Garden landscape dynamic planning method and system based on augmented reality

Abstract

The invention relates to the technical field of man-machine interaction and augmented reality, and discloses a garden landscape dynamic planning method and system based on augmented reality. The method comprises the steps of obtaining a garden scene calling instruction input by a user based on augmented reality equipment, extracting corresponding data from a BIM model database and a plant growth database, and performing fusion to generate garden scene interaction data; based on a user simulation request, calling the generative adversarial network to generate a plant community dynamic evolution model; performing three-dimensional projection and rendering based on the display instruction to generate a garden simulation scene; dynamically adjusting the plant community evolution model based on user interaction input, generating an adjusted model and rendering in real time; and based on the optimization request, calling an AI analysis module, evaluating garden scene ecological indexes according to the ecological balance function, generating optimization suggestions and feeding back the optimization suggestions. According to the method, the interactive planning capability of plant community dynamic simulation, spatial layout and ecological optimization in garden landscape design is improved, and the user participation degree and the design scientificity are enhanced.

Description

Technical Field

[0001] This invention relates to the field of human-computer interaction and augmented reality application technology, and in particular to a dynamic planning method and system for garden landscape based on augmented reality. Background Technology

[0002] As landscape design demands increasing visualization, intelligence, and ecological harmony, traditional methods based on two-dimensional drawings and static renderings are insufficient to comprehensively consider the dynamic growth, spatial distribution, and ecological suitability of plant communities in complex scenarios. Existing landscape design systems lack the ability to support real-time interactive design via augmented reality devices, preventing users from dynamically adjusting key parameters such as plant species, density, and growth cycles in a WYSIWYG manner, and hindering real-time preview of adjustments. Furthermore, current technologies lack mechanisms to dynamically generate plant community evolution models by combining BIM models of the landscape space with plant growth databases, failing to dynamically display the evolution of plant communities across different time periods, seasons, or growth stages. This limits the accurate prediction and presentation of landscape effects within the actual usage period. Particularly when users change plant species or adjust spatial layouts, the lack of automatic system analysis and feedback regarding ecological balance, spatial coordination, and light suitability leads to ecological inconsistencies and spatial conflicts in landscape design. Therefore, there is an urgent need for a system and method based on augmented reality devices that can interactively adjust garden landscapes through various means such as user input, voice, gestures, and menu selection. This system should automatically generate dynamic plant community models by combining garden spatial data and plant growth data, and render and display the effects in real time through human-computer interaction. Simultaneously, it should conduct ecological rationality analysis on the user-adjusted garden design scheme and output optimization suggestions to improve the dynamic planning and real-time interactive capabilities of plant growth evolution, spatial layout, and ecological balance in the garden landscape design process. Summary of the Invention

[0003] This invention provides a dynamic planning method and system for garden landscape based on augmented reality, which addresses the problem of how to dynamically plan and interactively adjust garden landscape design parameters and visualize the effects of plant landscapes across multiple time periods by combining user interaction input, BIM model data, and plant growth database data obtained from augmented reality devices with dynamically generated plant community evolution models, and through real-time human-computer interaction and AI analysis. This improves the ability to dynamically plan and interactively adjust plant growth evolution, spatial layout, and ecological balance in the garden landscape design process.

[0004] To address the aforementioned technical problems, this invention provides a dynamic planning method for garden landscapes based on augmented reality, comprising: A dynamic planning method for garden landscape based on augmented reality, characterized in that the method includes the following steps: Based on augmented reality devices, the system obtains user input commands to invoke garden scenes, and in response to these commands, extracts data from the BIM model and plant database, and integrates them to generate interactive garden scene data. Based on the user-input simulation request command, a generative adversarial network is invoked to dynamically simulate the interactive data of the garden scene, generating a dynamic evolution model of the plant community. The generative adversarial network generates the dynamic evolution model of the plant community through the following function: ; Where G(z|D,Q) is the generator, z is the noise vector, D is the standardized garden scene interaction data, Q is the simulation request sequence; GAN is the adversarial function. Based on the user's input display instructions, the dynamic evolution model of the plant community is projected and rendered in three dimensions on an augmented reality device to generate a garden simulation scene. Based on the user's interactive input in the garden simulation scene, the dynamic evolution model of the plant community is dynamically adjusted to generate the adjusted dynamic evolution model, and the adjusted garden simulation scene is rendered and displayed in real time. Based on the user-inputted optimization request, the AI ​​analysis module is invoked to automatically analyze the ecologically relevant data in the acquired garden simulation scene. Regarding the ecological balance of the plant community, the following ecological balance function E is used: a (x) Conduct ecological indicator assessment: ; Where: E a (x) represents the current comprehensive score of the ecological balance of the garden scene; To adjust the plant community density index (distribution of plant quantity per unit area) in the post-garden simulation scene. The overall suitability coefficient for the duration of light exposure received by the plant in the current environment; α is an indicator of spatial harmony between plants and other landscape elements within a scene. a β a γ a These are the weighting coefficients for community density, light suitability, and spatial coordination, which are dynamically adjusted according to the ecological optimization model to satisfy α. a +β a +γ a =1.

[0005] Furthermore, in the augmented reality-based dynamic planning method for garden landscapes, before generating the interactive data of the garden scene, the standardized fusion process includes: The spatial structure data and plant community data are fused according to preset weighting coefficients to generate fused data. The weighting coefficients include the weight λ1 of the spatial structure data and the weight λ2 of the plant community data.

[0006] Furthermore, in the augmented reality-based dynamic planning method for garden landscapes, the simulation request sequence includes time period parameters and seasonal type parameters, which are set by user input. The generative adversarial network dynamically generates dynamic evolution models of plant communities under different time periods and seasons based on the simulation request sequence.

[0007] Furthermore, in the augmented reality-based dynamic planning method for garden landscapes, the three-dimensional projection and rendering steps include: Based on the on-site spatial environment parameters obtained by the augmented reality device, the dynamic evolution model of the plant community is subjected to adaptive rendering processing of spatial location, scale and lighting. The rendered simulation scene is then projected into the user's spatial environment and dynamically integrated with the real space.

[0008] Furthermore, the user's interactive input in the garden simulation scene includes plant type selection, plant distribution density adjustment, and growth cycle setting. The interactive input is parsed into a parameter adjustment sequence and input into the plant community dynamic evolution model for dynamic adjustment.

[0009] Furthermore, the adjusted dynamic evolution model generated by the dynamic adjustment includes: Based on the original dynamic evolution model and the incremental adjustment value generated by the parameter adjustment sequence, the adjusted dynamic evolution model is generated by the following formula: ; Where M is the original dynamic evolution model, ΔM(A) is the dynamic adjustment increment corresponding to the parameter adjustment sequence A, and M' is the adjusted dynamic model.

[0010] Furthermore, the ecological analysis includes a comprehensive evaluation of the plant community density, light suitability, and spatial coordination in the adjusted garden simulation scene, and an ecological balance score is calculated based on the ecological balance function to generate optimization suggestions that include plant distribution adjustment, density optimization, and spatial reconstruction.

[0011] Furthermore, the ecological optimization suggestions are displayed through the graphical interface, voice prompts, or interactive menus of the augmented reality device, allowing users to further adjust the garden simulation scene based on the optimization suggestions.

[0012] Furthermore, the dynamic rendering and ecological feedback form a closed-loop interaction mechanism. After the ecological optimization suggestion is fed back, the user adjusts the parameters based on the suggestion and automatically enters the next round of plant community dynamic evolution model generation and rendering update process.

[0013] Furthermore, the dynamic planning method for garden landscapes includes real-time tracking of the user's perspective and spatial position using augmented reality devices, and adaptive adjustment of the perspective of the dynamically rendered garden simulation scene to ensure the continuity and spatial consistency of the landscape display during user interaction.

[0014] The following are its main beneficial effects: (1) This invention obtains the garden scene call command input by the user through augmented reality device, integrates the data of BIM model and plant growth database, dynamically generates standardized garden scene interactive data, realizes the unified processing of spatial structure data and plant community growth data in the process of garden landscape design, solves the problem of data dispersion and inability to generate interactive data in traditional garden design, and improves the data processing efficiency and real-time interactive capability of garden landscape design.

[0015] (2) This invention generates a plant community growth and evolution model dynamically through a generative adversarial network and combines it with augmented reality devices for three-dimensional projection and rendering display. This allows users to view the dynamic growth effects of plants under different time periods and seasonal conditions, solving the problem that traditional garden design cannot intuitively display plant growth and evolution, improving the dynamic preview and simulation effect of garden landscape schemes, and enhancing the scientific nature of users' design decisions.

[0016] (3) The present invention dynamically adjusts the plant community configuration parameters through user interaction commands in augmented reality devices, and combines the AI ​​analysis module to conduct ecological rationality analysis on the adjusted garden scene, automatically outputs optimization suggestions such as plant distribution and spatial layout, forming a closed-loop process of design, interactive adjustment and ecological assessment, solving the problem that traditional garden design cannot be dynamically adjusted and lacks ecological optimization feedback, and improving the intelligence and ecological adaptability of dynamic planning of garden landscape. Attached Figure Description

[0017] Figure 1 A flowchart illustrating the augmented reality-based dynamic planning method for garden landscapes provided in this application embodiment; Figure 2 A structural block diagram of an augmented reality-based dynamic planning system for garden landscapes provided in this application embodiment. Detailed Implementation

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.

[0019] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0020] Example 1: Refer to Figure 1 This is a flowchart illustrating a dynamic planning method for garden landscapes based on augmented reality provided in an embodiment of the present invention. The process may include at least steps S100-S500: S100. Based on the augmented reality device, obtain the garden scene call command input by the user, and extract data from the BIM model and plant database in response to the command, and merge them to generate garden scene interactive data.

[0021] S200: Based on the simulation request command input by the user, call the generative adversarial network to perform dynamic simulation of the interactive data of the garden scene, generate a dynamic evolution model of the plant community, and return it for the user to operate.

[0022] S300: Based on the user's input display instructions, the dynamic evolution model is projected and rendered in three dimensions on the augmented reality device to generate a garden simulation scene for user interaction.

[0023] S400: Based on the user's interactive input in the garden simulation scene, dynamically adjust the dynamic evolution model of the plant community, and update and render the adjusted simulation scene in real time.

[0024] S500: Based on the user's input optimization request, the AI ​​analysis module is invoked to perform ecological analysis on the adjusted garden simulation scene, generate optimization suggestions, and feed them back to the augmented reality device for further user interaction.

[0025] Step S100 includes at least steps S110-S130: S110. Obtain the garden scene invocation command input by the user through the augmented reality device, parse the site number and plant type parameters in the garden scene invocation command, and obtain the scene data request parameter sequence.

[0026] Specifically, the augmented reality devices include, but are not limited to, augmented reality headsets (such as HoloLens), augmented reality glasses, or mobile devices with spatial mapping capabilities.

[0027] The garden scene invocation command input by the user through the augmented reality device includes a spatial layout invocation request for the garden scene, and the invocation request includes at least the target site number parameter ID. s With the target plant type parameter V s The ID s The V is used to indicate the specific location of a garden space. s Used to indicate the selection of target plant varieties.

[0028] Furthermore, the augmented reality device parses the site number parameter ID input by the user. s With plant type parameter V s This forms a sequence of scene data request parameters: ; Where p1, p2, ..., p n This indicates a comprehensive data request related to the garden scene, including spatial dimensions, topographic and geomorphological numbers, soil type numbers, target plant numbers, planting density indicators, sunshine duration parameters, etc., which will serve as the basis for subsequent extraction from the database.

[0029] The result of generating the parameter sequence P will serve as the basis for data extraction in step S120 and will be integrated into subsequent dynamic simulation calls to ensure the continuity of the data chain.

[0030] S120. Extract spatial structure data B={b1,b2,...,b...} corresponding to the scene data request parameter sequence P from the BIM model database. k}, and extract the corresponding plant community growth data T={t1,t2,...,t from the plant growth database. m}, and perform data fusion according to the following fusion function: Furthermore, in response to the scene data request parameter sequence P, the BIM model database is invoked to obtain the spatial structure dataset B={b1,b2,...,b} corresponding to the target garden space. k},in: b1 represents the three-dimensional terrain data of the target area; b2 represents the spatial coordinate reference frame; b3 represents the location data of existing hard paving and structures within the site; b k This indicates other spatial constraints.

[0031] Simultaneously, the plant growth database is invoked, based on the target plant type parameter V. s Extract community growth parameter data related to the growth of garden plants, T={t1,t2,...,t m},in: t1 represents the target plant life cycle parameters (such as the start of the growth cycle and the end of the growth cycle). t2 represents indicators such as crown width, plant height, and root distribution of the target plant; t3 represents the data from the community competition, light, and humidity response model; t m This indicates other growth behavior indicators (such as the interaction coefficient within the community).

[0032] After extraction, the spatial structure data B and the plant community growth data T are weighted and fused based on the following fusion function: ; in, , These are the spatial data weights and plant growth data weights, determined according to the actual scenario requirements. D represents the standardized garden scene interaction data.

[0033] Note: The standardized garden scene interactive data D, by integrating BIM spatial information and plant growth behavior parameters, provides a complete data foundation for the subsequent generation of dynamic simulation models, ensuring that the simulation results have site adaptability and scientific validity of plant growth.

[0034] S130. The standardized garden scene interaction data D is sent to the augmented reality device for subsequent use by the garden scene dynamic simulation module to form integrated initial garden scene interaction data. Specifically, the standardized garden scene interaction data D is synchronized to the local cache of the augmented reality device, stored in a structured data format, and used by the subsequent dynamic simulation model as the basic data source for generating the dynamic evolution model.

[0035] Furthermore, the standardized garden scene interaction data D includes the following structured fields: ; Among them, the field (b1,b2,...,b k ) and (t1,t2,...,t m ) correspond to the spatial structure data and plant community data extracted from S120, respectively, and have been based on λ1s , λ 2s It integrates and standardizes data, providing a unified data interface format.

[0036] Furthermore, the standardized garden scene interaction data D, as input data for dynamic simulation calls, directly serves as the simulation data source for the plant dynamic simulation model generation module (generative adversarial network) in S200, ensuring that the simulation input used in S200 originates from this standardized processing result, thereby guaranteeing the real-time performance and accuracy of the subsequent dynamic simulation model.

[0037] Step S200 includes at least steps S210-S230: S210. Based on the augmented reality device receiving the simulation request command input by the user, the time period parameter and season type parameter in the simulation request command are parsed to obtain the simulation request sequence Q={q1,q2,...,q x}, and associated with the garden scene interaction data D Specifically, after generating and caching the standardized garden scene interactive data D (S130), the augmented reality device obtains a simulation request instruction input by the user through the augmented reality device's interactive interface. The simulation request instruction includes at least the simulation time period parameter T. p and seasonal type parameter S p It is used to indicate the target time series and landscape state that need to be simulated.

[0038] Furthermore, by analyzing the time period parameter T in the simulation request command... p and seasonal type parameter S p Generate simulation request parameters, denoted as the simulation request sequence: ; Where q1,q2,...,q x These represent parameter nodes corresponding to different simulation time points and seasonal scenarios, including but not limited to plant life cycle states and environmental response parameters (such as temperature and light).

[0039] Furthermore, the simulation request sequence Q is associated with the standardized garden scene interaction data D generated in the aforementioned S130 step to form a joint data pair (D,Q), which is then used for the next step of generating adversarial network dynamic simulation processing.

[0040] S220. Based on Generative Adversarial Network (GAN), dynamic growth and evolution simulation is performed using garden scene interaction data D and simulation request sequence Q.

[0041] Specifically, for the joint data pair (D,Q), the built-in generative adversarial network (GAN) is invoked to simulate the dynamic growth and evolution of the plant community. The generative adversarial network includes a generator and a discriminator, which are used to generate plant community simulation data under multiple time series and dynamic environmental change conditions.

[0042] The generator part uses the joint data pair (D,Q) and the noise vector z s As input, dynamic simulation is performed based on the following generator function: ; Where: G( |D,Q) represents the generator G based on the noise vector Generate a dynamic evolution model of plant community using the combined data pairs (D,Q); The introduced random noise vector is used to enhance the diversity and naturalness of the generative model; D represents the standardized garden scene interaction data output from step S130; Q is the simulation request sequence generated by S210.

[0043] The generating function G( The plant community dynamic evolution model output by |D,Q) is denoted as: ; in, This represents a dynamic evolution model of plant communities generated under specific time periods and seasonal conditions, including multi-dimensional simulation data such as plant growth morphology, spatial distribution, size, and color at various simulation time points.

[0044] Furthermore, the generator G is dynamically calibrated based on the discriminator D to ensure the rationality of the generated model and the dynamic balance process that conforms to ecological logic. The discriminator performs the following discrimination: ; in, This represents the discrimination result output by the discriminator, used to dynamically adjust the output accuracy of the generator.

[0045] Finally, the dynamic evolution model M of the plant community after dynamic adversarial training equilibrium was obtained. s .

[0046] S230. Return the plant community dynamic evolution model M to the augmented reality device for subsequent dynamic rendering and display module to call, and prepare for interactive output of simulation results.

[0047] Furthermore, the plant community dynamic evolution model M generated by the generative adversarial network is used... sThe M is synchronously returned to the graphics rendering interface of the augmented reality device, serving as the basis for subsequent dynamic rendering and interactive display modules. s Include: ; Where m1, m2, ..., m x These correspond to the plant dynamic data fragments generated at each simulation time point in the simulation request sequence Q.

[0048] Specifically, the plant community dynamic evolution model Each data segment mᵢ includes the following sub-data: The three-dimensional spatial coordinates Lᵢ of a single plant organism; The growth dimensions of individual plants (height Hᵢ, crown width Wᵢ); Dynamic indicators of community light intensity, density, and color, such as Cᵢ; Interactive association identifiers are used for subsequent user ID adjustments.

[0049] Furthermore, the plant community dynamic evolution model M returned to the augmented reality device s It will be directly called in the next module S300 for three-dimensional visualization, dynamic rendering, and interactive display, providing a scenario basis for users to make interactive adjustments based on simulation results.

[0050] Step S300 includes at least steps S310-S330: S310. Based on the augmented reality device receiving the user's input display request instruction, the plant community dynamic evolution model M is invoked, and the scene data S(t) of the current display time node is extracted from it, wherein: Specifically, the augmented reality device acquires display request commands issued by the user through graphical interface, voice input, gesture input, etc., and the display request commands include a target time node parameter t. s This is used to indicate the specific time point at which the garden landscape needs to be presented.

[0051] Furthermore, the augmented reality device parses and displays the time node parameter t in the request command. s The plant community dynamic evolution model M generated and returned from the aforementioned S230 step s In the process, extract the time node t. s The corresponding scene data yields the time node scene data S(t) defined by the following formula. s ): ; in: This indicates at time point t s The following is a simulated garden landscape scene; This represents the plant community dynamic evolution model. Corresponding time point The model snapshot data; Q is the simulation request sequence generated in the aforementioned S210 step.

[0052] The time node scene data S( It includes various interactive parameters and visualization data, including: Spatial coordinates of each individual plant in a plant community ; Plant individual size information (height) Crown width ); Plant color and material information ; Associated interaction identifiers This is for subsequent user interaction adjustments and calls.

[0053] S320, regarding the time node scene data S( ) Perform 3D graphics rendering processing and execute spatial matching through a spatial environment adaptive algorithm. Furthermore, the time node scene data S( The data is input into the spatial rendering module inside the augmented reality device to perform three-dimensional graphics rendering processing, thereby generating the corresponding virtual garden scene.

[0054] Specifically, the augmented reality device captures environmental parameters based on spatial perception units (such as depth cameras, SLAM systems, etc.). It includes information such as the geometric boundaries of the site space, the location of obstacles, and the distribution of light.

[0055] Based on the following spatial rendering function (·), to transfer the scene data and on-site environmental parameters Perform adaptive spatial matching to obtain the rendering result. : ; in: For the final 3D rendered image data used in augmented reality devices; (·) is the spatial rendering matching function, which is calculated in real time based on scene data and the on-site environment; Scene data at the target time point; To enhance the on-site environmental parameters currently captured by the augmented reality device.

[0056] The rendering result Further includes: Three-dimensional data of plant models generated according to real-world spatial proportions; Coordinate position data corrected after spatial adaptation; Rendering results of adaptive ambient lighting and material effects.

[0057] The The rendering results will be used as data for dynamic display in the next step.

[0058] S330, render the result The system dynamically displays a simulated garden scene that allows for visual interaction with the user, providing a basis for subsequent parameter adjustment modules. Furthermore, the rendering result The data is transmitted to the display and interaction module of the augmented reality device and dynamically displayed on the holographic display screen of the device, forming a garden simulation scene that users can adjust parameters for.

[0059] Specifically, the simulated garden scene display includes: Three-dimensional dynamic plant community images; Visible changes in growth status (such as color changes and size expansion over time); Real-time alignment between the scene space and the real environment; Associated interaction points (based on) (Recognition), allowing users to adjust input later through gestures, voice, menus, etc.

[0060] Furthermore, the garden simulation scene provides an intuitive interface for the subsequent S400 dynamic parameter interaction adjustment module to dynamically adjust plant community parameters, and maintains consistency with the aforementioned rendered scene data. and scene model Real-time binding ensures that user adjustments directly affect the currently displayed simulation model.

[0061] Step S400 includes at least steps S410-S430: S410. Obtain the user's parameter adjustment interaction commands in the garden simulation scene through the augmented reality device. The parameter adjustment interaction commands include adjustments to plant type, distribution density, and growth cycle, and are parsed into an adjustment parameter sequence A={a1,a2,...,aᵧ}. Specifically, the augmented reality device displays a simulated garden scene dynamically rendered and output by S330. At that time, based on graphical menus, gesture interaction, or voice input, it receives operation commands from users to adjust the plant community in the current garden scene.

[0062] Furthermore, the user-input parameter adjustment interaction instructions at least include: The plant type adjustment parameter a1 indicates whether to replace or add a target plant species; The distribution density adjustment parameter a2 indicates the adjustment of the planting density of plants per unit area in the scene; The growth cycle adjustment parameter a3 indicates the adjustment of the plant's growth characteristics, such as the start and end times of its life cycle and flowering time. Other personalized parameters related to the configuration of garden plants (such as color, viewing period) a4,...,aᵧ.

[0063] The augmented reality device parses the user's adjustment interaction commands to obtain a sequence of adjustment parameters: ; Where γ represents the number of adjustment terms input by the user, and the adjustment parameter sequence A will be input into the dynamic evolution model in subsequent steps. (Generated by S230) Make real-time adjustments.

[0064] S420. Input the adjusted parameter sequence A into the plant community dynamic evolution model M, and adjust the model based on the following dynamic update function. Specifically, the plant community dynamic evolution model output from step S230 is invoked. The adjustment parameter sequence A is input to the dynamic model processing unit for dynamic adjustment.

[0065] For the aforementioned adjustment parameter sequence A, combined with the original dynamic evolution model The simulation model is incrementally corrected based on the following dynamic update function to obtain the adjusted dynamic evolution model. : ; in: This represents the adjusted model of plant community dynamics; Δ (A) is the dynamic model increment calculated based on the adjusted parameter sequence A, specifically including: ; Where: ϕⱼ is the dynamic adjustment influence coefficient corresponding to each adjustment parameter aⱼ, used to control the proportion of the influence of the adjustment parameter on the overall community model; aⱼ is the j-th adjustment content in the adjustment parameter sequence A.

[0066] Furthermore, the dynamic update function can realize the dynamic evolution model structure update based on the user's real-time adjustment commands, and apply it in real time. Generate new ,Should It includes updated information such as new plant types, distribution density, and growth characteristics.

[0067] The Data structures and Maintaining consistency includes: Plant spatial coordinates L'ᵢ; Plant size information: H'ᵢ, W'ᵢ; Dynamic indicators such as color and growth cycle C'ᵢ; Adjusted interaction identifier ID'ᵢ.

[0068] S430. The adjusted dynamic evolution model M' is returned to the augmented reality device for real-time updating and rendering, forming a garden simulation scene R'(t) that changes synchronously with user interaction, for continuous user operation. Furthermore, the adjusted plant community dynamic evolution model M' s The data is then returned to the augmented reality device for dynamic rendering, generating an updated garden simulation scene R'(t) in real time. s This is used to instantly display the effects of user-adjusted garden landscapes.

[0069] Specifically, based on the following update rendering function f' s (·), the dynamic evolution model M' s With the on-site environmental parameter L s Perform adaptive spatial matching to generate an updated rendering scene R'(t) s ): ;

[0070] Where: R'(t) s () represents a dynamic garden scene that reflects user adjustments; f' s (·) is the adaptive update rendering function, which generates graphic data in real time in response to user adjustments; M' s The adjusted dynamic evolution model; L s To augment the real-time capture of on-site environmental parameters by the augmented reality device (consistent with S320).

[0071] The R'(t) s )Include: The latest spatial distribution, size, color, and other appearance characteristics of plants; The effects of new plant types and their distribution density adjustment; Dynamic growth cycle changes (such as flowering, leaf fall, etc.).

[0072] Furthermore, the R'(t) s As an instant feedback scenario for user interaction in augmented reality devices, it allows users to input dynamic parameters for the next round, realizes cyclical interactive operations, and maintains data continuity with the subsequent S500 ecosystem optimization analysis module.

[0073] Step S500 includes at least steps S510-S530: S510, obtaining the optimization request instruction input by the user in the augmented reality device, and extracting the adjusted garden simulation scene R'(t) as the analysis input. Specifically, based on the augmented reality device, the user's view during the display of the dynamically adjusted garden simulation scene R'(t) after S430 is obtained. s In the simulation, an optimization request command is input through the interactive interface. The optimization request command is used to request an ecological optimization assessment of the current garden simulation scene.

[0074] Furthermore, in response to the optimization request instruction, the augmented reality device extracts the adjusted garden simulation scene R'(t) from the current simulation display state. s ), as data input for ecological analysis, wherein R'(t) s )Include: Adjusted plant spatial distribution parameter L'ᵢ; Plant size characteristic parameters H'ᵢ, W'ᵢ; Plant community density distribution index ; Plant light exposure duration and fitness parameters ; Spatial compatibility parameters of plants with other objects in a scene ; Other growth data fields related to plant ecology.

[0075] The R'(t) s This will serve as input for subsequent AI ecosystem optimization analysis and will be integrated into the calculation of the ecosystem balance function in the S520 module.

[0076] S520. Call the AI ​​analysis module to evaluate the ecological indicators of the garden simulation scene R'(t) based on the ecological balance function E(x), and calculate as follows: Specifically, the AI ​​analysis module is invoked to analyze the garden simulation scene R'(t) obtained by the aforementioned S510. s The system automatically analyzes ecological data in the data, focusing on the ecological balance of plant communities, based on the following ecological balance function E. a (x) Conduct ecological indicator assessment: ; Where: E a (x) represents the current comprehensive score of the ecological balance of the garden scene; To adjust the plant community density index (distribution of plant quantity per unit area) in the post-garden simulation scene. The overall suitability coefficient for the duration of light exposure received by the plant in the current environment; α is an indicator of spatial harmony between plants and other landscape elements within a scene. a β a γ a These are the weighting coefficients for community density, light suitability, and spatial coordination, which are dynamically adjusted according to the ecological optimization model to satisfy α. a +β a +γ a =1.

[0077] Furthermore, the AI ​​analysis module, based on R'(t) s The comprehensive ecological balance score E of the current scenario is calculated using real-world scenario data extracted from the data. a (x), used for generating subsequent optimization suggestions, the E a The (x) value will be used directly as the basis for optimization in step S530.

[0078] S530. Based on the ecological indicator evaluation result E(x), generate corresponding ecological optimization suggestions O={o1,o2,...,o...} k The optimization suggestions will be presented to the user in an interactive manner using augmented reality devices, allowing the user to further adjust the garden simulation scene. Furthermore, based on the ecological balance score E calculated in step S520... a (x), the AI ​​analysis module automatically generates corresponding ecological optimization suggestions, resulting in an optimization suggestion set: ; Wherein: o1 represents optimization suggestions for community density, such as increasing or decreasing the plant distribution density in a specific area; o2 represents adjustment suggestions for plant light requirements, such as changing shade-tolerant / sun-tolerant plant species; o3 represents adjustment suggestions for spatial layout, such as increasing gaps or rearranging specific communities; o k This indicates other proposed adjustments related to ecological balance.

[0079] Specifically, the optimization suggestion O is automatically generated by the AI ​​module in conjunction with the ecological assessment results, and is displayed to the user through the visual interactive interface of the augmented reality device, including: Graphical optimization suggestions (e.g., dynamic indicators for suggested areas); Voice prompts or text descriptions; Optimization options are available for users to click and select.

[0080] Furthermore, each suggestion in the optimization suggestion O is bound to a corresponding parameter modification scheme, allowing users to perform subsequent operations directly in the augmented reality device via gestures, voice, or menu selection. The data structure of the optimization suggestion is consistent with the parameter adjustment instructions manually input by the user in the aforementioned S410 step, ensuring a closed-loop data chain for use in the next round of S410-S430 iterations.

[0081] The key innovations of this invention include: (1) This invention proposes for the first time a dynamic generation mechanism for interactive data of garden scenes based on augmented reality devices. By combining user input instructions, it automatically extracts BIM spatial structure data and plant community growth data to form standardized interactive data for subsequent dynamic simulation and interactive adjustment, thereby improving the data dynamic generation capability of garden landscape design.

[0082] (2) This invention proposes a method for dynamically generating plant community growth and evolution models based on generative adversarial networks. By combining the simulation requests input by the user, dynamic models of plant communities that conform to different time series and seasonal changes are automatically generated, thereby improving the realism and diversity of plant dynamic growth simulation in landscape design.

[0083] (3) This invention proposes a mechanism for dynamically adjusting the plant community model based on user interaction input. Combined with augmented reality devices, it responds in real time to user adjustments to parameters such as plant type, distribution density, and growth cycle, and dynamically updates the garden simulation scene, thereby improving the human-computer interaction efficiency and real-time adjustment capability of garden design.

[0084] (4) This invention proposes a method for ecological balance analysis of garden landscape based on AI analysis module. For the garden simulation scene adjusted by the user, it automatically analyzes ecological indicators such as plant community density, light suitability, and spatial coordination, and outputs scientific optimization suggestions to improve the ecological rationality and sustainability of garden landscape scheme.

[0085] (5) This invention is the first to construct an integrated system of human-computer interaction and dynamic feedback based on augmented reality, which connects the entire process of garden landscape from scene generation, dynamic simulation, interactive adjustment to ecological optimization, forming a complete closed loop of dynamic planning, real-time adjustment and ecological feedback, improving the intelligent level of dynamic design of garden landscape, and conforming to the development trend of interactive technology and dynamic generation.

[0086] The following are its main beneficial effects: (1) This invention obtains the garden scene call command input by the user through augmented reality device, integrates the data of BIM model and plant growth database, dynamically generates standardized garden scene interactive data, realizes the unified processing of spatial structure data and plant community growth data in the process of garden landscape design, solves the problem of data dispersion and inability to generate interactive data in traditional garden design, and improves the data processing efficiency and real-time interactive capability of garden landscape design.

[0087] (2) This invention generates a plant community growth and evolution model dynamically through a generative adversarial network and combines it with augmented reality devices for three-dimensional projection and rendering display. This allows users to view the dynamic growth effects of plants under different time periods and seasonal conditions, solving the problem that traditional garden design cannot intuitively display plant growth and evolution, improving the dynamic preview and simulation effect of garden landscape schemes, and enhancing the scientific nature of users' design decisions.

[0088] (3) The present invention dynamically adjusts the plant community configuration parameters through user interaction commands in augmented reality devices, and combines the AI ​​analysis module to conduct ecological rationality analysis on the adjusted garden scene, automatically outputs optimization suggestions such as plant distribution and spatial layout, forming a closed-loop process of design, interactive adjustment and ecological assessment, solving the problem that traditional garden design cannot be dynamically adjusted and lacks ecological optimization feedback, and improving the intelligence and ecological adaptability of dynamic planning of garden landscape.

[0089] Example 2: Figure 2 A structural block diagram of an augmented reality-based dynamic planning system for garden landscapes, according to an embodiment of the present invention, is shown. Figure 2 As shown, the structure may include: The garden scene data fusion module 10 is used to respond to the garden scene call command input by the user through the augmented reality device. Based on the target site number and plant type parameters input by the user through voice, gesture or menu, it extracts the garden spatial structure data corresponding to the site number from the BIM model database, and extracts the plant community growth cycle data and plant spatial distribution data related to the plant type parameters from the plant growth database. Then, it standardizes and formats the above-mentioned extracted data and merges them to generate garden scene interactive data for subsequent dynamic simulation and interactive operation, and ensures that the spatial data and plant growth data can be synchronously called in the augmented reality device, supporting real-time rendering and parameter adjustment.

[0090] The garden landscape dynamic simulation module 20 is used to automatically generate a dynamic plant community model that matches the user's request parameters after receiving a garden landscape simulation request input by the user through the augmented reality device, based on the garden scene interaction data and calling a generative adversarial network that includes plant life cycle algorithms and plant community evolution logic. This model can reflect the growth dynamics of plants under different time periods and seasonal changes, and includes multi-dimensional data such as the spatial location, size, color, and density of individual plants within the community. After generation, the dynamic plant community model serves as a dynamic rendering data source for subsequent display by the augmented reality device.

[0091] The augmented reality dynamic rendering module 30 is used to generate a three-dimensional visualized garden landscape simulation scene in real time based on a dynamic plant community model. By extracting plant community data at the user's requested time point from the dynamic plant community model and combining it with the on-site spatial environment parameters collected in real time by the augmented reality device, it performs three-dimensional graphics rendering and spatial matching, and blends the simulated plant landscape with the spatial environment according to the real scale. The simulation is then dynamically projected onto the display interface of the augmented reality device, realizing a visualized presentation of the plant community dynamically adjusting with changes in time and environment, allowing users to interactively observe, analyze, and design adjustments.

[0092] The garden landscape parameter interactive adjustment module 40 is used to receive parameter adjustment commands from the user via gestures, voice, interactive menus, etc., to adjust parameters such as plant type, planting density, and growth cycle, based on the dynamic display of the garden simulation scene on the augmented reality device. It parses these commands into a standardized parameter adjustment sequence in real time, calls the dynamic plant community model, and dynamically applies the user's adjustment commands to the internal parameters of the model to form a new adjusted model. At the same time, it automatically renders and updates the adjusted model, so that the adjustment results are synchronously reflected in the garden simulation scene displayed on the augmented reality device. This realizes a real-time closed-loop interactive process between user adjustment and system feedback, thereby meeting the user's dynamic design adjustment needs.

[0093] The ecological optimization analysis and feedback module 50 is used to respond to the ecological optimization request issued by the user through the augmented reality device after the user completes parameter adjustment. Based on the adjusted garden simulation scene, it extracts key ecological indicators such as plant community density, light suitability, and spatial coordination, and calls the AI ​​analysis module to perform ecological balance analysis. Through the preset ecological balance model, it automatically evaluates the rationality of the current garden landscape configuration, and further generates optimization suggestions for plant density, light configuration, spatial distribution, etc. based on the analysis results. The optimization suggestions are fed back to the user through the graphical interface, voice prompts, interactive menus, etc. of the augmented reality device, so that the user can continue to adjust the garden scene according to the system suggestions, thereby completing the dynamic interactive cycle of design-feedback-optimization.

[0094] Obviously, the embodiments described above are only some embodiments of this application, not all embodiments. The accompanying drawings show preferred embodiments of this application, but do not limit the patent scope of this application. This application can be implemented in many different forms; rather, the purpose of providing these embodiments is to provide a more thorough and comprehensive understanding of the disclosure of this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this application's specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of patent protection of this application.

Claims

1. A dynamic planning method for garden landscape based on augmented reality, characterized in that, The method includes the following steps: Based on augmented reality devices, the system obtains user input commands to invoke garden scenes, and in response to these commands, extracts data from the BIM model and plant database, and integrates them to generate interactive garden scene data. Based on the user-input simulation request command, a generative adversarial network is invoked to dynamically simulate the interactive data of the garden scene, generating a dynamic evolution model of the plant community. The generative adversarial network generates the dynamic evolution model of the plant community through the following function: ； Where G(z|D,Q) is the generator, z is the noise vector, D is the standardized garden scene interaction data, Q is the simulation request sequence; GAN is the adversarial function. Based on the user's input display instructions, the dynamic evolution model of the plant community is projected and rendered in three dimensions on an augmented reality device to generate a garden simulation scene. Based on the user's interactive input in the garden simulation scene, the dynamic evolution model of the plant community is dynamically adjusted to generate the adjusted dynamic evolution model, and the adjusted garden simulation scene is rendered and displayed in real time. Based on the user-inputted optimization request, the AI ​​analysis module is invoked to automatically analyze the ecologically relevant data in the acquired garden simulation scene. Regarding the ecological balance of the plant community, the following ecological balance function E is used: a (x) Conduct ecological indicator assessment: ； Among them: E a (x) represents the current comprehensive score of the ecological balance of the garden scene; To adjust the plant community density index (distribution of plant quantity per unit area) in the post-garden simulation scene. The overall suitability coefficient for the duration of light exposure received by the plant in the current environment; α is an indicator of spatial harmony between plants and other landscape elements within a scene. a β a γ a These are the weighting coefficients for community density, light suitability, and spatial coordination, which are dynamically adjusted according to the ecological optimization model to satisfy α. a +β a +γ a =1.

2. The augmented reality-based dynamic planning method for garden landscapes according to claim 1, characterized in that, Before generating the garden scene interaction data, the standardized fusion process includes: The spatial structure data and plant community data are fused according to preset weighting coefficients to generate fused data. The weighting coefficients include the weight λ1 of the spatial structure data and the weight λ2 of the plant community data.

3. The augmented reality-based dynamic planning method for garden landscapes according to claim 1, characterized in that, The simulation request sequence includes time period parameters and season type parameters, which are set by user input. The generative adversarial network dynamically generates plant community dynamic evolution models under different time series and seasons based on the simulation request sequence.

4. The augmented reality-based dynamic planning method for garden landscapes according to claim 1, characterized in that, The three-dimensional projection and rendering steps include: Based on the on-site spatial environment parameters obtained by the augmented reality device, the dynamic evolution model of the plant community is subjected to adaptive rendering processing of spatial location, scale and lighting. The rendered simulation scene is then projected into the user's spatial environment and dynamically integrated with the real space.

5. The augmented reality-based dynamic planning method for garden landscapes according to claim 1, characterized in that, The user's interactive input in the garden simulation scene includes plant type selection, plant distribution density adjustment, and growth cycle setting. The interactive input is parsed into a parameter adjustment sequence and input into the plant community dynamic evolution model for dynamic adjustment.

6. The augmented reality-based dynamic planning method for garden landscapes according to claim 1, characterized in that, The adjusted dynamic evolution model generated by the dynamic adjustment includes: Based on the original dynamic evolution model and the incremental adjustment value generated by the parameter adjustment sequence, the adjusted dynamic evolution model is generated by the following formula: ； Where M is the original dynamic evolution model, ΔM(A) is the dynamic adjustment increment corresponding to the parameter adjustment sequence A, and M' is the adjusted dynamic model.

7. The augmented reality-based dynamic planning method for garden landscapes according to claim 1, characterized in that, The ecological analysis includes a comprehensive evaluation of plant community density, light suitability, and spatial coordination in the adjusted garden simulation scene, and calculates an ecological balance score based on the ecological balance function to generate optimization suggestions that include plant distribution adjustment, density optimization, and spatial reconstruction.

8. The augmented reality-based dynamic planning method for garden landscapes according to claim 1, characterized in that, The ecological optimization suggestions are presented through the graphical interface, voice prompts, or interactive menus of augmented reality devices, allowing users to further adjust the garden simulation scene based on these suggestions.

9. The augmented reality-based dynamic planning method for garden landscapes according to claim 1, characterized in that, The dynamic rendering and ecological feedback form a closed-loop interaction mechanism. After the ecological optimization suggestion is fed back, the user adjusts the parameters based on the suggestion and automatically enters the next round of plant community dynamic evolution model generation and rendering update process.

10. The augmented reality-based dynamic planning method for garden landscapes according to claim 1, characterized in that, The dynamic planning method for garden landscapes includes real-time tracking of the user's perspective and spatial position using augmented reality devices, adaptively adjusting the perspective of the dynamically rendered garden simulation scene to ensure the continuity and spatial consistency of the landscape display during user interaction.

Images