A landscape planting planning and management system based on terrain environment analysis

Abstract

The present application discloses a landscape garden planting planning management system based on terrain environment analysis, and relates to the technical field of garden planting planning. The system includes the following modules: a data acquisition and processing module, which collects environmental data of the target garden area; a data feature extraction module, which generates a multi-dimensional environmental feature vector; an environmental zone division module: clustering areas with similar environmental characteristics into the same category, forming different environmental zones, and obtaining different environmental zone characteristics; a plant demand module, which forms an intelligent plant demand map; an environmental zone plant matching module, which intelligently matches plant species suitable for each environmental zone, and outputs a recommended planting plan through a deep learning model. Using data visualization technology, the environmental zone characteristics and plant matching schemes are graphically displayed, and an environmental zone division map and a plant planting recommendation map are generated, which intuitively display the location, characteristics and recommended plant species of different environmental zones.

Description

Technical Field

[0001] The invention relates to the technical field of garden planting planning, and in particular to a landscape garden planting planning management system based on terrain environment analysis. Background Art

[0002] Garden plants are plant materials suitable for landscaping, including woody and herbaceous flowering, foliage or fruiting plants, as well as protective plants and economic plants suitable for gardens, green spaces and scenic spots. Plants used for indoor flower decoration are also garden plants. Garden plants are divided into two categories: woody garden plants and herbaceous garden plants. In addition, they also include ferns, aquatic plants, cactus succulents and carnivorous plants. As people's living standards continue to improve, people's pursuit of the quality of garden plants is gradually increasing.

[0003] At present, garden plants are often planned and managed through artificial selection during planting, which can easily place plants in an environment that is not conducive to their growth, thus causing problems such as difficulty in plant growth.

[0004] The document with the prior art publication number CN115936304A provides a landscape planting planning and management system based on terrain environment analysis, which obtains the terrain information of the target garden and the sunshine conditions in different areas, and automatically deletes plants that are not suitable for planting in combination with the sunshine conditions, soil pH information and rainfall, thereby reducing the workload of selecting plants. Plants are classified according to the different lighting conditions that can be obtained by garden plants at different levels, so that the matching of different height levels of garden plants is more reasonable, and preferred and secondary plants are provided according to the user's intentions and needs, thereby improving the efficiency of garden planning.

[0005] Although the above-mentioned system and method have shown remarkable results in the field of garden planting planning, this method does not take into account the specific environmental differences in different areas of the garden, such as light intensity, altitude, water accumulation, etc., and does not consider the rationality of different plants in the same area. The classification of plants is relatively simple and cannot generate detailed planting plans. In addition, this method requires the use of artificially drawn contour lines and simulated light data, which will lead to inaccurate data.

[0006] In view of this, we proposed a landscape planting planning and management system based on terrain environment analysis. Summary of the invention

[0007] The present invention provides a landscape gardening planting planning and management system based on terrain environment analysis, and the technical solution includes the following aspects.

[0008] According to one aspect of an embodiment of the present application, a landscape planting planning and management system based on terrain environment analysis is provided, and the system includes the following modules.

[0009] The data acquisition and processing module collects the environmental data of the target garden area and processes and standardizes the environmental data using data cleaning and normalization algorithms.

[0010] The data feature extraction module uses data feature extraction technology to extract data features from the processed environmental data, integrates the data features, and generates a multi-dimensional environmental feature vector.

[0011] The environmental zone division module applies an adaptive clustering algorithm to perform cluster analysis on garden areas based on environmental feature vectors, automatically determining the optimal number of clusters and boundaries; clustering areas with similar environmental characteristics into the same category to form different environmental zones and obtain different environmental zone characteristics.

[0012] The plant demand module builds a plant database and combines knowledge graph technology to associate plant characteristics with environmental zone characteristics to form an intelligent plant demand map.

[0013] The environmental zone plant matching module uses the plant demand map based on the characteristics of the environmental zones to intelligently match plant species suitable for each environmental zone, and outputs a recommended planting plan through a deep learning model.

[0014] The environmental zone visualization module uses data visualization technology to graphically display environmental zone characteristics and plant matching plans, generate environmental zone division maps and plant planting recommendation maps; and intuitively display the location, characteristics and recommended plant species of different environmental zones.

[0015] As an optional solution of the present application, the data acquisition and processing module arranges a sensor group in the target garden area to cover the monitoring of key environmental factors.

[0016] The sensor group includes, but is not limited to, an altimeter for measuring altitude, a light sensor for measuring the duration and intensity of sunlight, a temperature sensor, a precipitation sensor, a water accumulation detection sensor, a soil moisture sensor, and a pH sensor.

[0017] The sensor group collects environmental data in real time or at a preset time interval.

[0018] The sensor group transmits environmental data and stores the environmental data in the database, which adopts structured storage method.

[0019] As an optional solution of the present application, the data feature extraction module extracts the altitude data features, light data features, temperature data features, precipitation data features, water accumulation data features, soil moisture data features and soil pH data features of each sensor group detection area, integrates the data features, and generates a multi-dimensional environmental feature vector F.

[0020] As an optional solution for this application.

[0021] Elevation data features include mean altitude, altitude standard deviation, maximum altitude, minimum altitude, and altitude range.

[0022] The characteristics of light data include average daily light duration, maximum light duration, minimum light duration, average light intensity and standard deviation of light intensity.

[0023] Temperature data features include average temperature, maximum temperature, minimum temperature, temperature fluctuation range, and temperature standard deviation.

[0024] The characteristics of precipitation data include average daily precipitation, maximum precipitation, precipitation frequency, total precipitation and precipitation standard deviation.

[0025] The characteristics of waterlogging data include waterlogging frequency, average waterlogging depth, maximum waterlogging depth, waterlogging duration and waterlogging location distribution.

[0026] Soil moisture data characteristics include average soil moisture, maximum soil moisture, minimum soil moisture and soil moisture standard deviation.

[0027] The soil pH data characteristics include average pH, maximum pH, minimum pH and pH standard deviation.

[0028] As an optional solution of the present application, the environmental zone division module obtains a multi-dimensional environmental feature vector F from the data feature extraction module. The environmental feature vector F is used as the input data of the clustering algorithm. The adaptive clustering algorithm GMM clustering is selected for clustering analysis. The optimal GMM model is used to generate a corresponding environmental zone label for each data point Fi. Specifically, according to the calculated responsibility degree γ ik , each data point F i Assign it to the Gaussian distribution k that maximizes the responsibility, thereby determining the environmental zone label of each data point, label(F i )=argmax k γ ik ,The clustering results and environmental zone characteristics are stored in a database, which also includes the number of each environmental zone and the corresponding environmental feature vector.

[0029] As an optional solution of the present application, the plant demand module constructs a comprehensive plant database, which contains information such as the growth habits, light requirements, temperature tolerance range, humidity requirements, water tolerance range, etc. of various plants. Specifically, the plant characteristics of the plant database include plant name, growth habits, light requirements, temperature tolerance range, humidity requirements, water tolerance range, soil pH adaptation range, and specific requirements. Combined with knowledge graph technology, plant characteristics are associated with environmental zone characteristics. Knowledge graph is a graph structure that represents entities and their relationships, which can be used to store and query complex related information. The specific steps are to define entities and relationships. Entities include plant characteristics and environmental feature vector requirements. Relationships include adaptation and need. A knowledge graph is created. RDF graph database technology is used to store the relationship between plant characteristics and environmental zone characteristics in the graph database to form an intelligent plant demand graph.

[0030] As an optional solution of the present application, the environmental zone plant matching module uses the plant demand map to intelligently match plant species suitable for each environmental zone based on the environmental zone characteristics obtained by cluster analysis. Specifically, the environmental feature vector F is obtained from the environmental zone division module. k , query the knowledge graph, find the plant species that match the characteristics of the environmental zone, and use the query language SPARQL of the graph database to query the plants that adapt to these characteristics based on the environmental feature vector.

[0031] Calculate its fitness score, calculate its user preference score, and then calculate the comprehensive score of each plant.

[0032] According to the number of plant species required in the environmental zone, the rationality scores of different plant species combinations were calculated taking into account the symbiotic relationship, resource competition and ecological balance among the plants in the combination.

[0033] Based on the comprehensive score and rationality score, a score is generated for each plant matching scheme, and the scores are sorted to form a list. Users can determine the final plant matching scheme based on the scheme score list.

[0034] A specific planting plan is generated based on the final plant matching scheme, the area of ​​the environmental zone, and the environmental characteristics of the environmental zone. The planting plan includes the planting time and planting density of various plants in the plant matching scheme combination.

[0035] As an optional solution for this application, plants are matched to solution scores.

[0036] For each plant P j , calculate its fitness score S j , S j =w1γ ij 海拔 +w2γ ij 光照 +w3γij 温度 +w4γ ij 湿度 +w5γ ij 积水 +w6γ ij 酸碱度 .

[0037] Among them, γ ij * Indicates plant P j The fitness of the characteristic in environment zone i, fitness γ ij The value of can be calculated by the degree of match between the plant characteristics recorded in the plant requirement map and the characteristics of the environmental zone.

[0038] Calculate its user preference score. Users can provide plant height and maintenance requirements, and calculate the user preference score U of each plant through the knowledge graph j , U j =∑ k=1 m w k u jk , where U j It is a plant j The user preference score, w k is the weight of the kth user demand, u jk It is a plant j Corresponding to the matching degree of the kth requirement, users with higher professional level can also directly assign user preference scores to plants.

[0039] Calculate the comprehensive score M for each plant j , M j =w j S j +(1-w j ) j .

[0040] Users can also provide the required number of plant species in the environmental zone and query plant species whose comprehensive scores exceed the set threshold.

[0041] According to the number of plant species required in the environmental area, the rationality score R of different plant species combinations is calculated by considering factors such as the symbiotic relationship, resource competition and ecological balance among the plants in the combination. c , R c =2 / n(n-1)∑ i=1 n-1 ∑ j=i+1 n r ij .

[0042] Among them, r ij It is a plant i and Pj The rationality score between them, n is the number of plant species in the combination.

[0043] According to the comprehensive score M j and the rationality score R c , generate each plant matching scheme score T c , and sort them to form a list, T c =αM j +βR c .

[0044] Among them, α and β are weight parameters for adjusting the comprehensive score and rationality score.

[0045] As an optional solution of the present application, an environmental zone visualization module.

[0046] The environmental zone visualization module obtains data from the environmental zone division module, plant requirement module and deep learning model.

[0047] Based on the environmental zone classification results, data visualization tools are used to display the locations and boundaries of different environmental zones, using different colors or line styles to represent different environmental zones.

[0048] Use polygons to draw each environmental zone, ensure that the boundaries of each environmental zone are clearly visible, and mark the recalculated feature information within the environmental zone.

[0049] Add text labels or icons at appropriate locations in the environment area to indicate the numerical value or level of the environmental characteristics.

[0050] Display the recommended plant species and their matching scores in each environmental zone j , the final solution score T c and detailed planting plan drawings.

[0051] Provides interactive functions to allow users to operate and query the visualization results.

[0052] According to one aspect of an embodiment of the present application, a landscape planting planning and management method based on terrain environment analysis is provided, and the method includes the following steps.

[0053] S1. Collect environmental data and use data cleaning and normalization algorithms to process and standardize the collected data to ensure the accuracy and consistency of the data.

[0054] S2. Use data feature extraction technology to extract key features from the processed environmental data and integrate the features to generate a multi-dimensional environmental feature vector.

[0055] S3. Apply the adaptive clustering algorithm to perform cluster analysis on the garden area according to the environmental feature vector, automatically determine the optimal number of clusters and boundaries, and cluster areas with similar environmental characteristics into the same category to form different environmental zones.

[0056] S4. Build a comprehensive plant database, combine knowledge graph technology, associate plant characteristics with environmental zone characteristics, and form an intelligent plant demand map.

[0057] S5. Based on the environmental zone characteristics obtained through cluster analysis, the plant demand map is used in combination with the user's personalized needs to intelligently match plant species suitable for each environmental zone, and a recommended planting plan is output through a deep learning model.

[0058] As an optional solution of the present application, the method further includes the following steps.

[0059] S6. Use data visualization technology to graphically display the characteristics of environmental zones and plant matching plans, generate environmental zone division maps and plant planting recommendation maps, and intuitively display the location, characteristics and recommended plant species of different environmental zones.

[0060] Compared with the prior art, the present application has the following beneficial effects.

[0061] 1. Through multi-dimensional environmental feature extraction and adaptive clustering algorithm, the accurate division of garden area environment is achieved, and the accuracy and scientificity of environmental zoning are improved.

[0062] 2. The combination of the comprehensive plant database and knowledge graph technology has formed an intelligent plant demand map, which automatically matches plants with environmental characteristics and improves the accuracy and efficiency of plant species selection. Through the association query function of the knowledge graph, users can easily obtain all plant species and their detailed information that match a specific environmental zone, providing strong support for subsequent planting planning.

[0063] 3. The training and optimization of deep learning models further enhances the intelligent recommendation capability of plant matching solutions. By learning from a large amount of historical data, the model can deeply understand the complex relationship between plant needs and environmental feature vectors, thereby providing users with more accurate and personalized plant matching solutions.

[0064] 4. With the addition of the environmental zone visualization module, users can clearly see the location and boundaries of different environmental zones, environmental characteristics, and recommended plant species through graphical display. It provides interactive functions, allowing users to operate and query the visualization results, further enhancing the user's sense of participation and experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 A schematic diagram of the structure of a landscape planting planning and management system based on terrain environment analysis provided for one embodiment of the present application.

[0066] Figure 2 A flow chart of a landscape planting planning and management method based on terrain environment analysis is provided for one embodiment of the present application. DETAILED DESCRIPTION

[0067] In order to make the purpose, technical scheme and advantages of the implementation of the present invention clearer, the technical scheme in the embodiment of the present invention will be described in more detail below in conjunction with the drawings in the embodiment of the present invention. In the drawings, the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The described embodiments are part of the embodiments of the present invention, not all of the embodiments.

[0068] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0069] The embodiments and directional terms described below with reference to the accompanying drawings are exemplary and intended to be used to explain the present invention, but should not be construed as limiting the present invention.

[0070] like Figure 1 As shown, a landscape planting planning and management system based on terrain environment analysis includes the following modules.

[0071] The data acquisition and processing module collects the environmental data of the target garden area and processes and standardizes the environmental data using data cleaning and normalization algorithms.

[0072] The data feature extraction module uses data feature extraction technology to extract data features from the processed environmental data, integrates the data features, and generates a multi-dimensional environmental feature vector.

[0073] The environmental zone division module applies an adaptive clustering algorithm to perform cluster analysis on garden areas based on environmental feature vectors, automatically determining the optimal number of clusters and boundaries; clustering areas with similar environmental characteristics into the same category to form different environmental zones and obtain different environmental zone characteristics.

[0074] The plant demand module builds a plant database and combines knowledge graph technology to associate plant characteristics with environmental zone characteristics to form an intelligent plant demand map.

[0075] The environmental zone plant matching module uses the plant demand map based on the characteristics of the environmental zones to intelligently match plant species suitable for each environmental zone, and outputs a recommended planting plan through a deep learning model.

[0076] The environmental zone visualization module uses data visualization technology to graphically display environmental zone characteristics and plant matching plans, generate environmental zone division maps and plant planting recommendation maps; and intuitively display the location, characteristics and recommended plant species of different environmental zones.

[0077] As an optional solution of the present application, a data acquisition and processing module arranges a sensor group in the target garden area to ensure coverage of monitoring of all key environmental factors. The sensor group includes but is not limited to an altimeter for measuring altitude, a light sensor for light duration and intensity, a temperature sensor, a precipitation sensor, a water accumulation detection sensor, a soil moisture sensor and a pH sensor. The sensor group collects environmental data in real time or at a scheduled time according to a preset time interval. The sensor group transmits the environmental data and stores the environmental data in a database, which uses a structured storage method.

[0078] The environmental data are processed and standardized using data cleaning and normalization algorithms; ensuring that the organization and management of the data are convenient for subsequent processing and analysis. Data cleaning includes missing value filling, outlier detection and correction, and removal of duplicate data. Specifically, missing values ​​in environmental data are filled using interpolation, mean filling, or a prediction model; abnormal data points are detected and corrected using an isolation forest algorithm; duplicate collection records are removed using timestamps and sensor IDs; the cleaned environmental data are normalized using a data normalization algorithm to ensure that the dimensions of data from different sensors are consistent; the data are scaled to the [0, 1] interval to facilitate comparison between different features; and the cleaned and normalized data are stored in a high-performance database for subsequent data feature extraction and analysis.

[0079] As an optional solution of the present application, the data feature extraction module extracts the altitude data features, illumination data features, temperature data features, precipitation data features, water accumulation data features, soil moisture data features and soil pH data features of each sensor group detection area, wherein the altitude data features include the average altitude, altitude standard deviation, maximum altitude, minimum altitude and altitude range; the illumination data features include the average daily illumination duration, maximum illumination duration, minimum illumination duration, average illumination intensity and standard deviation of illumination intensity; the temperature data features include the average temperature, maximum temperature, minimum temperature , temperature fluctuation range and temperature standard deviation, precipitation data characteristics include average daily precipitation, maximum precipitation, precipitation frequency, total precipitation and precipitation standard deviation, waterlogging data characteristics include waterlogging frequency, average waterlogging depth, maximum waterlogging depth, waterlogging duration and waterlogging location distribution, soil moisture data characteristics include average soil moisture, maximum soil moisture, minimum soil moisture and soil moisture standard deviation, soil pH data characteristics include average pH, maximum pH, minimum pH and pH standard deviation. The above extracted data features are integrated to generate a multi-dimensional environmental feature vector F.

[0080] As an optional solution of the present application, the environmental zone division module obtains a multi-dimensional environmental feature vector F from the data feature extraction module. The environmental feature vector F will be used as the input data of the clustering algorithm. The adaptive clustering algorithm Gaussian Mixture Model (GMM) clustering is selected for clustering analysis, and the optimal GMM model is used to generate a corresponding environmental zone label for each data point Fi.

[0081] Set the initial number K of Gaussian distributions (set a larger initial value), use the expectation maximization (EM) algorithm for training, calculate the probability of each data point belonging to each Gaussian distribution in E steps, and calculate the responsibility γ ik , responsibility γ ik Indicates that at a given data point F i In the case of , the probability that the point belongs to the kth Gaussian distribution, the calculation formula for the responsibility degree is, γ ik =[π k N(F i ∣μ k ,Σ k )] / [∑ j=1 K π j N(F i ∣μ j ,Σ j )], where π k is the mixing coefficient of the kth Gaussian distribution, which represents the weight of the distribution in the overall model. i ∣μ k ,Σ k) are k Gaussian distributions at data point F i The probability density function value at k and Σ k are the mean vector and covariance matrix of the kth Gaussian distribution, N(F i ∣μ j ,Σ j ) is the data point F i The total probability density under all Gaussian distributions, M steps according to the probability update Gaussian distribution (responsibility γ ik ) parameters, responsibility degree γ ik Calculate the new mean, covariance and mixing coefficient by weighted average, and update the mixing coefficient π k , π k =N k / N, where N k =∑ i=1 N γ ik represents the number of valid samples belonging to the kth Gaussian distribution, N is the total number of data points, and the updated mean μ k , μ k =1 / N k ∑ i=1 N γ ik F i , update the covariance Σ k ,Σ k =1 / N k ∑ i=1 N γ ik (F i -μ k )(F i -μ k ) T , repeat the E step and the M step until convergence, use the Bayesian Information Criterion (BIC) or the Akaike Information Criterion (AIC) to select the optimal number of Gaussian distributions K, select the K that minimizes the BIC or AIC, and use the obtained optimal GMM model to generate the corresponding environmental zone label for each data point Fi.

[0082] Specifically, according to the calculated responsibility γ ik , each data point F i Assign it to the Gaussian distribution k that maximizes the responsibility, thereby determining the environmental zone label of each data point, label(F i )=argmax k γ ik , summarize and describe the characteristics of each environmental zone, and store the clustering results and environmental zone characteristics in the database. The database also includes the number of each environmental zone and the corresponding environmental feature vector.

[0083] As an optional solution of the present application, the plant requirement module constructs a comprehensive plant database, which contains information such as the growth habits, light requirements, temperature tolerance range, humidity requirements, waterlogging tolerance range, etc. of various plants.

[0084] Specifically, the plant characteristics of the database include plant name, growth habits, light requirements, temperature tolerance range, humidity requirements, waterlogging tolerance range, soil pH adaptability range, and other specific requirements. Combined with knowledge graph technology, plant characteristics are associated with data features. Knowledge graph is a graph structure that represents entities and their relationships, which can be used to store and query complex related information.

[0085] The specific steps are to define entities and relationships. Entities include plants, environmental characteristics, and requirements. Relationships include adaptation and need. A knowledge graph is created. RDF graph database technology is used to store the relationship between plant characteristics and data characteristics in the graph database, such as PlantA--[AdaptsTo]-->LightRequirement, PlantA--[Requires]-->TemperatureTolerance; an intelligent plant requirement graph is formed.

[0086] As an optional solution of the present application, the environmental zone plant matching module uses the plant demand map system to intelligently match plant species suitable for each environmental zone based on the environmental zone characteristics obtained by cluster analysis.

[0087] Specifically, the environmental feature vector F is obtained from the environmental zone division module. k , that is, recalculate the altitude data features, light data features, temperature data features, precipitation data features, water accumulation data features, soil moisture data features and soil pH data features for the areas classified as the same environmental zone, and generate a multi-dimensional environmental feature vector F k , query the knowledge graph, find the plant species that match the environmental zone characteristics, use the query language SPARQL of the graph database, according to the environmental feature vector F k Look for plants adapted to these characteristics.

[0088] For each plant P j , calculate its fitness score S j , S j =w1γ ij 海拔 +w2γ ij 光照 +w3γ ij 温度 +w4γ ij 湿度 +w5γ ij 积水 +w6γ ij酸碱度 .

[0089] Among them, γ ij * Indicates plant P j The fitness of the characteristic in environment zone i, fitness γ ij The value of can be calculated by the degree of match between the plant characteristics recorded in the plant requirement map and the characteristics of the environmental zone.

[0090] Calculate its user preference score. Users can provide plant height and maintenance requirements, and calculate the user preference score U of each plant through the knowledge graph j , U j =∑ k=1 m w k u jk , where U j It is a plant j The user preference score, w k is the weight of the kth user demand, u jk It is a plant j Corresponding to the matching degree of the kth requirement, users with higher professional level can also directly assign user preference scores to plants.

[0091] Calculate the comprehensive score M for each plant j , M j =w j S j +(1-w j ) j .

[0092] Users can also provide the required number of plant species in the environmental zone and query plant species whose comprehensive scores exceed the set threshold.

[0093] According to the number of plant species required in the environmental area, the rationality score R of different plant species combinations is calculated by considering factors such as the symbiotic relationship, resource competition and ecological balance among the plants in the combination. c , R c =2 / n(n-1)∑ i=1 n-1 ∑ j=i+1 n r ij .

[0094] Among them, r ij It is a plant i and P j The rationality score between them, n is the number of plant species in the combination.

[0095] According to the comprehensive score M j and the rationality score R c, generate each plant matching scheme score T c , and sort them to form a list, T c =αM j +βR c .

[0096] Among them, α and β are weight parameters for adjusting the comprehensive score and rationality score; users can determine the final plant matching plan based on the plan score list.

[0097] A deep learning model is trained to identify the complex relationship between plant characteristics and environmental zone characteristics by using a combination of environmental zone characteristics, plant species, and plant matching schemes in historical data to annotate planting time and planting density.

[0098] According to the final plant matching scheme, the area of ​​the environmental zone, and the environmental characteristics of the environmental zone, a specific planting plan is generated. The planting plan includes the planting time and planting density of various plants in the plant matching scheme combination. T ; θ), where y^ is the predicted planting plan, including the planting time and planting density of various plants in the plant matching scheme combination, F T is a comprehensive feature vector formed by the combination of environmental zone characteristics, plant species and plant matching schemes, and θ is a model parameter. Input the new comprehensive feature vector F T to the trained deep learning model and output the recommended planting plan.

[0099] As an optional solution of the present application, the environmental zone visualization module obtains data from the environmental zone division module, the plant demand module and the deep learning model, specifically including the environmental zone classification results, the plant species in each environmental zone and their matching scores M j , plant planting type scheme score T c , detailed planting plan, planting time and density. According to the environmental zone classification results, data visualization tools are used to display the location and boundaries of different environmental zones, and different colors or line styles are used to represent different environmental zones.

[0100] Use polygons to draw each environmental zone, ensure that the boundaries of each environmental zone are clearly visible, and mark the recalculated feature information within the environmental zone.

[0101] Add text labels or icons at appropriate locations in the environment area to indicate the numerical value or level of the environmental characteristics.

[0102] Display the recommended plant species and their matching scores in each environmental zone j , the final solution score T c and detailed planting plan drawings.

[0103] Provides interactive functions to allow users to operate and query the visualization results.

[0104] Please refer to Figure 2 ,A landscape planting planning and management method based on terrain environment analysis includes the following steps.

[0105] S1. Collect environmental data and use data cleaning and normalization algorithms to process and standardize the collected data to ensure the accuracy and consistency of the data.

[0106] S2. Use data feature extraction technology to extract key features from the processed environmental data and integrate the features to generate a multi-dimensional environmental feature vector.

[0107] S3. Apply the adaptive clustering algorithm to perform cluster analysis on the garden area according to the environmental feature vector, automatically determine the optimal number of clusters and boundaries, and cluster areas with similar environmental characteristics into the same category to form different environmental zones.

[0108] S4. Build a comprehensive plant database that includes information on the growth habits, light requirements, temperature tolerance range, humidity requirements, etc. of various plants. Combined with knowledge graph technology, associate plant characteristics with environmental zone characteristics to form an intelligent plant demand map.

[0109] S5. Based on the environmental zone characteristics obtained through cluster analysis, the plant demand map is used in combination with the user's personalized needs to intelligently match plant species suitable for each environmental zone, and a recommended planting plan is output through a deep learning model.

[0110] S6. Use data visualization technology to graphically display the characteristics of environmental zones and plant matching plans, generate environmental zone division maps and plant planting recommendation maps, and intuitively display the location, characteristics and recommended plant species of different environmental zones.

Claims

1. A landscape planting planning and management system based on terrain environment analysis, characterized in that: The system includes the following modules: A data collection and processing module collects environmental data of the target garden area and processes and standardizes the environmental data using data cleaning and normalization algorithms; The data feature extraction module uses data feature extraction technology to extract data features from the processed environmental data, integrates the data features, and generates a multi-dimensional environmental feature vector; The environmental zone division module applies an adaptive clustering algorithm to perform cluster analysis on the garden area according to the environmental feature vector, and automatically determines the optimal number of clusters and boundaries; clusters areas with similar environmental characteristics into the same category to form different environmental zones and obtain different environmental zone characteristics; The environmental zone division module obtains the multi-dimensional environmental feature vector F from the data feature extraction module, and the environmental feature vector F will be used as the input data of the clustering algorithm; The adaptive clustering algorithm GMM clustering is selected for cluster analysis, and the optimal GMM model is used to generate the corresponding environmental zone label for each data point Fi; According to the calculated responsibility γ ik , each data point F i Assign it to the Gaussian distribution k that maximizes the responsibility, thereby determining the environmental zone label of each data point, label(F i )=argmax k γ ik ; The clustering results and the environmental zone characteristics are stored in a database, wherein the database also includes the number of each environmental zone and the corresponding environmental characteristic vector; Plant demand module, builds a plant database, combines knowledge graph technology, associates plant characteristics with environmental zone characteristics, and forms an intelligent plant demand map; The environmental zone plant matching module uses the plant demand map to intelligently match plant species suitable for each environmental zone based on the characteristics of the environmental zone, and outputs a recommended planting plan through a deep learning model; The environmental zone visualization module uses data visualization technology to graphically display the environmental zone characteristics and plant matching plans, and generates environmental zone division maps and plant planting recommendation maps.

2. A landscape gardening planting planning and management system based on terrain environment analysis according to claim 1, characterized in that: Data acquisition and processing module: The data acquisition and processing module arranges a sensor group in the target garden area; The sensor set includes, but is not limited to, an altimeter for measuring altitude, a light sensor for measuring the duration and intensity of sunlight, a temperature sensor, a precipitation sensor, a water accumulation detection sensor, a soil moisture sensor, and a pH sensor; The sensor group collects environmental data in real time or at scheduled intervals according to preset time intervals; The sensor group transmits environmental data and stores the environmental data in the database, which adopts structured storage method.

3. The landscape gardening planting planning and management system based on terrain environment analysis according to claim 1 is characterized in that: The data features include altitude data features, light data features, temperature data features, precipitation data features, water accumulation data features, soil moisture data features and soil pH data features; Integrate the data features to generate a multi-dimensional environmental feature vector F.

4. A landscape gardening planting planning and management system based on terrain environment analysis according to claim 3, characterized in that: Altitude data features include mean altitude, altitude standard deviation, maximum altitude, minimum altitude, and altitude range; The characteristics of light data include average daily light duration, maximum light duration, minimum light duration, average light intensity and standard deviation of light intensity; Temperature data features include average temperature, maximum temperature, minimum temperature, temperature fluctuation range, and temperature standard deviation; The characteristics of precipitation data include daily average precipitation, maximum precipitation, precipitation frequency, total precipitation, and precipitation standard deviation; The characteristics of waterlogging data include waterlogging frequency, average waterlogging depth, maximum waterlogging depth, waterlogging duration, and waterlogging location distribution; Soil moisture data characteristics include average soil moisture, maximum soil moisture, minimum soil moisture and soil moisture standard deviation; The soil pH data characteristics include average pH, maximum pH, minimum pH and pH standard deviation.

5. The landscape gardening planting planning and management system based on terrain environment analysis according to claim 1 is characterized in that: Plant Requirements Module: The plant requirements module builds a plant database that contains the growth habits, light requirements, temperature tolerance range, humidity requirements, and water tolerance range of various plants; The plant characteristics of the plant database include plant name, growth habits, light requirements, temperature tolerance range, humidity requirements, waterlogging tolerance range, soil pH adaptation range, and specific requirements; Combined with knowledge graph technology, plant characteristics are associated with environmental characteristics. Knowledge graph is a graph structure that represents entities and their relationships, and is used to store and query complex related information. The specific steps are to define entities and relationships. Entities include plant characteristics and environmental feature vector requirements. Relationships include adaptation and needs. A knowledge graph is created. RDF graph database technology is used to store the relationship between plant characteristics and environmental zone characteristics in the graph database to form an intelligent plant demand graph.

6. The landscape gardening planting planning and management system based on terrain environment analysis according to claim 1 is characterized in that: The environmental zone plant matching module uses the plant demand map to intelligently match plant species suitable for each environmental zone based on the environmental zone characteristics obtained by cluster analysis; Obtain the environmental feature vector F from the environmental zone division module k , query the knowledge graph, find the plant species that match the environmental zone characteristics, use the query language SPARQL of the graph database, according to the environmental feature vector F k Look for plants adapted to these characteristics; Calculate its fitness score, calculate its user preference score, and then calculate the comprehensive score of each plant; According to the number of plant species required in the environmental zone, the rationality scores of different combinations of plant species are calculated by considering the symbiotic relationship, resource competition and ecological balance among the plants in the combination; Generate a score for each plant matching scheme based on the comprehensive score and rationality score, sort them, and form a list. Users can determine the final plant matching scheme based on the scheme score list; A specific planting plan is generated based on the final plant matching scheme, the area of ​​the environmental zone, and the environmental characteristics of the environmental zone. The planting plan includes the planting time and planting density of various plants in the plant matching scheme combination.

7. A landscape gardening planting planning and management system based on terrain environment analysis according to claim 6, characterized in that: Plant Matching Scheme Score: For each plant P j , calculate its fitness score S j , S j =w1γ ij 海拔 +w2γ ij 光照 +w3γ ij 温度 +w4γ ij 湿度 +w5γ ij 积水 +w6γ ij 酸碱度 ; Among them, γ ij * Indicates plant P j The fitness of the characteristic in environment zone i, fitness γ ij The value of can be calculated by the matching degree between the plant characteristics recorded in the plant requirement map and the characteristics of the environmental zone; Calculate its user preference score. Users can provide plant height and maintenance requirements, and calculate the user preference score U of each plant through the knowledge graph j , U j =∑ k=1 m w k u jk , where U j It is a plant j The user preference score, w k is the weight of the kth user demand, u jk It is a plant j Corresponding to the matching degree of the kth requirement, users with higher professional level can also directly assign user preference scores to plants; Calculate the comprehensive score M for each plant j , M j =w j S j +(1-w j ) j ; Users can also provide the desired number of plant species in the environmental zone and query plant species with a comprehensive score exceeding a set threshold; According to the number of plant species required in the environmental area, the rationality score R of different plant species combinations is calculated by considering factors such as the symbiotic relationship, resource competition and ecological balance among the plants in the combination. c , R c =2 / n(n-1)∑ i=1 n-1 ∑ j=i+1 n r ij ; Among them, r ij It is a plant i and P j The rationality score between them, n is the number of plant species in the combination; According to the comprehensive score M j and the rationality score R c , generate each plant matching scheme score T c , and sort them to form a list, T c =αM j +βR c ; Among them, α and β are weight parameters for adjusting the comprehensive score and rationality score.

8. The landscape gardening planting planning and management system based on terrain environment analysis according to claim 1 is characterized in that: Environmental Zone Visualization Module: The environmental zone visualization module obtains data from the environmental zone division module, the plant requirement module and the deep learning model; Use data visualization tools to display the locations and boundaries of different environmental zones based on the environmental zone classification results, and use different colors or line styles to represent different environmental zones; Use polygons to draw each environmental zone, ensure that the boundaries of each environmental zone are clearly visible, and mark the recalculated feature information within the environmental zone; Add text labels or icons at appropriate locations in the environmental area to indicate the numerical value or level of the environmental characteristics; Display the recommended plant species and their matching scores in each environmental zone j , the final solution score T c and detailed planting plan drawings; Provides interactive functions to allow users to operate and query the visualization results.

9. The management method of a landscape gardening planting planning and management system based on terrain environment analysis according to claim 1 is characterized in that: The method comprises the following steps: S1. Collect environmental data and use data cleaning and normalization algorithms to process and standardize the collected data to ensure the accuracy and consistency of the data; S2. Using data feature extraction technology, extract key features from the processed environmental data, and integrate the features to generate a multi-dimensional environmental feature vector; S3, applying an adaptive clustering algorithm to perform cluster analysis on the garden area according to the environmental feature vector, automatically determining the optimal number and boundaries of clusters, clustering areas with similar environmental characteristics into the same category, and forming different environmental zones; S4. Build a comprehensive plant database, combine knowledge graph technology, associate plant characteristics with environmental zone characteristics, and form an intelligent plant demand map; S5. Based on the environmental zone characteristics obtained through cluster analysis, the plant demand map is used in combination with the user's personalized needs to intelligently match plant species suitable for each environmental zone, and a recommended planting plan is output through a deep learning model.

10. A landscape gardening planting planning and management method based on terrain environment analysis according to claim 9, characterized in that: The method further comprises the following steps: S6. Use data visualization technology to graphically display the characteristics of environmental zones and plant matching plans, generate environmental zone division maps and plant planting recommendation maps, and intuitively display the location, characteristics and recommended plant species of different environmental zones.

Images