Landscaping maintenance comprehensive management system and method

Abstract

The invention discloses a landscaping maintenance comprehensive management system and method, and relates to the technical field of intelligent management, and the method comprises the following steps: presetting a greening area, carrying out the first analysis, constructing a maintenance topology, carrying out the second analysis and the third analysis, obtaining a first spatial feature and a first corresponding relation, predicting the maintenance cost, and carrying out the maintenance management. And the third analysis is used for updating the prediction result. According to the invention, based on a cost prediction model of a first spatial feature, annual labor cost prediction errors are controlled in a purchasing link, budget addition in the midway is avoided, through a third analysis real-time closed loop, dynamic pruning or rearrangement is carried out on abnormally high-cost nodes, and topological nodes have carbon sink attributes. According to the method, auditing carbon footprints are generated at the same time in each maintenance operation, a data base is provided for the garden greenbelt to enter CCER transactions, and on the premise that the landscape quality of the greenbelt is not reduced, the obvious effects of cost reduction, efficiency improvement, low carbon emission reduction and fine management are achieved.

Description

Technical Field

[0001] This invention relates to the technical field of intelligent management, and in particular to a comprehensive management system and method for garden greening maintenance. Background Technology

[0002] In recent years, the landscaping and greening maintenance industry has shifted from focusing on reconstruction rather than maintenance to emphasizing both construction and maintenance. The standardization and institutionalization of maintenance management have been continuously improved. Technologies such as the Internet of Things, big data, and cloud computing have been widely applied to maintenance management. Water volume is automatically adjusted based on soil moisture and meteorological data, enabling remote early warning and precise prevention and control of pests and diseases. The concept of sustainable development has taken root, and environmental protection technologies such as water-saving irrigation, organic fertilizers, and biological control have been widely applied. Rainwater harvesting systems, permeable paving, and ecological restoration technologies have also been incorporated into the maintenance system, promoting the improvement of the ecological function of urban green spaces.

[0003] Currently, Chinese invention patent CN114881461A discloses a method and system for managing landscaping and greening. This method generates a greening ratio based on urban building area and urban green area, and compares the greening ratio with a preset greening ratio threshold. If the greening ratio is less than the threshold, a greening base increment is generated based on the threshold, urban building area, and urban green area. This application displays the greening base increment through a display screen or mobile terminal using greening management equipment, allowing users to implement urban greening construction based on the displayed greening base increment, thereby facilitating the management of urban green area and maintaining a dynamic balance between buildings and green vegetation during urban construction. However, related technologies do not simplify complex landscaping data into indicator parameters to construct mapping relationships and quickly obtain management results, which is not conducive to the rapid response of data analysis. They only consider the numerical changes in landscaping area and do not consider the prediction of maintenance costs, which is not conducive to the economic transparency of greening management and has certain limitations. Summary of the Invention

[0004] The technical problem solved by this invention is that related technologies do not simplify complex garden data into indicator parameters to construct mapping relationships and quickly obtain management results, which is not conducive to the rapid response of data analysis. They only consider the numerical changes in garden area and do not consider the prediction of maintenance costs, which is not conducive to the economic transparency of greening management and has certain limitations.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: Firstly, a comprehensive management method for the maintenance of garden greening, comprising the following steps: Step S100: Construct a maintenance topology based on the historical green area data of the preset green areas; Step S200: Perform a second and third analysis on the historical green area data to obtain the first spatial characteristics and the first correspondence. Based on the first spatial characteristics and the first correspondence, predict the maintenance cost and obtain the prediction result. Step S300: Perform a fourth analysis on the influencing factors and actual results, and update the prediction results based on the results of the fourth analysis.

[0006] As a preferred embodiment of the comprehensive management method for garden greening maintenance described in this invention, the method for pre-setting green areas includes: A planar image of the garden is acquired, which is an image obtained by processing an overhead image and a ground image. The overhead image is obtained by a drone at a first altitude, and the pixels of the overhead image cover the entire area of ​​the garden. The ground image is obtained by a 360-degree camera combined with an automatically moving shooting robot. The height of the 360-degree camera is set to a second altitude. The shooting trajectory of the shooting robot is distributed on the curve formed by the midpoints of the main roads and all the paths in the garden, and the starting shooting position is located at the entrance of the garden. The trajectory points where the shooting robot stops each time are set as key points. After reaching the key point, a shooting command is sent to the 360-degree camera. The movement of the shooting robot causes the 360-degree camera to move. After the robot completes its inspection, a first number of ground images are acquired. According to the image processing software, adaptive tree filtering is performed to reduce the transparency of the pixels corresponding to the tree canopy point cloud to a first value, and at the same time, the pixels corresponding to the tree canopy point cloud are covered on top of the pixels of the corresponding ground images. Obtain the names of greenery within the garden area, retrieve the green plant database, input the greenery names into the green plant database, and match the color values ​​corresponding to the greenery names. The color values ​​represent the color values ​​of the greenery in different seasons and different growth cycles. The growth cycle includes the seedling stage, the seedling establishment stage, the rapid growth stage, the near-maturity stage, and the senescence stage. Select the color values ​​of any green plant species for any season, and mark the pixels in the planar image that belong to the color values ​​as first markers. Set the marked areas as green areas. Iterate through the color values ​​of each green plant species in that season to obtain the green areas of each plant species in that season. Number each green area and represent it as X. i Where i is a natural number, when the season changes, the green area will switch the color value of the first marker to the color value of the corresponding plant species in the corresponding season to complete the real-time color value adjustment.

[0007] As a preferred embodiment of the comprehensive management method for garden greening maintenance described in this invention, after the greening areas are set up, any numbered greening area is selected, and the historical greening area data corresponding to the greening area is extracted. The historical greening area data is represented by the historical growth cycle, historical maintenance steps, corresponding historical maintenance dates, corresponding historical maintenance manpower, corresponding historical greening area, and corresponding historical labor cost of each green plant in the greening area. The historical labor cost is represented by the total wages paid to the manpower on the corresponding dates. The historical maintenance steps include leaf pruning, crown pruning, branch pruning, pest control, fertilization, replanting, soil conditioning, irrigation, and root strength testing. Different plant species correspond to different maintenance steps, and the time sequence of the maintenance steps is determined by expert experience.

[0008] As a preferred embodiment of the comprehensive management method for landscaping maintenance described in this invention, the method for performing a first analysis of historical green area data includes: Select any greening species in any season and any green area, and number the historical growth cycles according to their chronological order. The historical growth cycle number is Y. j , where j is a natural number; Select any historical growth cycle with a given number, obtain the sequence of historical maintenance steps corresponding to that historical growth cycle, and number the historical maintenance steps as T. m , where m is a natural number, and by traversing each historical growth cycle, the number of the historical maintenance steps corresponding to each historical growth cycle is obtained. Select the historical maintenance date and the corresponding historical maintenance manpower quantity for any historical maintenance step number, calculate the first sum of the historical maintenance manpower quantity for each historical maintenance date, calculate the average of the first sum, traverse each historical maintenance step for the greening type, and obtain the average of the first sum corresponding to each historical maintenance step. The green area number, the historical growth cycle number, the historical maintenance step number, and the average of the first sum are set as the first analysis result of the season, and a maintenance topology is constructed based on the first analysis result.

[0009] As a preferred embodiment of the comprehensive management method for landscaping maintenance described in this invention, the method for constructing a maintenance topology based on the first analysis result includes: The node ID is set, and the expression for the node ID is: S N P iL =Y j T m ; Among them, S N P iL Let N be the node ID corresponding to the greening of the i-th area in the N-th season, where the values ​​of N for spring, summer, autumn and winter are 1, 2, 3 and 4 respectively; Set the first node attribute, which is the average of the first sum value; Set the edge ID, and the method for setting the edge ID includes: Divide the dates according to the four seasons, set the start date and end date of the season, and in chronological order, set the first node ID between the start date and the end date as the start ID, set the last node ID between the start date and the end date as the end ID, and set the edge IDs from the start ID to the edge IDs. Set an edge attribute, where the edge attribute is the number of days in the season, that is, the number of days corresponding to the difference between the end date and the start date; Using edge IDs as boundaries, node IDs as input data, and a directed graph as output, we obtain the maintenance topology map of each area in the given season. By traversing each season, we obtain the maintenance topology map of each area in each season.

[0010] As a preferred embodiment of the comprehensive management method for garden greening maintenance described in this invention, the method includes: performing a second analysis on historical maintenance dates and historical greening areas to obtain first spatial characteristics, and performing a third analysis on historical greening areas and historical labor costs to obtain a first correspondence. The first spatial feature represents the diffusion characteristics of greening in each area; The first correspondence represents the relationship between green area and labor cost; The analysis method for the first spatial feature includes: Obtain any green area in a season, obtain each historical maintenance date and its corresponding historical green area in the green area, calculate the first time interval between adjacent historical maintenance dates in chronological order, calculate the first difference between the historical green areas corresponding to adjacent historical maintenance dates, calculate the first ratio between the first difference and the first time interval, iterate through each first ratio in the season, calculate the average value of the first ratio, set the average value of the first ratio as the first spatial feature of the area in the season, iterate through each season to obtain the first spatial feature of the area in each season; By traversing various green areas and seasons, the primary spatial characteristics of each area in each season are obtained. The method for constructing the first correspondence includes: Obtain any green area in the season, calculate the first sum of historical labor costs for each green area in the season, obtain the initial historical green area of ​​the green area, and obtain the first spatial characteristics of the green area in the season. With the first sum as the dependent variable and the initial historical green area and the first spatial characteristics as independent variables, construct a mapping equation and set the mapping equation as the first correspondence.

[0011] As a preferred embodiment of the comprehensive management method for landscaping maintenance described in this invention, the calculation expression for the first correspondence is: A = F(B); Where A is the dependent variable, F is the coefficient and constant corresponding to the independent variable in the mapping equation, i.e. the mapping rule, and B is the independent variable.

[0012] As a preferred embodiment of the comprehensive management method for landscaping maintenance described in this invention, the method for predicting maintenance costs based on first spatial characteristics and a first correspondence to obtain the prediction result includes: Select any green area, obtain the predicted date, determine the predicted season corresponding to the predicted date according to the seasonal determination criteria, calculate the second difference between the predicted date and the start date corresponding to the predicted season, calculate the total number of days in the predicted season, calculate the second ratio between the second difference and the total number of days, calculate the first product of the second ratio and the corresponding first spatial feature, input the first product into the first correspondence, and obtain the first sum value corresponding to the first product. Set the first sum as the prediction result, which represents the predicted labor cost for garden maintenance between the start date of the season and the prediction date.

[0013] As a preferred embodiment of the comprehensive management method for garden greening maintenance described in this invention, a fourth analysis is performed on the influencing factors and actual results to obtain the fourth analysis results. The influencing factors include the temperature change rate and the extreme weather rate. The temperature change rate is expressed as the annual average temperature change rate, and the extreme weather rate is expressed as the annual average probability of extreme weather occurrence. The actual results are the predicted results obtained through experiments, which represent the actual labor costs for garden maintenance between the start date of the season and the predicted date. The method for obtaining the third analysis results includes: Calculate the third difference between the actual labor cost and the predicted labor cost. Using the third difference as the dependent variable and the temperature change rate and extreme weather rate as independent variables, perform a multiple regression analysis to obtain a multiple regression equation. Set the multiple regression equation as the fourth analysis result. Methods for updating the prediction results based on the fourth analysis include: Obtain the current temperature change rate and the current extreme weather rate, obtain the third difference through a multiple regression equation, calculate the weighted sum of the third difference and the prediction result to obtain the second sum, and update the second sum as the new prediction result.

[0014] Secondly, a comprehensive management system for garden and green space maintenance includes a construction module, an analysis module, and an update module; The construction module constructs a maintenance topology based on the historical green area data of the preset green areas; The analysis module performs a second and a third analysis on historical green area data to obtain a first spatial feature and a first correspondence. Based on the first spatial feature and the first correspondence, the maintenance cost is predicted to obtain the prediction result. The update module performs a fourth analysis on the influencing factors and actual results, and updates the prediction results based on the results of the fourth analysis.

[0015] The beneficial effects of this invention are as follows: Based on the cost prediction model of the first spatial characteristics, the annual labor cost prediction error is controlled at the procurement stage to avoid mid-term budget increases. Through real-time closed-loop analysis, abnormally high cost nodes are dynamically pruned or rearranged. The topological nodes are attached with carbon sink attributes, so that each maintenance operation generates an auditable carbon footprint at the same time, providing a data foundation for parks and green spaces to enter CCER trading. Without reducing the quality of green space landscape, the invention achieves predictable, controllable and auditable park and green space maintenance costs, with significant effects of cost reduction and efficiency improvement, low carbon emission reduction and refined management. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the basic process of a comprehensive management method for landscaping maintenance provided in one embodiment of the present invention. Detailed Implementation

[0017] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0018] Example, refer to Figure 1 As an embodiment of the present invention, a comprehensive management method for the maintenance of garden greening is provided, comprising the following steps: Step S100: Construct a maintenance topology based on the historical green area data of the preset green areas; Step S200: Perform a second and third analysis on the historical green area data to obtain the first spatial characteristics and the first correspondence. Based on the first spatial characteristics and the first correspondence, predict the maintenance cost and obtain the prediction result. Step S300: Perform a fourth analysis on the influencing factors and actual results, and update the prediction results based on the results of the fourth analysis.

[0019] This invention, based on a cost prediction model with first spatial characteristics, controls the annual labor cost prediction error at the procurement stage, avoiding mid-term budget increases. Through real-time closed-loop analysis, it dynamically prunes or rearranges abnormally high-cost nodes. The topological nodes are associated with carbon sink attributes, enabling each maintenance operation to generate an auditable carbon footprint. This provides a data foundation for green spaces to enter the CCER trading market. Without reducing the quality of the green space landscape, it achieves predictable, controllable, and auditable green space maintenance costs, resulting in significant cost reduction, efficiency improvement, low-carbon emission reduction, and refined management effects.

[0020] The pre-planning methods for green areas include: A planar image of the garden is acquired, which is an image obtained by processing an overhead image and a ground image. The overhead image is obtained by a drone at a first altitude, and the pixels of the overhead image cover the entire area of ​​the garden. The ground image is obtained by a 360-degree camera combined with an automatically moving shooting robot. The height of the 360-degree camera is set to a second altitude. The shooting trajectory of the shooting robot is distributed on the curve formed by the midpoints of the main roads and all the paths in the garden, and the starting shooting position is located at the entrance of the garden. The trajectory points where the shooting robot stops each time are set as key points. After reaching the key point, a shooting command is sent to the 360-degree camera. The movement of the shooting robot causes the 360-degree camera to move. After the robot completes its inspection, a first number of ground images are acquired. According to the image processing software, adaptive tree filtering is performed to reduce the transparency of the pixels corresponding to the tree canopy point cloud to a first value, and at the same time, the pixels corresponding to the tree canopy point cloud are covered on top of the pixels of the corresponding ground images. Obtain the names of greenery within the garden area, retrieve the green plant database, input the greenery names into the green plant database, and match the color values ​​corresponding to the greenery names. The color values ​​represent the color values ​​of the greenery in different seasons and different growth cycles. The growth cycle includes the seedling stage, the seedling establishment stage, the rapid growth stage, the near-maturity stage, and the senescence stage. Select the color values ​​of any green plant species for any season, and mark the pixels in the planar image that belong to the color values ​​as first markers. Set the marked areas as green areas. Iterate through the color values ​​of each green plant species in that season to obtain the green areas of each plant species in that season. Number each green area and represent it as X. i Where i is a natural number, when the season changes, the green area will switch the color value of the first marker to the color value of the corresponding plant species in the corresponding season to complete the real-time color value adjustment.

[0021] In practice, the system achieves canopy transparency for the first time at the image level by fusing aerial and 360° ground-level images and using adaptive tree filtering to forcibly reduce the transparency of tree canopy pixels before overlaying them onto the ground pixels. This eliminates the blind spots of traditional pure aerial photography by removing the canopy from the boundaries of lower garden paths, features, and ground cover. It directly utilizes a four-dimensional color chart from the green plant database (type, season, growth cycle, color value) to perform pixel-level matching on the fused true-color image. Without the need for manual outlining along shrub edges, plants of the same species, condition, and seasonal phase can be marked as closed areas in one go, achieving color-based spot definition and spot-based boundary definition. When the season changes, the system only needs to replace the color chart to re-identify and delineate areas, eliminating the need for re-flying or field re-surveying. This ensures that the boundaries of green areas update seamlessly with natural phenological growth, achieving one-time construction. This method is applicable year-round. The camera robot automatically inspects along the centerline of the road network, with key points and shooting commands being hard-triggered to ensure uniform and complete ground-based shooting angles. The starting point is fixed at the park entrance, forming a repeatable and comparable benchmark image chain. This provides a unified coordinate framework for the evolution analysis of subsequent years. The pre-defined Xi-numbered areas already have attributes such as type, season, area, and center coordinates, which can be directly linked to subsequent maintenance topology nodes without manual coding. This achieves one-click integration of images, vectors, topology, and costs, laying a data foundation for refined budgeting and work assignment. This method is significantly superior to traditional manual mapping or pure aerial photography solutions in five dimensions: occlusion completion, automatic segmentation, seasonal phase refresh, trajectory reuse, and topology pre-positioning. It shifts the pre-setting process of green areas from field-driven to algorithm-driven, achieving a maintenance-free effect of "one flight, year-round management".

[0022] After the green areas are set up, select any numbered green area and extract the historical green area data corresponding to that green area; The historical greening area data is represented by the historical growth cycle, historical maintenance steps, corresponding historical maintenance dates, corresponding historical maintenance manpower, corresponding historical greening area, and corresponding historical labor cost of each green plant in the greening area. The historical labor cost is represented by the total wages paid to the manpower on the corresponding dates. The historical maintenance steps include leaf pruning, crown pruning, branch pruning, pest control, fertilization, replanting, soil conditioning, irrigation, and root strength testing. Different plant species correspond to different maintenance steps, and the time sequence of the maintenance steps is determined by expert experience.

[0023] The methods for the first analysis of historical green space data include: Select any greening species in any season and any green area, and number the historical growth cycles according to their chronological order. The historical growth cycle number is Y. j , where j is a natural number; Select any historical growth cycle with a given number, obtain the sequence of historical maintenance steps corresponding to that historical growth cycle, and number the historical maintenance steps as T. m , where m is a natural number, and by traversing each historical growth cycle, the number of the historical maintenance steps corresponding to each historical growth cycle is obtained. Select the historical maintenance date and the corresponding historical maintenance manpower quantity for any historical maintenance step number, calculate the first sum of the historical maintenance manpower quantity for each historical maintenance date, calculate the average of the first sum, traverse each historical maintenance step for the greening type, and obtain the average of the first sum corresponding to each historical maintenance step. The green area number, the historical growth cycle number, the historical maintenance step number, and the average of the first sum are set as the first analysis result of the season, and a maintenance topology is constructed based on the first analysis result.

[0024] In practice, by numbering the historical growth cycle as Yj in chronological order and the maintenance steps as Tm in sequence, the system automatically establishes a two-layer time-series coordinate system for cycles and steps. This allows any subsequent calculations to be performed on a unified time axis, avoiding sequential errors. The average of the first sum is used as the node weight to dilute outliers in single-time manual labor, ensuring that the topological edge weights reflect long-term stable levels of manpower demand and preventing extreme data from misleading path optimization. The maintenance step sequence determined by expert experience is directly written into the node connection relationship, ensuring that the inviolable operational logic of pruning before crown pruning and pest control after fertilization can be guaranteed without an additional rule engine. This achieves knowledge as structure, with four-dimensional attributes—cycle, step, area, and average—written into the node at once. Subsequent cost prediction, scheduling, and anomaly detection can directly run shortest path, maximum flow, or clustering algorithms on the graph without further format conversion, achieving graph construction as modeling. Each node carries the original cycle number and step number, supporting reverse tracing back to specific historical dates and manual records, forming a complete chain of "computable, interpretable, and auditable" graph structure, providing a continuously reliable foundation for refined maintenance management.

[0025] Methods for constructing maintenance topologies based on the results of the first analysis include: The node ID is set, and the expression for the node ID is: S N P iL =Y j T m ; Among them, S N P iL Let N be the node ID corresponding to the greening of the i-th area in the N-th season, where the values ​​of N for spring, summer, autumn and winter are 1, 2, 3 and 4 respectively; Set the first node attribute, which is the average of the first sum value; Set the edge ID, and the method for setting the edge ID includes: Divide the dates according to the four seasons, set the start date and end date of the season, and in chronological order, set the first node ID between the start date and the end date as the start ID, set the last node ID between the start date and the end date as the end ID, and set the edge IDs from the start ID to the edge IDs. Set an edge attribute, where the edge attribute is the number of days in the season, that is, the number of days corresponding to the difference between the end date and the start date; Using edge IDs as boundaries, node IDs as input data, and a directed graph as output, we obtain the maintenance topology map of each area in the given season. By traversing each season, we obtain the maintenance topology map of each area in each season.

[0026] In practice, node IDs are used to compress the four elements of season, region, cycle, and step into a unique string, giving each node its own spatiotemporal coordinates and avoiding name conflicts. Edge IDs adopt a start ID → end ID structure, directly mapping the order of operations within a season. Edge attributes use the number of days in the season to quantitatively describe the time span, thus transforming the order of pruning branches before crowns, as in expert experience, into computable directed edges. The generated maintenance topology map can support path search, schedule projection, and resource conflict detection without additional manual verification, realizing that time sequence is graph structure and days are edge weights, providing a standardized and reusable graph model foundation for subsequent cost prediction and dynamic scheduling.

[0027] A second analysis of historical maintenance dates and historical green areas yields the first spatial characteristics, and a third analysis of historical green areas and historical labor costs yields the first correspondence. The first spatial feature represents the diffusion characteristics of greening in each area; The first correspondence represents the relationship between green area and labor cost; The analysis method for the first spatial feature includes: Obtain any green area in a season, obtain each historical maintenance date and its corresponding historical green area in the green area, calculate the first time interval between adjacent historical maintenance dates in chronological order, calculate the first difference between the historical green areas corresponding to adjacent historical maintenance dates, calculate the first ratio between the first difference and the first time interval, iterate through each first ratio in the season, calculate the average value of the first ratio, set the average value of the first ratio as the first spatial feature of the area in the season, iterate through each season to obtain the first spatial feature of the area in each season; By traversing various green areas and seasons, the primary spatial characteristics of each area in each season are obtained. The method for constructing the first correspondence includes: Obtain any green area in the season, calculate the first sum of historical labor costs for each green area in the season, obtain the initial historical green area of ​​the green area, and obtain the first spatial characteristics of the green area in the season. With the first sum as the dependent variable and the initial historical green area and the first spatial characteristics as independent variables, construct a mapping equation and set the mapping equation as the first correspondence.

[0028] The calculation expression for the first correspondence is: A = F(B); Where A is the dependent variable, F is the coefficient and constant corresponding to the independent variable in the mapping equation, i.e. the mapping rule, and B is the independent variable.

[0029] In practice, by calculating the ratio of the difference in green area between adjacent maintenance dates to the time interval and averaging the results, the first spatial feature quantitatively characterizes the speed of green space diffusion, making the "speed of green growth" a comparable numerical indicator. Then, with the first sum of labor costs as the dependent variable and the initial area and the first spatial feature as independent variables, a mapping equation is constructed. For the first time, diffusion speed is incorporated into the cost model, achieving a precise correspondence that different diffusion speeds result in different costs for the same area. This equation can be directly used for budget projection of new areas without the need to re-collect historical labor records, achieving the effect of spatial features as cost factors, and providing a calculable mathematical basis for subsequent dynamic pricing and resource allocation.

[0030] Methods for predicting maintenance costs based on first spatial features and first correspondence include: Select any green area, obtain the predicted date, determine the predicted season corresponding to the predicted date according to the seasonal determination criteria, calculate the second difference between the predicted date and the start date corresponding to the predicted season, calculate the total number of days in the predicted season, calculate the second ratio between the second difference and the total number of days, calculate the first product of the second ratio and the corresponding first spatial feature, input the first product into the first correspondence, and obtain the first sum value corresponding to the first product. Set the first sum as the prediction result, which represents the predicted labor cost for garden maintenance between the start date of the season and the prediction date.

[0031] A fourth analysis is conducted on the influencing factors and actual results to obtain the fourth analysis results. The influencing factors include the temperature change rate and the extreme weather rate. The temperature change rate is expressed as the annual average temperature change rate, and the extreme weather rate is expressed as the annual average probability of extreme weather events. The actual results are the predicted results obtained through experiments, which represent the actual labor costs for garden maintenance between the start date of the season and the predicted date. The method for obtaining the third analysis results includes: Calculate the third difference between the actual labor cost and the predicted labor cost. Using the third difference as the dependent variable and the temperature change rate and extreme weather rate as independent variables, perform a multiple regression analysis to obtain a multiple regression equation. Set the multiple regression equation as the fourth analysis result. Methods for updating the prediction results based on the fourth analysis include: Obtain the current temperature change rate and the current extreme weather rate, obtain the third difference through a multiple regression equation, calculate the weighted sum of the third difference and the prediction result to obtain the second sum, and update the second sum as the new prediction result.

[0032] In practice, the difference between the predicted date and the start of the season is mapped to the ratio of progress within the season, and multiplied with the first spatial feature. This combines the time progress with the speed of green space expansion, allowing any date to be converted into an "equivalent area growth," providing dynamic input for the first correspondence and realizing "date equals area equivalent." The first product after process conversion is directly substituted into the first correspondence to output the cumulative labor cost before the predicted date, without waiting for the end of the season to perform statistics. This achieves the instantaneous cost extrapolation capability of "input on the same day, output on the same day." A multiple regression equation is constructed with temperature change rate and extreme weather rate as independent variables and the difference between actual cost and predicted cost as the dependent variable, quantifying macro-climate disturbances into a quantifiable value. The addition and subtraction of correction terms give the forecast results climate flexibility, avoiding systematic deviations caused by "using the model indefinitely". The third difference is calculated in real time using the current temperature change rate and extreme weather rate, and then weighted with the original forecast result to synthesize the second sum value, which is refreshed online. This ensures that the forecast value evolves in sync with climate intelligence, realizing a closed-loop operation of "forecast-measurement-correction". It keeps the cost estimate and actual consumption in the same trend and direction in the long term. The new forecast results can be directly used for next month's funding plan, staffing schedule and material procurement. Management does not need to wait until the end of the season to review, realizing refined cost control of "knowing beforehand, adjusting during the process and verifying afterward". It provides a complete technical path for the transformation of landscaping maintenance from experience-based allocation to algorithm-driven.

[0033] This invention, based on a cost prediction model with first spatial characteristics, controls the annual labor cost prediction error at the procurement stage, avoiding mid-term budget increases. Through real-time closed-loop analysis, it dynamically prunes or rearranges abnormally high-cost nodes. The topological nodes are associated with carbon sink attributes, enabling each maintenance operation to generate an auditable carbon footprint. This provides a data foundation for green spaces to enter the CCER trading market. Without reducing the quality of the green space landscape, it achieves predictable, controllable, and auditable green space maintenance costs, resulting in significant cost reduction, efficiency improvement, low-carbon emission reduction, and refined management effects.

[0034] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0035] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A comprehensive management method for landscaping maintenance, characterized in that, The method comprises the following steps: Step S100, constructing a maintenance topology according to preset historical greening plot data of the greening plot, the first analysis being used to obtain a greening plot number of any season, a historical growth cycle number, a historical maintenance step number, and an average value of a first sum value; Step S200, performing second and third analyses on the historical greening plot data to obtain a first spatial feature and a first corresponding relationship, and predicting a maintenance cost according to the first spatial feature and the first corresponding relationship to obtain a prediction result; Step S300, performing a fourth analysis on influencing factors and actual results, and updating the prediction result according to a fourth analysis result.

2. The method of claim 1, wherein the method further comprises: The preset method of the greening plot comprises: ​ A plan image of the garden is obtained, the plan image of the garden being an image obtained by processing a bird's-eye image and a ground image, the bird's-eye image being obtained by a drone at a first height, and pixels of the bird's-eye image covering the entire range of the garden, the ground image being obtained by a 360-degree camera combined with a shooting robot automatically moving to shoot, the 360-degree camera being arranged at a second height, a shooting track of the shooting robot being distributed on a curve formed by midpoints of each large road and each small road of the garden and a starting shooting position being located at a garden entrance, each stop track point of the shooting robot being set as a key point, and a shooting instruction being sent to the 360-degree camera after reaching the key point, the shooting robot moving to drive the 360-degree camera to move, a first number of ground images being obtained after the robot completes inspection, and adaptive tree filtering being performed according to an image processing software, a pixel transparency of a tree crown point cloud being reduced to a first value, and the pixel transparency of the tree crown point cloud being overlaid on a corresponding ground image pixel. A greening name in the garden range is obtained, a green plant database is called, the greening name is input into the green plant database, a color value corresponding to the greening name is matched, the color value being represented as color values of different seasons and different growth periods of the greening, and the growth period comprising a seedling period, a slow-growing period, a fast-growing period, a near-mature period and a senescence period. Select any green plant species in any season of each color value, mark the pixels belonging to the color value in the plan image first, set the region after the first mark as the green patch area, traverse each green plant species in the season of each color value, get the green patch area of each green plant species in the season, number each green patch area, and the numbered green patch area is represented as X i Wherein i is a natural number, when the season changes, the green patch area respectively switches the first marked color value to the color value of the corresponding season of the plant species, and completes the real-time color value adjustment.

3. The method of claim 2, wherein the method further comprises: determining a location of the mobile device; and determining a location of the plant; and determining a distance between the location of the mobile device and the location of the plant. After the greening plot is set, a greening plot numbered is selected, and historical greening plot data corresponding to the greening plot is extracted. The historical greening plot data is represented as historical growth periods, historical maintenance steps, corresponding historical maintenance dates, corresponding historical labor quantities, corresponding historical greening areas and corresponding historical labor costs of each green plant in the greening plot, and the historical labor cost being represented as a total amount of wages paid to the corresponding date labor. The historical maintenance steps comprise leaf pruning, crown pruning, branch pruning, insect prevention, fertilizer supplement, plant supplement, soil adjustment, irrigation and root strength testing, wherein different green plant species correspond to different maintenance steps, and a time sequence of the maintenance steps is obtained by expert experience.

4. The method of claim 3, wherein the method further comprises: The method for performing the first analysis on the historical greening plot data comprises: ​ Select any season, any green area of green species, according to the time sequence of historical growth cycle, the historical growth cycle is numbered, the numbering of the historical growth cycle is Y j wherein j is a natural number; Select any numbered historical growth cycle, obtain the sequence of historical maintenance steps corresponding to the historical growth cycle, number the historical maintenance steps, and the number of historical maintenance steps is T m wherein m is a natural number, and each historical growth cycle is obtained by traversing the historical maintenance steps corresponding to each historical growth cycle; Select the historical maintenance date corresponding to any numbered historical maintenance step and the corresponding historical maintenance labor quantity, calculate the first sum of historical maintenance labor quantities at each historical maintenance date, calculate the average of the first sum, traverse each historical maintenance step of the green area, and obtain the average of the first sum corresponding to each historical maintenance step; Set the green area number, the number of the historical growth period, the number of the historical maintenance step, and the average of the first sum as the first analysis result of the season, and construct the maintenance topology according to the first analysis result.

5. The method of claim 4, wherein the method further comprises: The method for constructing the maintenance topology according to the first analysis result comprises: ​ Set the node ID, and the expression of the node ID is: S N P iL =Y j T m ; S N P iL is the node ID corresponding to the greening of the ith region of the Nth season, wherein the values of N corresponding to spring, summer, autumn and winter are 1, 2, 3 and 4, respectively. Set the first node attribute, which is the average of the first sum; Set the edge ID, and the setting method of the edge ID comprises: According to the division of the four seasons, set the start date and the end date of the season, set the first node ID between the start date and the end date as the start ID, set the last node ID between the start date and the end date as the end ID, and set the edge ID as the start ID to the edge ID; Set the edge attribute, which is the number of days of the season, that is, the number of days corresponding to the difference between the end date and the start date; Take the edge ID as the boundary and the node ID as the input data, and obtain the maintenance topology graph of each area in the season by taking the directed graph as the output, and traverse each season to obtain the maintenance topology graph of each area in each season.

6. The method of claim 1, wherein the method further comprises: Secondly analyze the historical maintenance date and the historical green area to obtain a first spatial feature, and thirdly analyze the historical green area and the historical labor cost to obtain a first corresponding relationship; ​ The first spatial feature represents the diffusion feature of the green area of each area; The first corresponding relationship represents the corresponding relationship between the green area and the labor cost; The analysis method of the first spatial feature comprises: Obtain any green area in a season, obtain each historical maintenance date and the corresponding historical green area in the green area, calculate the first time interval between adjacent historical maintenance dates in time sequence, calculate the first difference value of the historical green area corresponding to adjacent historical maintenance dates, calculate the first ratio of the first difference value to the first time interval, calculate the average of the first ratio, set the average of the first ratio as the first spatial feature of the area in the season, and traverse each season to obtain the first spatial feature of the area in each season; Traverse each green area and each season to obtain the first spatial feature of each area in each season; The construction method of the first corresponding relationship comprises: Obtain any green area in the season, calculate the first sum of each historical labor cost of the green area in the season, obtain the initial historical green area of the green area, and obtain the first spatial feature of the green area in the season, construct a mapping equation by taking the first sum as the dependent variable and taking the initial historical green area and the first spatial feature as the independent variable, and set the mapping equation as the first corresponding relationship.

7. The method of claim 6, wherein the method further comprises: The calculation expression of the first corresponding relationship is: ​ A=F(B); Wherein, A is the dependent variable, F is the coefficient and constant corresponding to the independent variable of the mapping equation, that is, the mapping rule, and B is the independent variable.

8. The method of claim 7, wherein the method further comprises: The method for predicting the maintenance cost according to the first spatial feature and the first corresponding relationship includes: ​ Selecting any green area, obtaining a prediction date, determining the prediction season corresponding to the prediction date according to the season determination standard, calculating a second difference value between the prediction date and the starting date corresponding to the prediction season, calculating the total number of days of the prediction season, calculating a second ratio value of the second difference value and the total number of days, calculating a first product of the second ratio value and the corresponding first spatial feature, inputting the first product into the first corresponding relationship to obtain a first sum value corresponding to the first product; Setting the first sum value as the prediction result, and the prediction result represents the predicted labor cost for maintaining the garden between the starting date of the season and the prediction date.

9. The method of claim 1, wherein the method further comprises: The fourth analysis result is obtained by performing a fourth analysis on the influencing factors and the actual result, the influencing factors include a temperature change rate and an extreme weather rate, the temperature change rate is represented as an annual average temperature change rate, the extreme weather rate is represented as an annual average probability of extreme weather occurrence, and the actual result is represented as an actual labor cost for maintaining the garden between the starting date of the season and the prediction date. ​ The method for obtaining the third analysis result includes: Calculating a third difference value between the actual labor cost and the predicted labor cost, performing multiple regression analysis on the third difference value as the dependent variable and the temperature change rate and the extreme weather rate as the independent variables, obtaining a multiple regression equation, and setting the multiple regression equation as the fourth analysis result; The method for updating the prediction result according to the fourth analysis result includes:

10. A comprehensive management system for garden maintenance, which is used to execute the comprehensive management method for garden maintenance according to claim 1, characterized in that, Obtaining the current temperature change rate and the current extreme weather rate, obtaining the third difference value through the multiple regression equation, performing weighted calculation on the third difference value and the prediction result to obtain a second sum value, and updating the second sum value as a new prediction result. The method includes a construction module, an analysis module and an updating module; The construction module constructs a maintenance topology according to the historical green area data of the preset green area; The analysis module performs a second analysis and a third analysis on the historical green area data to obtain the first spatial feature and the first corresponding relationship, and predicts the maintenance cost according to the first spatial feature and the first corresponding relationship to obtain the prediction result; The updating module performs a fourth analysis on the influencing factors and the actual result, and updates the prediction result according to the fourth analysis result.

Images