost machine learning model displays outstanding applicability for flood risk assessment and hydrological scenario simulation in shallow mountainous areas. An in-depth analysis of the GI spatial metrics identified by SHAP interpretation suggests that the fragmentation resulting from increased PD of closed-canopy deciduous broad-leaved forests, together with the impact of grassland PLAND on the runoff coefficient, are the core driving factors of stormwater runoff generation in these mountainous contexts. Additionally, the shape and configuration of grassland patches may further promote stormwater runoff. Accordingly, in the process of optimizing GI spatial arrangements in shallow mountainous areas, enhancing the connectivity of closed-canopy deciduous broad-leaved forest while reducing the size of large grassland patches is found conducive to forming optimal GI layouts that reduce flood risk under extreme precipitation. Through the application of interpretable machine learning techniques, this research reveals the underlying mechanisms by which different GI spatial pattern metrics influence mountain runoff generation and, based on these findings, effectively reduces regional flood risk during extreme rainfall events. The methodological approach and practical guidance provided by this research offer robust technical support for flood-mitigating green space planning in similar shallow mountain terrains and contribute valuable experience for regional adaptation to intensified climate-driven stormwater challenges.

[Objective] This research addresses critical challenges in the documentation and research of classical Chinese gardens. As exemplary representatives of the World Cultural Heritage, Suzhou classical gardens are renowned for their intricate spatial compositions and profound cultural significance. However, current teaching and research predominantly rely on manual surveying and mapping data from the last century, such as the maps included in Liu Dunzhen’s publication, which no longer accurately reflect the current conditions. This research takes the Master-of-Nets Garden as an example, whose spatial layout has undergone multiple modifications, including the restoration of the Peony Courtyard in 2003, making it significantly different from what it is in existing maps. Traditional manual surveying methods are typically inefficient and subjective, particularly when documenting complex the morphological features such as rockery textures and architectural curves. Therefore, this research innovatively integrates modern digital surveying technologies, including 3D laser scanning and photogrammetry with intelligent image processing algorithms, such as the Canny edge detection and gradient analysis, to develop a comprehensive methodology for automated feature recognition and 2D drawing generation. Based on the case study of the Peony Courtyard, this research establishes a high-precision 3D point cloud model, aiming to provide reliable technical support and scientific basis for garden heritage conservation, academic research, and professional education, while addressing the critical limitation of historical maps in dynamically reflecting garden evolution.
[Methods] This research adopts a multi-source data fusion approach, systematically integrating three advanced surveying techniques. During the data acquisition stage, terrestrial photogrammetry is first employed using a GPS-equipped Nikon Z5 camera to capture 1,675 high-quality images under controlled conditions at fixed daily time slots, with the overlapping area between consecutive images exceeding 70%, comprehensively covering traditionally difficult-to-document concealed areas including interior spaces, eaves, and rockery caves. Secondly, oblique aerial photography is conducted using a DJI Mavic 2 Pro drone along five designed flight paths (one nadir and four oblique routes) capturing 188 georeferenced aerial images. Thirdly, the FARO Focus S350 3D laser scanner is deployed at 26 locations to capture high-precision data of complex morphological features such as building facades and rockeries. During the data processing stage, RealityCapture is used to integrate multi-source data, constructing a 3D point cloud model with millimeter-level precision. It is verified through 38 on-site measurements using steel tape that the model’s overall error rate at 0.71% ± 0.13% (mean ± SD), with particularly reliable accuracy in architectural and courtyard areas. During the intelligent mapping stage, this study employs the Canny edge detection algorithm, with its optimal high and low thresholds of 4 and 2 determined through repeated trials, to extract feature lines of objects. Subsequently, this research utilizes gradient threshold masks to categorize the feature lines into three hierarchical levels: outer contours, secondary contours, and texture lines, corresponding to thick, medium, and thin lines, respectively, ultimately generating professional-level 2D plans and sections. Lastly, special elements like vegetation are optimized through manual assistance to ensure the completeness and accuracy of drawings.
[Results] The experimental outcomes have significant advantages in multiple aspects. In terms of precision, the algorithm-generated 2D drawings maintain a stable error rate below 1%, substantially outperforming traditional manual surveying. Technically, the method successfully captures and represents subtle architectural curves and complex rockery textures that are challenging for conventional documentation. Systematic comparison with historical drawings reveals important layout modifications, such as the non-linear configuration of the Peony Courtyard’s eastern and western walls and their non-perpendicular relationship with the southern wall, with such findings corroborated by restoration photographs from the late 1950s. This research also accurately documents detailed changes including newly added rocks at the southeastern corner and morphological evolution of the steps of Hanbi Spring. Limitations include some blurred representations of interior furniture and certain windows or doors due to insufficient scanning coverage, and the need for manual parameter adjustment in complex rockery areas. Notably, the established 3D point cloud model offers comprehensive data advantages, supporting cross-sectional extraction and drawing generation from any viewpoint, overcoming the fixed-perspective limitation of traditional methods. This provides unprecedented technical possibilities for long-term monitoring and dynamic documentation of garden heritage. The entire methodology ensures professional accuracy while significantly improving efficiency, enabling multi-angle outputs from single data acquisition and greatly reducing repetitive field measurements.
[Conclusion] Through systematic technological development and empirical research, this research successfully validates the practical value of digital surveying and intelligent algorithms in the documentation and conservation of classical gardens. Technically, the research confirms the effectiveness of combining Canny edge detection with gradient threshold masking for feature extraction, establishing a complete intelligent workflow from the 3D point cloud model to 2D drawings. Regarding application value, the proposed methodology not only generates professional-level high-precision drawings, but also, through its unique traceability, enables dynamic documentation and analysis of garden evolution, providing a scientific basis for heritage monitoring and conservation decisions. Compared to traditional methods, the new technology demonstrates clear advantages in data completeness, workflow efficiency, and output accuracy, particularly excelling in documenting complex features such as rockery textures and architectural curves. Future research should focus on the following aspects: First, incorporating convolutional neural networks to enhance automated feature recognition and semantic segmentation; second, developing specialized modules for intelligent analysis of classical garden elements like rockery texture patterns and architectural components; third, establishing intelligent comparison systems between historical and current survey data for quantitative analysis of garden evolution. These innovations will advance the digital conservation of classical gardens from basic documentation to intelligent analysis, providing more robust technical support for sustainable cultural heritage conservation. The research outcomes are applicable not only to Suzhou classical gardens but can also be extended to other types of cultural heritage conservation practices, demonstrating broad application prospects and significant academic value.

[Objective] This research aims to explore the theory and practice of landscape design for campuses of primary and secondary schools in the context of contemporary high-density Chinese cities. The objective is to establish a comprehensive design framework based on the concept of “playscape” to transform school campuses from single-functional, enclosed educational facilities into vibrant, creative urban spaces that serve as nodes within the public realm. Confronted with challenges such as limited space, a rigid focus on academic performance, and physical and psychological segregation from surrounding communities, traditional campus design has proven inadequate. This research seeks to address these shortcomings by proposing a design philosophy that not only promotes the holistic development of children but also enhances the campus’s role as an open, inclusive, and vital community hub. The research intends to provide inspiring theoretical perspectives and practical pathways for the future innovation of campus landscape design and overall spatial design, grounded in asynthesis of the scientific principles of child development and the poetics of place-making.
[Methods] To achieve this objective, the research employs a multi-faceted methodological approach. First, it conducts a systematic historical review of the concepts of “play” and “playfulness” within the evolution of contemporary cultural, artistic, urban and architectural theories. This review traces the intellectual lineage from Schiller’s and Huizinga’s cultural theories to the critical practices of the Situationist International and Cedric Price, establishing “play” as a profound cultural phenomenon and a powerful tool for critiquing functionalist urbanism. Second, the research constructs an integrative design framework by synthesizing the “science” of child development with the “poetics” of playscape design. It systematically incorporates a five-dimensional design method, largely informed by the scientific findings in child psychology, behavioral science, and neuroscience, covering aspects such as risk assessment, embodied cognition, executive functions, social-emotional support, and benefits of the natural environment. This scientific dimension is interwoven with the poetic dimension, which draws inspiration from the pioneering works of artists and designers like Isamu Noguchi and Aldo van Eyck, focusing on place-making, aesthetic experience, and community engagement. Third, the research empirically illustrates and validates this framework through an in-depth case study of the landscape regeneration project at the Harbin Institute of Technology, Shenzhen Experimental School. This case study analyzes the specific strategies and methods applied to translate the playscape philosophy into a tangible design, particularly examining its adaptability within the constraints of a rapid, low-cost campus regeneration project.
[Results] The research yields several significant findings. The historical review confirms that “play” and “playfulness” have consistently served as a critical counter-narrative to rigid, functionalist approaches to urban design, acting as a vital source of urban vitality and community cohesion. The proposed “science-poetics” integrative framework proves to be a robust and effective tool for addressing the multifaceted challenges of campuses in high-density cities. The scientific dimension provides a clear, evidence-based rationale for design decisions, moving beyond intuition to create environments that precisely support children’s holistic development. The poetic dimension elevates the campus landscape from a mere functional backdrop to a meaningful “playable work of art” that fosters a sense of place and belonging. The case study of Harbin Institute of Technology, Shenzhen Experimental School demonstrates the framework’s practical applicability and adaptability. The findings show that even under significant constraints of time, budget, and space, the core values of the playscape can be effectively realized through “low-intervention, high-perception” strategies. The project successfully transforms a monotonous, single-functional sports ground into a dynamic, interactive playscape. Key results from the case study are summarized as follows. 1) The paradigm shift from a “functionalist” to a “developmentalist” landscape is achieved by creating a flowing topography and integrating natural elements. 2) “Spatial acupuncture”, a strategy of activating interstitial spaces, proves highly effective in maximizing the use of limited land resources. 3) The integration of “color therapy” with micro-topography shaping serves as a low-cost, high-impact method for enhancing spatial perception, guiding activities, and creating a positive emotional atmosphere. 4) The design successfully blurs the physical and psychological boundaries between the campus and the community, enhancing the school’s identity as a public node through strategies like the design of the “Sixth Facade” system. These results collectively demonstrate that the playscape philosophy, when adapted through context-specific, innovative strategies, offers a powerful pathway to overcoming the prevalent challenges in contemporary Chinese campus regeneration.
[Conclusion] This research concludes that “playscape” is an integrative design philosophy that masterfully combines scientific rationality with poetic and humanistic concerns, offering a systematic solution for the design of campuses of primary and secondary schools in high-density cities. The research establishes that a playscape-based approach can fundamentally reshape the campus, facilitating a paradigm shift from a passive, function-oriented environment to an active, educational space that scientifically empowers children’s growth. Furthermore, it serves as a critical catalyst for linking the campus to the city, transforming it from an isolated “island” into an open, creative, and inclusive hub that energizes the community. The “low-intervention, high-perception” strategies explored in this research offer a tangible and adaptable pathway for realizing innovative playscape designs within the common constraints of campus regeneration projects in China. Looking forward, the systematic application of the playscape philosophy to campus design is not merely an effective strategy for tackling current challenges, but also a vital step towards reimagining the school as a place full of creativity, inclusivity, and community vitality, which holds profound significance for nurturing future citizens and building harmonious, livable cities.

[Objective] In the context of profound demographic change and rapid urban restructuring, the spatial role of university campuses in Japan has undergone a fundamental transformation. Once conceived as inward-looking and self-sufficient “ivory tower” enclaves located on the urban periphery, campuses are increasingly being reconfigured as open and integrated nodes embedded within the metropolitan fabric. This paradigm shift is closely tied to Japan’s declining youth population, intensifying competition among universities, and evolving policy frameworks that regulate land use and higher education. Campus landscapes, in this process, are not merely ornamental green spaces but active agents of transformation that mediate the campus-city relations. The objective of this research is therefore to investigate how campus landscapes, as a spatial and social interface, respond to demographic pressures, policy incentives, and urban redevelopment agendas. By examining the synergistic evolution of universities and their host cities, the research aims to provide insights into the mechanisms that underpin this transformation and to extract lessons relevant to the forthcoming landscape transitions in Chinese higher education institutions.
[Methods] The research adopts a multi-scalar approach that combines historical trajectory analysis, case-based comparative study, and theoretical synthesis. First, the historical evolution of Japanese university campuses from 1945 to the present is traced and periodized into three major phases: the expansion phase (1945–1980s), when demographic booms and policy restrictions encouraged suburban relocation and the creation of enclosed, inward-looking campuses; the peak phase (1980s–2000s), marked by intensifying competition, partial return to urban centers, and the emergence of vertical and compact campus typologies; and the contraction phase (2000s to present), characterized by severe demographic decline, urban concentration, and increasing demands for publicness and integration. Second, representative case studies are selected from metropolitan Tokyo, regional cities, and newly developed urban districts. These are analyzed through spatial observation, planning documents, and secondary literature to identify common strategies and contextual variations. Third, the research synthesizes empirical findings into a typological framework of three strategic modes — “catalyst”, “regenerator”, and “stabilizer” — and further generalizes these into a theoretical three-pillar model composed of demographic dynamics, policy instruments, and spatial strategies. This model is used to explain the synergistic evolution mechanism of campus landscapes and urban environments.
[Results] The analysis shows that campus landscape transformation in Japan is not an isolated architectural endeavor but a systemic process shaped by demographic, institutional, and spatial forces. In newly developed urban areas and large-scale redevelopment zones, universities frequently operate as catalysts, strategically positioned to anchor emerging districts. Here, landscape strategies emphasize publicness, multi-functionality, and accessibility. For instance, the Toyosu Campus of Shibaura Institute of Technology integrates open terraces, green staircases, and community-oriented plazas that attract both students and local residents, thereby stimulating district-level vitality. In historic city centers and post-industrial neighborhoods, universities act as regenerators, using landscape interventions to repair urban fabric and reinvigorate cultural identity. Examples include the Kitasenju Campus of Tokyo Denki University, which deploys pedestrian linkages and unified pavement to soften campus — city boundaries, and Kyoto City University of Arts, which integrates riverside ecological restoration with cultural events to generate a “memory landscape”. In smaller regional cities, universities often serve as stabilizers, embedding themselves in local social and demographic structures through service-oriented landscapes and shared facilities. Fukuchiyama Public University, for example, co-locates community dining halls and elderly care facilities within its campus landscape, while university consortia in Kyoto pool resources to create a multi-institutional network of open sports fields, libraries, and cultural spaces accessible to local communities. The proposed three-pillar model explains the underlying mechanism of these transformations. Demographic decline provides the fundamental pressure, reducing the student-age population from over two million in the early 1990s to just above one million in the 2020s, with further decline projected. Policy instruments translate these demographic pressures into spatial outcomes, with such instruments ranging from restrictive measures such as the 1959 Factory Location Law to liberalizing interventions like the 1991 revision of university establishment standards, and most recently, the 2017 enrollment cap in central Tokyo. Spatial strategies, materialized through landscape design, serve as the ultimate vehicles through which demographic and policy drivers are enacted: Open courtyards, pedestrian corridors, cultural event spaces, and service-based green infrastructures become concrete manifestations of institutional adaptation. The interplay of these three pillars — demographics, policies, and spatial strategies — constitutes the synergistic evolution dynamic of campus landscapes and cities.
[Conclusion] Japanese experience shows campus landscapes have moved beyond their traditional role as green buffers to become strategic nodes of governance, cultural renewal, and social inclusion. By adopting roles of catalyst, regenerator, and stabilizer, campuses now shape urban growth, support community services, and sustain regional resilience. The proposed three-pillar model provides a structural lens for interpreting such changes. For China, where higher education faces slowed growth and demographic transition, these findings are highly relevant. Suburban university towns face the risk of under-use, while urban campuses must balance scarcity with public engagement. Japanese precedents suggest strategies of vertical compaction, boundary softening, and service-oriented integration can enhance publicness and urban alignment. Policymakers, meanwhile, should design flexible regulations balancing equity and autonomy. Future research should incorporate quantitative tools such as GIS metrics, user surveys, and cross-national comparison to further validate the three-pillar model and refine its applicability. Ultimately, campus landscapes must be understood not as passive backdrops but as active instruments in reshaping campus – city relations in an era of demographic and urban transformation.

[Objective] The quality of pedestrian environments is a crucial component of campus planning for comprehensive universities. As contemporary higher education increasingly emphasizes interdisciplinary communication, well-designed pedestrian environments can help foster interaction, strengthen campus identity, and promote active mobility. However, a great deal of research has identified the pedestrian-unfriendly conditions of university campuses in China, particularly those constructed in recent decades. Most empirical research has focus on sidewalks in campuses, while overlooking the characteristics and qualities of pedestrian spaces within squares, green spaces, and void spaces, revealing the limitations of quantitative evaluation. The lack of systematic characterization in existing studies also limits their applicability as practical guidances for campus pedestrian planning, in which urban design plays an integrative role. Therefore, a systematic spatial and design analysis of exemplary cases is needed. Singapore stands out for its well-developed pedestrian networks that effectively respond to tropical climatic conditions, support placemaking, and integrate with campus and urban systems. These qualities are particularly evident in its two comprehensive universities: the National University of Singapore (NUS) and Nanyang Technological University (NTU). This research examines the spatial and design characteristics of pedestrian spaces in NUS and NTU, aiming to extract strategies applicable to campus regeneration. The research first establishes a framework through a literature review that synthesizes key factors related to campus pedestrian environments and distinctive characteristics of Singapore’s pedestrian planning. Building on this framework, the research combines quantitative spatial analysis with qualitative mapping, on-site observation, and design analysis to identify the configurational and design characteristics of pedestrian spaces in the campuses of the both universities mentioned above. The findings further inform a discussion on design and management strategies for improving pedestrian environments in other universities.
[Methods] This research employs a mixed-methods approach for analyzing the design characteristics of pedestrian spaces. First, the research establishes an analysis framework based on a literature review. The review summarizes key spatial factors related to pedestrian on the campus from the perspectives of international research and distinctive characteristics of Singapore’s pedestrian system. The framework comprises four dimensions: overall spatial layout, public space design and placemaking, landscape integration, and infrastructural application. Second, on-site observation, spatial analysis and design analysis are conducted to unpack the characteristics of the campuses of the both universities. Specifically, observation comprises photography, measurement, hand sketching and mapping. Design characteristics are summarized in an inductive approach, according to the four-dimensional framework. In particular, to analyze the overall spatial layout, comprehensive mapping and configurational analysis are conducted. Mapping overlays walking spaces with campus functional layout and transportation systems. Based on the space syntax theory, a configurational analysis is conducted to measure the spatial relationships between one pedestrian space to another. This research employs two classical parameters, integration and choice, to measure to-potential and through-potential of each pedestrian space in relating to other spaces at local (400 m radius) and district (2,000 m radius) scales.
[Results] The analysis reveals that both NUS and NTU develop comprehensive pedestrian systems characterized by the above four aspects, overall layout, public space, landscape integration, and infrastructural application. First, continuous networks of covered walkways connect academic, residential, and service functions, as well as public transport nodes. The compact built form of the campuses of the both universities shortens pedestrian distance, promotes pedestrian activities, and makes pedestrian systems more efficient. Public transport routes are accessible to the campuses and share stops with campus shuttle, well-connected with covered walkways. Major covered walkways are laid out effectively to support local pedestrian activities, shown by high 400 m Choice values. While campus roads are supportive to the public transport for accessing to the campuses, shown by 2,000 m Choice. Besides, special attention should be paid to relations between multiple types of pedestrian spaces and campus roads and the configurational legibility of pedestrian spaces. Second, pedestrian spaces at NUS and NTU are characterized by their public space design and placemaking, which extend pedestrian nodes into multifunctional places that support studying, social interaction, and leisure. These node spaces are equipped with diverse forms of seats, lights, plants and equipment for better thermal comfort, encouraging encounter and stay. In particular, maintaining pedestrian spaces, both paths and nodes, at a human scale is crucial not only for users’ comfort while staying, but also for minimumizing impact on nature. Third, landscape integration balances aesthetic design with ecological and environmental performance. Shaded corridors, rain gardens, and terrain-responsive pathways enhance thermal comfort, support stormwater management, and strengthen ecological sustainability, while preserving the natural terrain and reinforcing campus identity. Fourth, infrastructure application comprises pedestrian-friendly elements and climate-responsive design. Natural ventilation, canopies, and semi-open transitions enhance microclimatic comfort, while durable materials, modular drainage systems, and traffic-calming measures improve accessibility, safety, and long-term maintainability. Despite these strengths, challenges persist, such as narrow pedestrian routes and limited connectivity with adjacent neighborhoods. Nevertheless, the two Singapore examples illustrate how progressive infill and adaptive design renewal can transform pedestrian systems into cohesive, efficient, and socially vibrant pedestrian environments.
[Conclusion] NUS and NTU demonstrate effective coverage of pedestrian spaces for core university functions with limited investment. Moreover, design strategies integrating pedestrian spaces with campus functions, public spaces, landscape resources, and transport systems can create walkable, vibrant, multi-functional and thermally comfortable pedestrian environments. Beyond spatial layout and design, the study offers more insights for universities subject to regeneration, particularly in China. First, despite different campus form, pedestrian space can be incrementally integrate with teaching and research spaces, amenities, and public transport. Second, creating vibrant pedestrian spaces and fostering placemaking of public space require accessible campus for pedestrian and open building public and transport spaces. A new Town & Gown relationship comprising accessible campus, shared facilities and open buildings means new management paradigms. Third, a compact building layout combined with human-scale public spaces forms the spatial foundation for active pedestrian space. This is achievable through the innovation of climate-response design elements, even under the current sunlight code of China. Fourth, digital analytical tools, such as behavioral tracking and environmental monitoring, can support participatory planning and performance evaluation. Successful campus pedestrian systems emerge from the coordination of spatial design, institutional management, and functional programming.