Digital intelligent management system and method for landscaping maintenance

Abstract

The invention discloses a digital intelligent management system and method for landscaping maintenance, and relates to the technical field of data processing, and the system comprises a data sensing module which is used for obtaining M pieces of garden region information and obtaining M pieces of garden state data streams; the parallel analysis module is used for performing parallel analysis on the M garden state data streams and outputting M garden growth analysis results; the strategy analysis module is used for acquiring a landscaping maintenance strategy library and determining M landscaping maintenance strategy parameters; and the digital intelligent management and control module is used for matching and activating to obtain a garden linkage maintenance equipment set and executing remote management and control of greening maintenance. The technical problems of resource waste and management and protection lag caused by extensive region division and poor maintenance strategy linkage in traditional garden management in the prior art are solved, and the technical effects of realizing precise intelligent management and control of garden maintenance by using big data acquisition and processing and improving the garden greening maintenance efficiency and the pest control real-time performance are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a digital intelligent management system and method for landscaping maintenance. Background Art

[0002] With the acceleration of the urbanization process, the contradiction between the expansion of landscaping scale and the demand for refined ecological management has become increasingly prominent. Traditional garden maintenance mainly relies on manual inspections and experience-based decision-making, resulting in problems such as vague area division, lagging pest and disease early warnings, and waste of water and fertilizer resources. In recent years, although technologies such as Internet of Things sensors and drone inspections have been introduced, existing solutions mostly adopt single data sources or fixed threshold analysis models, leading to poor regional adaptability and fragmented processing of multi-dimensional environmental parameters (such as soil moisture, vegetation growth, and pest and disease characteristics), making it difficult to achieve dynamic and precise maintenance. For example, extensive zoning easily causes uneven irrigation coverage, and the identification of pests and diseases depends on manual photo upload, resulting in response delays, while the maintenance strategy is disjointed from the equipment execution link, further reducing the resource utilization efficiency.

[0003] The prior art has technical problems of resource waste and management lag caused by extensive area division and poor linkage of maintenance strategies in traditional garden management. Summary of the Invention

[0004] This application provides a digital intelligent management system and method for landscaping maintenance, aiming to solve the technical problems of resource waste and management lag caused by extensive area division and poor linkage of maintenance strategies in traditional garden management in the prior art.

[0005] In view of the above problems, this application provides a digital intelligent management system and method for landscaping maintenance.

[0006] In the first aspect of this application, a digital intelligent management system for landscaping maintenance is provided. The system includes:

[0007] A data perception module, which is used to perform regional division coding on a target garden to obtain M pieces of garden area information, and deploy a perception network to collect and obtain M pieces of garden state data streams of the M pieces of garden area information; a parallel analysis module, which is used to construct a dual-channel garden analysis, the dual-channel garden analysis includes a landscaping prediction channel and a pest identification channel, and uses the dual-channel garden analysis to perform parallel analysis on the M pieces of garden state data streams, and outputs M pieces of garden growth analysis results; a strategy parsing module, which is used to obtain a landscaping maintenance strategy library, and use the landscaping maintenance strategy library to perform strategy parsing and associated correction on the M pieces of garden growth analysis results to determine M pieces of landscaping maintenance strategy parameters; a digital control module, which is used to match and activate based on the M pieces of landscaping maintenance strategy parameters and a list of garden maintenance equipment to obtain a set of garden-linked maintenance equipment, and perform remote control of greening maintenance on the target garden through the set of garden-linked maintenance equipment.

[0008] In the second aspect of the present application, a digital management method for landscaping maintenance is provided, and the method includes:

[0009] Perform regional division coding on a target garden to obtain M pieces of garden area information, deploy a perception network to collect and obtain M pieces of garden state data streams of the M pieces of garden area information; construct a dual-channel garden analysis, the dual-channel garden analysis includes a landscaping prediction channel and a pest identification channel, and use the dual-channel garden analysis to perform parallel analysis on the M pieces of garden state data streams, and output M pieces of garden growth analysis results; obtain a landscaping maintenance strategy library, and use the landscaping maintenance strategy library to perform strategy parsing and associated correction on the M pieces of garden growth analysis results to determine M pieces of landscaping maintenance strategy parameters; match and activate based on the M pieces of landscaping maintenance strategy parameters and a list of garden maintenance equipment to obtain a set of garden-linked maintenance equipment, and perform remote control of greening maintenance on the target garden through the set of garden-linked maintenance equipment.

[0010] One or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0011] A data perception module is used to divide and encode the target garden into regions to obtain M pieces of garden area information, and collect M garden status data streams of the M pieces of garden area information; a parallel analysis module is used to construct a dual-channel for garden analysis, perform parallel analysis, and output M garden growth analysis results; a strategy parsing module is used to obtain a landscaping maintenance strategy library, perform strategy parsing and correlation correction on the M garden growth analysis results, and determine M landscaping maintenance strategy parameters; a digital intelligent control module is used to match and activate based on the M landscaping maintenance strategy parameters and a list of garden maintenance equipment to obtain a set of garden linkage maintenance equipment, and perform remote control of greening maintenance on the target garden. It achieves the technical effect of realizing precise intelligent control of garden maintenance by using big data collection and processing, and improving the efficiency of landscaping maintenance and the real-time performance of pest control. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for description in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0013] Figure 1 FIG. is a schematic structural diagram of a digital intelligent management system for landscaping maintenance provided by an embodiment of the present application;

[0014] Figure 2 FIG. is a schematic flow diagram of a digital intelligent management method for landscaping maintenance provided by an embodiment of the present application.

[0015] Description of reference numerals: data perception module 10, parallel analysis module 20, strategy parsing module 30, digital intelligent control module 40. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] The present application provides a digital intelligent management system and method for landscaping maintenance, aiming to solve the technical problems of resource waste and management lag caused by extensive regional division and poor linkage of maintenance strategies in traditional garden management in the prior art.

[0017] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present application belong to the scope of protection of the present application.

[0018] Embodiment 1, as Figure 1As shown, the present application provides a digital management system for landscaping maintenance, the system comprising:

[0019] The data perception module 10 is used to perform regional division and encoding on the target garden to obtain M garden area information, and the perception network is deployed to collect and obtain M garden status data streams of the M garden area information.

[0020] Specifically, firstly, the target garden is divided into comprehensive and detailed regions based on various factors such as regional functions, geographical features and regional area. These division factors are of different importance and they are prioritized. Then, combined with the design and survey data of the target garden, a multi-level regional division is carried out, and finally M different garden areas are obtained. In order to facilitate management and data correspondence, a special coding system is designed for each area, including coding content, sequence and identification, to complete the coding and labeling of M garden areas, thereby generating M garden area information. Next, the perception network is deployed in a targeted manner according to these regional information, and various sensors and monitoring equipment are used to collect real-time data of each area, forming M garden status data streams reflecting the growth status of garden plants, soil environment, meteorological conditions and other conditions, providing strong data support for subsequent precise garden maintenance analysis and decision-making.

[0021] The parallel analysis module 20 is used to construct a garden analysis dual channel, which includes a garden greening prediction channel and a pest and disease identification channel. The garden analysis dual channel is used to perform parallel analysis on the M garden status data streams to output M garden growth analysis results.

[0022] Specifically, the parallel analysis module 20 constructs an efficient dual-channel system for garden analysis, which consists of a garden greening prediction channel and a pest and disease identification channel. During the construction process, a large number of garden greening historical data sets and garden pest and disease historical data sets are first collected, and the corresponding key indicators are extracted by in-depth analysis of the garden greening prediction target and the pest and disease identification target, and the garden greening prediction indicator set and the pest and disease identification indicator set are formed respectively. These indicator sets are used to classify and identify the corresponding historical data sets, so as to obtain the garden greening prediction channel and the pest and disease identification channel, and then the two are integrated in parallel. When receiving M garden status data streams from the data perception module, the dual channels simultaneously perform rapid and independent analysis on these data streams, starting from multiple dimensions such as plant growth trend prediction and early diagnosis of pests and diseases, and output M accurate garden growth analysis results, providing a scientific basis for subsequent maintenance decisions.

[0023] The strategy analysis module 30 is used to obtain a garden greening maintenance strategy library, use the garden greening maintenance strategy library to perform strategy analysis and correlation correction on the M garden growth analysis results, and determine M garden greening maintenance strategy parameters.

[0024] Specifically, the policy analysis module 30 first accesses and obtains the pre-constructed landscaping maintenance policy library, which is a knowledge collection formed based on a large amount of gardening maintenance experience, scientific research, and industry standards. It covers a rich variety of maintenance policies, with corresponding coping strategies for different garden growth conditions, plant variety characteristics, seasonal changes, etc. Using the M garden growth analysis results output by the parallel analysis module as the input basis, for the growth analysis results of each garden area, a comprehensive and detailed search is carried out in the policy library to screen out the maintenance policies that match the actual situation of this area. In this process, it comprehensively considers various factors such as the growth stage of plants, the type and severity of pests and diseases, soil fertility status, and climate conditions. After screening out the initially applicable policies, since each area in the garden is not completely independent, there may be ecological associations, environmental impacts, and other interactions among them, and the actual maintenance scenarios are complex and changeable. A single policy may not fully meet the requirements. Combining the characteristics of each area and the relationships between them, the initially screened policies are adjusted and optimized, weighing the priorities and synergies between different policies. After the above process of policy analysis and correlation correction, M landscaping maintenance policy parameters are finally determined. These parameters precisely specify the detailed standards and implementation plans for specific maintenance operations such as watering, fertilizing, pruning, and pest control in each garden area, providing key decision-making support for the subsequent digital intelligent control module to drive garden maintenance equipment for precise maintenance operations, and ensuring the scientific and efficient advancement of the entire landscaping maintenance work.

[0025] The digital intelligent control module 40 is used to match and activate based on the M landscaping maintenance policy parameters and the list of garden maintenance equipment to obtain a set of garden linkage maintenance equipment, and perform remote control of greening maintenance on the target garden through the set of garden linkage maintenance equipment.

[0026] Specifically, the digital intelligent control module 40 first receives the determined M landscaping maintenance policy parameters, which specify in detail the maintenance operations and standards required for each garden area. At the same time, it obtains the list of garden maintenance equipment, which covers the detailed information of various equipment such as irrigation equipment, pruning machinery, and pest control devices.

[0027] The matching and activation work begins. The maintenance policy parameters of each garden area are precisely compared with the attributes such as the equipment functions and applicable scopes in the equipment list. For example, for areas that require a large amount of watering, irrigation equipment with a flow rate and coverage range that meet the requirements is screened out; for areas with pests and diseases, corresponding pest control equipment is matched. Through this precise matching, a series of applicable equipment is activated, thus forming a set of garden linkage maintenance equipment.

[0028] Finally, with the help of the Internet of Things and automation control technologies, remote management and control of greening maintenance for the target garden is carried out through a set of garden linkage maintenance equipment. During the remote management and control process, precise control instructions are sent to each device according to the maintenance strategy parameters, such as controlling the opening time, watering duration, and water volume of irrigation equipment, commanding pruning equipment to prune according to specific shapes and sizes, and setting the chemical spraying dose and frequency of pest control equipment. In this way, efficient and precise maintenance of the entire target garden is achieved, ensuring the healthy growth of garden plants and enhancing the garden landscape effect.

[0029] In a possible implementation manner, the data perception module 10 further includes:

[0030] A garden area division factor acquisition unit, configured to acquire garden area division factors, where the garden area division factors include area functions, geographical features, and area sizes.

[0031] A garden division factor priority sequence acquisition unit, configured to perform priority sorting on the garden area division factors according to the garden area division factors to obtain a garden division factor priority sequence.

[0032] A garden division area acquisition unit, configured to perform multi-level division of the design and survey data of the target garden based on the garden division factor priority sequence to obtain M garden division areas.

[0033] A regional coding system design unit, configured to design a regional coding system according to the garden division factor priority sequence, where the regional coding system includes coding content, coding order, and coding identifiers.

[0034] A garden area information acquisition unit, configured to perform coding and marking on the M garden division areas according to the regional coding system to obtain the M garden area information.

[0035] Specifically, it is crucial to obtain comprehensive and accurate garden area division factors. Among them, the area function is an important basis for division, and different functional areas have different planning and maintenance focuses; geographical features cover terrain and landforms. For example, the drainage and irrigation methods of mountains and plains are different, affecting the growth environment of plants; soil types also determine the types of plants suitable for planting. Acidic soil is suitable for certain plants, while alkaline soil is suitable for others. The area size is related to the input scale of maintenance resources. Larger areas require more manpower, material resources, and time for management, and area factors need to be considered when planning irrigation systems and pest control ranges. By integrating these area function, geographical feature, and area size factors, targeted maintenance strategies are formulated for subsequent reasonable division of the garden area.

[0036] After obtaining the factors for garden area division, prioritize these factors to obtain the priority sequence of garden division factors. First, conduct in-depth research on each factor and consider its impact on the overall garden planning, plant growth, and maintenance management. For example, the regional function determines the core use of the garden. If it is an ornamental area, the selection and layout of plants will revolve around aesthetics; geographical features fundamentally limit the plant growth conditions, and different terrains and soil types will affect the survival and growth status of plants; the regional area affects the scale and efficiency of resource allocation. Through the comprehensive evaluation of these factors, use the analytic hierarchy process to determine the weight of each factor. Based on the weight, arrange the garden area division factors in an orderly manner to form a clear priority sequence.

[0037] After determining the priority sequence of garden division factors, conduct the actual division of the target garden area. Guided by this priority sequence, start processing the design survey data of the target garden. The design survey data contains rich information such as the topography, plant distribution, and infrastructure layout of the garden. According to the factor with the highest priority, such as the regional function, initially divide the target garden into large areas of different functional types, such as leisure areas, planting areas, and landscape areas. Then, based on the geographical feature factor ranked second, further subdivide within the already divided large areas. For example, in the planting area, divide it into sub-areas suitable for different plant growth according to different soil types and terrain undulations. Finally, combined with the regional area factor, reasonably adjust and optimize these sub-areas to ensure that each area meets the actual maintenance and management requirements in terms of area. After such a multi-level division process, finally obtain M clearly defined, functionally diverse garden division areas that conform to the actual situation of the garden, providing strong support for the subsequent precise management and maintenance of each area.

[0038] According to the determined priority sequence of garden division factors, design a regional coding system. In the design of the coding content, fully combine factors such as the regional functions and geographical features with higher priorities, and transform them into coding information with practical significance. For example, use numbers or letters to represent different regional functions, such as "A" representing the viewing area and "B" representing the leisure area; use specific code combinations to represent geographical features, such as "01" representing mountains and "02" representing plains. The coding order is determined according to the priorities of the garden division factors. First, reflect the coding related to regional functions, then the geographical feature coding, and finally integrate the information related to the regional area, ensuring that the coding order is logically clear and convenient for identification and interpretation. The coding identifier, as a special symbol or mark to distinguish different regional codes, is set at a specific position of the code. For example, add a unique symbol at the beginning or end of the code to quickly identify the coding type and the category it belongs to, enhancing the recognition and management efficiency of the code. By designing the coding content, reasonably arranging the coding order, and clarifying the coding identifier, a complete and practical regional coding system is constructed, laying a solid foundation for the precise management and data docking of garden areas.

[0039] After the design of the regional coding system is completed, it is necessary to code and label the previously divided M garden division areas to obtain M garden area information. According to the established regional coding system, assign the corresponding code to each garden division area one by one. For each area, operate according to the rules of coding content, coding order, and coding identifier. For example, a certain area belongs to the viewing area (corresponding to the symbol representing the viewing area in the coding content), is located in the plain (matching the code of the plain in the geographical feature coding), and the area is within a specific range (corresponding to the coding part related to the regional area). Combine these information according to the coding order and add a unique coding identifier to form a unique code for this area. In this way, code and label the M garden division areas in turn, and each area has a specific code. These codes and the regional attribute information they represent together constitute M garden area information. These information become the basic data of the entire garden digital management system.

[0040] In a possible implementation manner, the garden division factor priority sequence acquisition unit further includes:

[0041] A garden division factor maintenance impact data set acquisition unit, which is used to mine maintenance impact data based on the garden area division factors to obtain a garden division factor maintenance impact data set.

[0042] A garden division factor principal component information acquisition unit, which is used to perform standardization processing and principal component selection on the garden division factor maintenance impact data set to obtain garden division factor principal component information.

[0043] A garden division factor weight factor determination unit, configured to perform PCA weight allocation on each division factor in the garden area division factors based on the principal component information of the garden division factors, and determine the garden division factor weight factors.

[0044] A priority ranking unit, configured to rank the garden area division factors according to the garden division factor weight factors, and obtain the garden division factor priority sequence.

[0045] Specifically, based on the garden area division factors, maintenance impact data mining is performed, and a large amount of maintenance data related to division factors such as regional functions, geographical features, and regional areas is collected. This data covers the actual effects of different gardens under various maintenance measures, such as the growth conditions of plants in different functional areas, the occurrence frequencies of pests and diseases in different geographical feature areas, and the consumption of maintenance resources in different area regions. By using data mining techniques, maintenance impact data closely related to the garden division factors is extracted from the massive data, and then the garden division factor maintenance impact data set is obtained.

[0046] This data set is subjected to standardization processing and principal component selection. The standardization processing can eliminate the influence brought about by different dimensions and value ranges between different data, making the data comparable. On this basis, using the principal component analysis method, the most critical principal components for garden maintenance are selected from the numerous impact data, thereby obtaining the principal component information of the garden division factors. These principal component information condense the main characteristics of the original data and can more concisely and effectively reflect the relationship between each division factor and garden maintenance.

[0047] Based on the principal component information of the garden division factors, the principal component analysis (PCA) algorithm is used to perform weight allocation on each division factor in the garden area division factors. The PCA algorithm determines the contribution degree of each principal component to the original division factor by calculating the correlation between the principal components and the original division factors, and then obtains the garden division factor weight factors. These weight factors intuitively reflect the relative importance of each division factor to garden maintenance.

[0048] The garden area division factors are ranked according to the garden division factor weight factors. The larger the weight factor, the higher the importance of the corresponding division factor in garden maintenance, and its priority is also higher. After such ranking, the garden division factor priority sequence is finally obtained. This sequence provides an important basis for scientifically and reasonably dividing the garden area and formulating targeted maintenance strategies in the follow-up, and helps to improve the efficiency and quality of garden maintenance work.

[0049] In a possible implementation manner, the parallel analysis module 20 further includes:

[0050] A historical dataset collection unit for collecting and obtaining a historical dataset of landscaping and a historical dataset of garden pests and diseases.

[0051] An index extraction unit for obtaining landscaping prediction targets and pest and disease identification targets, extracting indexes for the landscaping prediction targets and pest and disease identification targets, and obtaining a landscaping prediction index set and a pest and disease identification index set.

[0052] A classification identification training unit for respectively performing classification identification training on the historical dataset of landscaping and the historical dataset of garden pests and diseases by using the landscaping prediction index set and the pest and disease identification index set, and obtaining a landscaping prediction channel and a pest and disease identification channel.

[0053] A garden analysis dual-channel construction unit for integrating the landscaping prediction channel and the pest and disease identification channel in parallel to form and construct the garden analysis dual-channel.

[0054] Specifically, widely collect and obtain a historical dataset of landscaping and a historical dataset of garden pests and diseases. These datasets have a wide range of sources, covering the growth data of garden plants, maintenance measures and effect records under different seasons, different years and different climatic conditions, as well as detailed information such as the types, occurrence times and damage degrees of pests and diseases.

[0055] Clarify the landscaping prediction targets, including predicting the growth rate of plants, the time of flowering and fruiting, the health status of plants, etc.; at the same time, determine the pest and disease identification targets, such as identifying the types of pests and diseases, judging the infection range and severity of pests and diseases, etc. After clarifying the targets, start from multiple aspects to extract indexes. For landscaping prediction, collect the physiological data of plants, such as the height of plants, the number of leaves, the chlorophyll content, etc.; environmental data, such as temperature, humidity, light duration and intensity, etc.; and maintenance data, including watering frequency, fertilizer types and dosages, etc. These data comprehensively constitute the landscaping prediction index set. For the pest and disease identification targets, the extracted indexes cover the morphological characteristics of pests and diseases, such as the body shape, color and markings of pests, the colony morphology of pathogens, etc.; the symptom manifestations of plants being invaded, such as the discoloration, perforation and withering of leaves, whether the plants are deformed or not; and the environmental factors of disease occurrence, such as the surrounding vegetation situation, soil acidity and alkalinity, etc. These indexes jointly form the pest and disease identification index set, providing a core basis for subsequent classification identification training and channel construction.

[0056] For the construction of the landscaping prediction channel, the classification and labeling training is carried out on the landscaping historical dataset using the landscaping prediction index set, thereby obtaining multiple branch landscaping index predictors. These predictors predict the growth status of garden plants from different perspectives. Then, these branch predictors are fused with equal weights to make them work together. Subsequently, through verification and evaluation optimization, the historical data and actual situation are used to test and adjust the fusion result, continuously improving the accuracy and reliability of the prediction, and then generating the landscaping prediction channel. When constructing the pest and disease identification channel, the convolutional neural network technology is used to carry out the classification and labeling training on the garden pest and disease historical dataset based on the pest and disease identification index set, obtaining multiple branch pest and disease index identifiers. After that, these branch identifiers are verified and evaluated to judge the accuracy of their pest and disease identification, and the fusion coefficient is calculated. This fusion coefficient is determined based on the pest and disease index decisive factor and the identification accuracy rate. Through it, the branch pest and disease index identifiers are weighted and fused to integrate the advantages of multiple identifiers, and finally the pest and disease identification channel is obtained.

[0057] From the perspective of architecture design, a parallel processing architecture mode is adopted to build a unified data input interface for the two channels, ensuring that the garden status data stream from the data perception module can be transmitted to these two channels simultaneously and without conflict. Then, the output results of the two channels are unifiedly planned, and a comprehensive output interface is designed to integrate the plant growth trend and health status prediction data output by the landscaping prediction channel with the identification results such as the types of pests and diseases, the infection range, and the severity output by the pest and disease identification channel. Through such integration, a comprehensive garden growth analysis result can be obtained at one time, including both the prediction of the plant growth trend and the relevant information of pests and diseases, providing rich and comprehensive data support for the subsequent strategy analysis module to formulate accurate maintenance strategies, greatly improving the efficiency and scientific nature of garden management, and ensuring the healthy and stable development of the garden ecosystem.

[0058] In a possible implementation manner, the classification and labeling training unit further includes:

[0059] The branch landscaping index predictor acquisition unit is used to carry out the classification and labeling training on the landscaping historical dataset using the landscaping prediction index set to obtain the branch landscaping index predictors.

[0060] The landscaping prediction channel generation unit is used to fuse the branch landscaping index predictors with equal weights and optimize through verification and evaluation to generate the landscaping prediction channel.

[0061] The branch pest and disease index identifier acquisition unit is used to carry out the classification and labeling training on the garden pest and disease historical dataset based on the pest and disease identification index set using the convolutional neural network to obtain the branch pest and disease index identifiers.

[0062] The pest and disease identification channel acquisition unit is used to verify, evaluate and calculate the fusion coefficient of the branch pest and disease index identifier, obtain the fusion coefficient of the branch identifier, and perform weighted fusion on the branch pest and disease index identifier based on the fusion coefficient of the branch identifier to obtain the pest and disease identification channel.

[0063] Specifically, the decision tree algorithm is used to classify and label the training of the landscaping historical data set, so as to obtain the branch landscaping index predictor. First, the landscaping prediction index set is used as the feature attribute of the decision tree, and the actual growth results of the plants corresponding to each sample in the landscaping historical data set are used as the class labels. Starting from the root node, the decision tree algorithm selects an optimal index from many prediction indexes to divide the data based on criteria such as information gain or Gini index. For example, taking the index of plant height as an example, if the class purity of the data can be maximally improved after dividing the data according to a certain height threshold, then this index and threshold are selected for division to generate left and right child nodes. Then, recursively repeat the above process on the child nodes, continuously subdivide the data until a predetermined stop condition is met, such as the number of samples in the node is too small or the class purity reaches an extremely high level, etc. In this way, a decision tree is constructed, which is a branch landscaping index predictor that can predict the growth state of plants, such as growth trend, health degree, etc., according to the input landscaping prediction index data.

[0064] Obtain multiple branch landscaping index predictors, which are obtained by classifying and labeling the training of the landscaping historical data set and predict the growth conditions of garden plants from different dimensions respectively. Then, these predictors are fused in an equal-weight manner, and each branch predictor has the same influence in the new comprehensive model, and their prediction results are equally treated and integrated. After fusion, in order to ensure the accuracy and reliability of the generated prediction channel, verification and evaluation optimization work will be carried out. A large amount of actual landscaping data that has not participated in the training is used as the verification set, and the output results of the prediction channel are compared and analyzed with the actual plant growth conditions in the verification set. According to the differences between the two, the parameters of the prediction channel are adjusted using optimization algorithms, such as adjusting the correlation weights between predictors, optimizing the internal algorithm logic, etc. After multiple rounds of verification, evaluation and parameter adjustment, the accuracy and stability of the prediction channel are gradually improved, and finally a landscaping prediction channel that can accurately predict the growth trend of garden plants and provide strong support for landscaping maintenance decisions is generated.

[0065] Use a convolutional neural network to conduct classification and labeling training on the historical dataset of garden pests and diseases based on the pest and disease identification index set, so as to obtain a branch pest and disease index identifier. First, convert the pest and disease identification index set into an input data format that can be processed by the convolutional neural network. For example, quantify the morphological characteristics of pests and diseases, the symptoms of plants being damaged, etc. into numerical matrices. Then, build a convolutional neural network model, which includes multiple convolutional layers, pooling layers, and fully connected layers. In the training stage, input the sample data with clear pest and disease category labels in the historical dataset of garden pests and diseases into the convolutional neural network in sequence. The convolutional layer slides the convolutional kernel over the data to extract local features in the data, and the pooling layer downsamples the feature map to reduce the data volume and retain key features. As the data is passed through the network layer by layer, the model continuously learns the association pattern between pest and disease features and categories. After multiple rounds of training, adjust the weight and bias parameters of the network so that the model can accurately output the corresponding pest and disease category prediction results according to the input pest and disease identification index data. At this time, the trained model becomes a branch pest and disease index identifier, which can be used to initially identify the types and related information of pests and diseases in the garden.

[0066] Conduct verification and evaluation. Use a validation dataset independent of the training set. Input the data into each branch identifier, compare the pest and disease identification results output by it with the true labels in the validation dataset, and use indicators such as accuracy, recall rate, and F1 value to comprehensively measure the performance of each branch identifier. Then carry out the calculation work of the fusion coefficient. Determine the fusion coefficient of each branch identifier according to the decisive factors of pest and disease indicators and the identification accuracy obtained from the verification and evaluation. For the branch identifier that plays a key role in pest and disease identification and has a high identification accuracy, give a larger fusion coefficient; otherwise, give a smaller coefficient. Finally, perform weighted fusion on all branch identifiers based on the calculated fusion coefficients of the branch identifiers. When new input data arrives, each branch identifier first gives its own identification result, and then weights and combines these results according to their respective fusion coefficients to obtain the final pest and disease identification result. In this way, a pest and disease identification channel with better comprehensive performance and more accurate identification is constructed.

[0067] In a possible implementation manner, the pest and disease identification channel acquisition unit further includes:

[0068] The decisive factor determination unit of pest and disease indicators is used to evaluate the importance of the pest and disease identification index set based on the historical dataset of garden pests and diseases, and determine the decisive factors of pest and disease indicators.

[0069] The fusion coefficient calculation formula construction unit is used to construct a fusion coefficient calculation formula: where D i represents the fusion coefficient of the i-th branch identifier, Identify the decisive factor of the pest and disease index for the i-th branch identifier, β i Identify the recognition accuracy rate of the i-th branch identifier, D i As increases, it increases.

[0070] The fusion coefficient calculation unit is used to perform performance verification evaluation and fusion coefficient calculation on the branch pest and disease index identifiers respectively based on the fusion coefficient calculation formula, and obtain the fusion coefficient of the branch identifier.

[0071] Specifically, extract information from the historical dataset of garden pests and diseases, which covers various pest and disease cases occurring at different times and locations, as well as multi-dimensional data such as corresponding environmental conditions and plant varieties. Then, use statistical analysis methods, such as correlation analysis, to calculate the correlation coefficient between each pest and disease recognition index and the actual occurrence of pests and diseases (such as pest and disease types, damage levels, etc.). The higher the correlation coefficient, the closer the relationship between the index and the pests and diseases, and the greater its importance in the recognition process. At the same time, adopt the feature importance evaluation function of the random forest algorithm, which judges the importance of the index by calculating the contribution degree of each index to reducing the uncertainty of the sample label during the construction of the decision tree. Combining the results of these statistical analysis and machine learning methods, rank the importance of each index in the pest and disease recognition index set, screen out those indexes that play a key role and have a great influence on pest and disease recognition, and determine these indexes as the decisive factors of the pest and disease index.

[0072] Construct a dedicated fusion coefficient calculation formula, that is In this formula, D i represents the fusion coefficient of the i-th branch identifier, and its size directly affects the weight of the branch identifier in the final pest and disease recognition result; represents the decisive factor of the pest and disease index of the i-th branch identifier, reflecting the importance of the key index relied on by the branch identifier; β i is the recognition accuracy rate of the i-th branch identifier, reflecting the performance of the branch identifier. And, D i As increases, it increases. This means that the branch identifier with a more critical pest and disease index decisive factor and a higher recognition accuracy rate has a larger fusion coefficient and a higher weight in the final recognition result.

[0073] According to the constructed calculation formula of the fusion coefficient, the performance verification and evaluation and the calculation of the fusion coefficient are carried out for each branch pest and disease index identifier respectively. During the performance verification and evaluation process, a large amount of test data will be used to test the actual performance of each branch identifier to ensure its reliability. Then, the fusion coefficient corresponding to each branch identifier is calculated through the formula, and these fusion coefficients will be used for the weighted fusion of the branch pest and disease index identifiers subsequently, so as to obtain a more accurate pest and disease identification channel and provide strong support for the prevention and control of garden pests and diseases.

[0074] In a possible implementation manner, the policy parsing module 30 further includes:

[0075] A garden greening maintenance label policy library acquisition unit, configured to classify and label the garden greening maintenance policy library according to the garden greening maintenance element information, so as to obtain a garden greening maintenance label policy library.

[0076] A greening maintenance policy parameter threshold acquisition unit, configured to perform policy retrieval and parsing in the garden greening maintenance label policy library based on the M garden growth analysis results, so as to obtain M greening maintenance policy parameter thresholds.

[0077] A target greening maintenance policy parameter acquisition unit, configured to perform global optimization within the M greening maintenance policy parameter thresholds to obtain M target greening maintenance policy parameters.

[0078] A garden greening maintenance policy parameter determination unit, configured to perform correlation analysis and equilibrium correction on the M target greening maintenance policy parameters to determine M garden greening maintenance policy parameters.

[0079] Specifically, according to the garden greening maintenance element information, the garden greening maintenance policy library is classified and labeled. All kinds of policies in the maintenance policy library are carefully classified according to maintenance elements such as plant maintenance types (such as watering, fertilizing, pruning, etc.), garden area functions (viewing area, leisure area, etc.), and seasonal characteristics, and corresponding labels are added to each kind of policy, so as to construct a well-organized garden greening maintenance label policy library.

[0080] When determining the threshold values of the M greening maintenance strategy parameters, the M garden growth analysis results are used as the retrieval basis. These analysis results contain rich information such as the growth trend of garden plants and the occurrence of pests and diseases. Each garden growth analysis result is compared with the labels in the greening maintenance label strategy library. For example, if the growth analysis result of a certain garden area shows that the plants show signs of water shortage and have minor pests and diseases, retrieve the maintenance strategies with the relevant labels of "plant water shortage" and "minor pests and diseases" in the strategy library. For each retrieved strategy, deeply analyze its content and extract the key parameters related to maintenance operations, such as the amount and frequency of watering, the concentration and dosage of pesticides for pest control, etc. Since the same maintenance scenario corresponds to multiple different strategies, the parameter values in these strategies will vary. Integrate the value ranges of these parameters to determine a reasonable parameter threshold interval for each garden area, and finally obtain the threshold values of the M greening maintenance strategy parameters.

[0081] According to the threshold values of the M greening maintenance strategy parameters, initialize the particle swarm space of the M greening maintenance strategies. In this space, each particle represents a potential combination of greening maintenance strategy parameters. These combinations are randomly generated within the previously determined parameter threshold range, providing diverse initial solutions for subsequent optimization. Clearly define the greening maintenance objectives, which include promoting the healthy growth of plants, controlling pests and diseases, and enhancing the landscape effect. Deeply analyze the effect indicators of these objectives and convert them into quantifiable indicators, such as plant growth height, pest and disease incidence rate, landscape beauty score, etc. Then, through data correlation fitting, analyze the relationship between different maintenance strategy parameters and these quantifiable indicators, and construct a fitness function for the greening maintenance effect. Finally, use the constructed fitness function for the greening maintenance effect to perform iterative global optimization within the particle swarm space of the M greening maintenance strategies. In each iteration, the particle adjusts its position according to its own historical optimal position and the global optimal position of the entire particle swarm, that is, changes the combination of maintenance strategy parameters. By continuously calculating the fitness value of each particle and comparing their magnitudes, gradually screen out the particle with the maximum fitness. After multiple rounds of iteration, finally determine the M target greening maintenance strategy parameters with the maximum particle fitness. These parameters are the combination of strategy parameters that are most conducive to achieving the greening maintenance objectives within the given parameter threshold range.

[0082] Combine the information of M garden areas, such as the functions, geographical features, and plant species distributions of each area, and conduct an analysis of the associated impacts on the M target greening maintenance strategy parameters. This analysis process will deeply explore the impacts of the unique attributes of each garden area on the maintenance strategy parameters. For example, in the densely planted ornamental area, the fertilization strategy may be affected by space limitations and landscape requirements and cannot be simply implemented according to general standards. Through this analysis, quantify the influence degree of each area's attributes on the maintenance strategy parameters, and then obtain M greening maintenance influence coefficients. These coefficients reflect the degree of closeness of the association between different garden areas and the maintenance strategy parameters. Based on the obtained M greening maintenance influence coefficients, use the Nash equilibrium theory to correct the M target greening maintenance strategy parameters, so that when the strategies of other participants remain unchanged, the adjustment of one's own strategy will not bring better results. In the garden maintenance scenario, each garden area can be regarded as a participant, and the maintenance strategy parameters are decision variables. According to the greening maintenance influence coefficients, adjust the target greening maintenance strategy parameters, comprehensively consider the mutual influences between different areas, avoid over-optimizing a certain area and affecting the maintenance effects of other areas, and seek a balanced combination of maintenance strategy parameters among the areas. After such Nash equilibrium correction, finally determine the M greening and landscaping maintenance strategy parameters, so as to provide a scientific and reasonable basis for the precise maintenance of the entire garden and ensure that each area of the garden reaches the best state under unified maintenance management.

[0083] In a possible implementation manner, the target greening maintenance strategy parameter acquisition unit further includes:

[0084] A greening maintenance strategy particle swarm space initialization unit, configured to initialize an M greening maintenance strategy particle swarm space according to the M greening maintenance strategy parameter thresholds.

[0085] A fitness function construction unit, configured to obtain the target of greening and landscaping maintenance, perform effect index analysis and data association fitting on the target of greening and landscaping maintenance, and construct a greening and landscaping maintenance effect fitness function.

[0086] An iterative global optimization unit, configured to perform iterative global optimization in the M greening maintenance strategy particle swarm space by using the greening and landscaping maintenance effect fitness function, and determine the M target greening maintenance strategy parameters with the maximum particle fitness.

[0087] Specifically, clarify the threshold values of these M greening maintenance strategy parameters. For each garden area (a total of M areas), construct the corresponding particle swarm space respectively. In each particle swarm space, a large number of particles will be randomly generated, and each particle represents a unique combination of greening maintenance strategy parameters. The parameter values in these combinations are strictly within the corresponding parameter threshold ranges. For example, if the threshold of fertilization amount in a certain area is 0.5 - 2 kg per square meter and the threshold of watering frequency is 3 - 7 times per week, then the fertilization amount parameter of the particle will randomly take values between 0.5 - 2 kg, and the watering frequency parameter will be randomly determined between 3 - 7 times per week, thus forming a diverse set of initial parameter combinations. These particles together constitute the greening maintenance strategy particle swarm space.

[0088] When constructing the fitness function for the effect of garden greening maintenance, first determine the goals of garden greening maintenance, which mainly cover several key aspects such as promoting plant growth, enhancing landscape effects, and preventing and controlling pests and diseases. Analyze the effect indicators for these goals. For example, select the plant growth indicator G (which can be measured by a comprehensive value calculated from plant height, number of leaves, etc.), the landscape effect indicator L (quantified by landscape aesthetic scores), and the pest and disease prevention and control indicator P (represented by the reciprocal of the pest and disease infection rate, the lower the infection rate, the larger this value). Then conduct data correlation fitting. By analyzing historical maintenance data, it is found that plant growth is greatly affected by the fertilization amount and watering amount, the landscape effect is related to plant layout and pruning frequency, and pest and disease prevention and control is closely related to the dosage of pesticides. These relationships are fitted through linear relationships, and the relationships between plant growth and maintenance factors are obtained as G = a1x1 + a2x2 (x1 represents the fertilization amount, x2 represents the watering amount, a1 and a2 are the corresponding influence coefficients), the relationship between landscape effect and maintenance factors is L = b1y1 + b2y2 (y1 represents the parameter for adjusting plant layout, y2 represents the pruning frequency, b1 and b2 are influence coefficients), and the relationship between pest and disease prevention and control and maintenance factors is P = c1z1 (z1 represents the dosage of pesticides, c1 is the influence coefficient). Based on the above analysis, construct a simple fitness function for the effect of garden greening maintenance F = w1G + w2L + w3P, where w1, w2, and w3 are the weights of the three aspects of plant growth, landscape effect, and pest and disease prevention and control respectively, and w1 + w2 + w3 = 1. The setting of the weights is determined according to the focus of different gardens. For example, in a garden that focuses on plant growth, the value of w1 is relatively large. Through this function, the maintenance effect under different maintenance strategy combinations can be quantified, which is convenient for screening out the most suitable plan for the maintenance goals among numerous strategies.

[0089] Use the constructed fitness function for the effect of landscaping maintenance to conduct iterative global optimization within the space of M landscaping maintenance strategy particle swarms, and then determine the M target landscaping maintenance strategy parameters. In each landscaping maintenance strategy particle swarm space, numerous particles each represent a different combination of landscaping maintenance strategy parameters. In the initial state, these particles are randomly distributed within the established parameter threshold range. Taking a certain particle swarm as an example, one of the particles may represent a specific combination of fertilization amount, watering frequency, and pruning cycle. When starting the iterative optimization, first substitute the parameter combination represented by each particle into the fitness function for the effect of landscaping maintenance for calculation to obtain a fitness value, which reflects the degree of fit of this set of maintenance strategies to achieving the maintenance goal. In each iteration process, the particle will adjust its position by referring to the position corresponding to the optimal fitness value obtained in its own history (i.e., the combination of maintenance strategy parameters that once performed best) and the global optimal position currently found by the entire particle swarm (the parameter combination with the largest fitness value among all current particles), that is, change the combination of maintenance strategy parameters it represents. For example, if a certain particle finds that moving in the direction of the global optimal particle can improve its fitness value, it will correspondingly adjust parameters such as fertilization amount and watering frequency. With each iteration, the entire particle swarm will gradually move in a more optimal direction, and the fitness value will continuously increase. After multiple rounds of iteration, in each particle swarm, the combination of landscaping maintenance strategy parameters represented by the particle with the largest fitness value gradually stands out. The parameter combinations corresponding to these particles with the largest fitness values respectively selected from the M particle swarms are the finally determined M target landscaping maintenance strategy parameters. These parameter combinations can maximize the satisfaction of the landscaping maintenance goal after comprehensively considering various factors such as plant growth, landscape effect, and pest control, providing a scientific and effective guiding basis for the refined maintenance of the garden.

[0090] In a possible implementation manner, the determining of the M landscaping maintenance strategy parameters further includes:

[0091] A greening maintenance influence coefficient acquisition unit, configured to perform associated influence analysis on the M target landscaping maintenance strategy parameters according to the M garden area information to obtain M greening maintenance influence coefficients.

[0092] A Nash equilibrium correction unit, configured to perform Nash equilibrium correction on the M target landscaping maintenance strategy parameters based on the M greening maintenance influence coefficients to determine the M landscaping maintenance strategy parameters.

[0093] Specifically, comprehensively sort out the information of M garden areas, which covers various aspects such as the geographical location, topography, distribution of plant varieties, regional functional positioning (such as leisure area, viewing area, etc.), and past maintenance history. For each garden area, analyze the unique information characteristics corresponding to the target greening maintenance strategy parameters one by one. Taking a certain garden area as an example, if the area is located at the tuyere position and mostly planted with shallow-rooted plants, then when analyzing the target greening maintenance strategy parameters, its wind prevention measures (such as the height and material of the enclosure) will be affected by the geographical location and plant varieties. From the perspective of geographical location, the wind force is relatively large at the tuyere, so the height of the enclosure may need to be increased; from the perspective of plant varieties, shallow-rooted plants have weak wind resistance and require higher stability and protection range of the enclosure. Through in-depth research on this influence relationship and using the method of quantitative analysis, establish a regression model, taking each factor in the regional information as an independent variable and the target greening maintenance strategy parameters as a dependent variable, and calculate the influence degree of each factor on the parameters. After comprehensively considering all relevant factors, obtain a value that can quantify the influence degree of the regional information on the target greening maintenance strategy parameters. This value is a greening maintenance influence coefficient. In this way, conduct correlation influence analysis on each of the M garden areas one by one, and finally obtain M greening maintenance influence coefficients, which provide a strong basis for subsequent precise adjustment of the maintenance strategy parameters.

[0094] The key stage is to use these coefficients to correct the Nash equilibrium of the M target greening maintenance strategy parameters, and then determine the final M landscape gardening maintenance strategy parameters. The core of the Nash equilibrium theory lies in that in a system involving multiple decision-making subjects and mutual influence, the decision of each subject should not only consider its own interests, but also take into account the decisions of other subjects to achieve a balanced state. At this time, any subject unilaterally changing its decision cannot obtain better results. In the landscape gardening maintenance scenario, the M garden areas are the M decision-making subjects, and the target greening maintenance strategy parameters of each area are their respective decision variables. Based on the obtained M greening maintenance influence coefficients, analyze the mutual connection and influence among the areas. For example, increasing the watering frequency in area A may lead to an increase in soil humidity in the surrounding area B, affecting the growth environment of the plants in area B, and then changing the originally set watering and fertilization strategy parameters in area B. At this time, according to the Nash equilibrium principle, it is necessary to comprehensively consider the interests of each area and the overall landscape gardening maintenance goal. Adjust the target greening maintenance strategy parameters of each area, while meeting the basic maintenance needs of each area, ensure the balance and stability of the entire garden system. During the adjustment process, continuously try different parameter combinations, calculate the maintenance effects of each area under each combination, and perform weight distribution according to the greening maintenance influence coefficients. Through multiple iterations and optimizations, finally determine a set of maintenance strategy parameters that achieve Nash equilibrium among the areas, that is, the M landscape gardening maintenance strategy parameters. These parameters not only consider the unique needs of each area, but also take into account the mutual influence among the areas, and can realize the scientific and efficient maintenance of the entire garden.

[0095] Example 2, based on the same inventive concept as the digital intelligent management system for landscape gardening maintenance in the foregoing example, as Figure 2 shown, this application provides a digital intelligent management method for landscape gardening maintenance. The method in the embodiments of this application and the system embodiments are based on the same inventive concept. Among them, the method includes:

[0096] Step S100: Divide and code the target garden into areas to obtain M garden area information, and deploy a sensing network to collect and obtain M garden state data streams of the M garden area information.

[0097] Step S200: Construct a dual-channel garden analysis. The dual-channel garden analysis includes a landscape gardening prediction channel and a pest identification channel. Use the dual-channel garden analysis to perform parallel analysis on the M garden state data streams and output M garden growth analysis results.

[0098] Step S300: Obtain a landscape gardening maintenance strategy library, and use the landscape gardening maintenance strategy library to perform strategy analysis and correlation correction on the M garden growth analysis results to determine M landscape gardening maintenance strategy parameters.

[0099] Step S400: Based on the matching and activation of the M landscaping maintenance strategy parameters and the list of garden maintenance equipment, a set of garden linkage maintenance equipment is obtained, and remote control of greening maintenance of the target garden is performed through the set of garden linkage maintenance equipment.

[0100] Furthermore, step S100 further includes:

[0101] Step S110: Obtain the factors for garden area division, where the factors for garden area division include regional functions, geographical features, and regional area.

[0102] Step S120: Sort the factors for garden area division according to the factors for garden area division to obtain a priority sequence of garden division factors.

[0103] Step S130: Based on the priority sequence of garden division factors, perform multi-level regional division on the design and survey data of the target garden to obtain M garden division regions.

[0104] Step S140: According to the priority sequence of garden division factors, design a regional coding system, where the regional coding system includes coding content, coding order, and coding identification.

[0105] Step S150: Label the M garden division regions according to the regional coding system to obtain the M garden area information.

[0106] Furthermore, step S120 further includes:

[0107] Step S121: Mine the maintenance impact data based on the factors for garden area division to obtain a garden division factor maintenance impact data set.

[0108] Step S122: Perform standardization processing and principal component selection on the garden division factor maintenance impact data set to obtain the principal component information of garden division factors.

[0109] Step S123: Based on the principal component information of garden division factors, perform PCA weight assignment on each division factor in the factors for garden area division to determine the weight factors of garden division factors.

[0110] Step S124: Sort the factors for garden area division according to the weight factors of garden division factors to obtain the priority sequence of garden division factors.

[0111] Furthermore, step S200 further includes:

[0112] Step S210: Collect and obtain the historical data set of landscaping and the historical data set of garden pests and diseases.

[0113] Step S220: Obtain the landscaping prediction target and the pest identification target, extract indicators for the landscaping prediction target and the pest identification target, and obtain a landscaping prediction indicator set and a pest identification indicator set.

[0114] Step S230: Use the landscaping prediction indicator set and the pest identification indicator set to perform classification and identification training on the landscaping historical data set and the garden pest historical data set respectively, and obtain a landscaping prediction channel and a pest identification channel.

[0115] Step S240: Integrate the landscaping prediction channel and the pest identification channel in parallel to form the constructed garden analysis dual channel.

[0116] Further, step S230 further includes:

[0117] Step S231: Use the landscaping prediction indicator set to perform classification and identification training on the landscaping historical data set, and obtain a branch landscaping indicator predictor.

[0118] Step S232: Perform equal-weight fusion and verification and evaluation optimization on the branch landscaping indicator predictor to generate a landscaping prediction channel.

[0119] Step S233: Use a convolutional neural network to perform classification and identification training on the garden pest historical data set based on the pest identification indicator set, and obtain a branch pest indicator identifier.

[0120] Step S234: Perform verification and evaluation and fusion coefficient calculation on the branch pest indicator identifier, obtain a branch identifier fusion coefficient, and perform weighted fusion on the branch pest indicator identifier based on the branch identifier fusion coefficient to obtain a pest identification channel.

[0121] Further, step S234 further includes:

[0122] Step S2341: Perform importance evaluation on the pest identification indicator set based on the garden pest historical data set to determine the pest indicator decisive factor.

[0123] Step S2342: Construct a fusion coefficient calculation formula: where D i represents the fusion coefficient of the i-th branch identifier, identifies the pest indicator decisive factor of the i-th branch identifier, β i identifies the recognition accuracy of the i-th branch identifier, D i increases as increases.

[0124] Step S2343: Based on the fusion coefficient calculation formula, perform performance verification and evaluation on the branch pest and disease index recognizers respectively, and calculate the fusion coefficients of the branch recognizers to obtain the branch recognizer fusion coefficients.

[0125] Further, step S300 further includes:

[0126] Step S310: Classify and label the landscaping maintenance strategy library according to the landscaping maintenance element information to obtain the landscaping maintenance labeled strategy library.

[0127] Step S320: Based on the M garden growth analysis results, perform strategy retrieval and analysis in the landscaping maintenance labeled strategy library to obtain M greening maintenance strategy parameter thresholds.

[0128] Step S330: Perform global optimization within the M greening maintenance strategy parameter thresholds to obtain M target greening maintenance strategy parameters.

[0129] Step S340: Perform correlation analysis and equilibrium correction on the M target greening maintenance strategy parameters to determine M landscaping maintenance strategy parameters.

[0130] Further, step S330 further includes:

[0131] Step S331: According to the M greening maintenance strategy parameter thresholds, initialize the M greening maintenance strategy particle swarm spaces.

[0132] Step S332: Obtain the landscaping maintenance target, perform effect index analysis and data correlation fitting on the landscaping maintenance target, and construct a landscaping maintenance effect fitness function.

[0133] Step S333: Use the landscaping maintenance effect fitness function to perform iterative global optimization within the M greening maintenance strategy particle swarm spaces to determine M target greening maintenance strategy parameters with the maximum particle fitness.

[0134] Further, step S340 further includes:

[0135] Step S341: Perform correlation impact analysis on the M target greening maintenance strategy parameters according to the M garden area information to obtain M greening maintenance impact coefficients.

[0136] Step S342: Based on the M greening maintenance impact coefficients, perform Nash equilibrium correction on the M target greening maintenance strategy parameters to determine the M landscaping maintenance strategy parameters.

[0137] It should be noted that the above order of the embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. In addition, the specific embodiments of this specification have been described. Moreover, the processes depicted in the drawings do not necessarily require the specific order or consecutive order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0138] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included within the protection scope of the present application.

[0139] This specification and the drawings are only exemplary descriptions of the present application and are considered to have covered any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. A digital management system for landscaping maintenance, characterized in that, The system includes: A data perception module, which is used to divide and code the target garden into regions, obtain M garden area information, and sense and collect M garden state data streams of the M garden area information through network deployment; A parallel analysis module, which is used to construct a dual-channel garden analysis. The dual-channel garden analysis includes a landscape gardening prediction channel and a pest and disease identification channel, and parallelly analyze the M garden state data streams by using the dual-channel garden analysis to output M garden growth analysis results; A strategy parsing module, which is used to obtain a landscape gardening maintenance strategy library, and perform strategy parsing and correlation correction on the M garden growth analysis results by using the landscape gardening maintenance strategy library to determine M landscape gardening maintenance strategy parameters; A digital control module, which is used to match and activate based on the M landscape gardening maintenance strategy parameters and a list of garden maintenance equipment to obtain a set of garden linkage maintenance equipment, and perform remote control of landscape gardening maintenance on the target garden through the set of garden linkage maintenance equipment.

2. The digital intelligent management system for landscaping maintenance according to claim 1, wherein The data perception module further includes: A garden area division factor acquisition unit, which is used to acquire garden area division factors, and the garden area division factors include regional functions, geographical features, and regional areas; A garden division factor priority sequence acquisition unit, which is used to sort the garden area division factors according to the garden area division factors to obtain a garden division factor priority sequence; A garden division area acquisition unit, which is used to perform multi-level division of the design survey data of the target garden based on the garden division factor priority sequence to obtain M garden division areas; A regional coding system design unit, which is used to design a regional coding system according to the garden division factor priority sequence, and the regional coding system includes coding content, coding order, and coding identification; A garden area information acquisition unit, which is used to code and label the M garden division areas according to the regional coding system to obtain the M garden area information.

3. The digital intelligent management system for landscaping maintenance according to claim 2, characterized in that, The garden division factor priority sequence acquisition unit further includes: A garden division factor maintenance impact data set acquisition unit, which is used to mine maintenance impact data based on the garden area division factors to obtain a garden division factor maintenance impact data set; A garden division factor principal component information acquisition unit, which is used to perform standardization processing and principal component selection on the garden division factor maintenance impact data set to obtain garden division factor principal component information; A garden division factor weight factor determination unit, which is used to perform PCA weight allocation on each division factor in the garden area division factors based on the garden division factor principal component information to determine garden division factor weight factors; A priority sorting unit, which is used to sort the garden area division factors according to the garden division factor weight factors to obtain the garden division factor priority sequence.

4. A digital management system for landscaping maintenance as described in claim 1, characterized in that, The parallel analysis module further includes: A historical data set acquisition unit, which is used to collect and obtain a historical data set of landscape gardening and a historical data set of garden pests and diseases; An index extraction unit, configured to obtain a landscaping prediction target and a pest identification target, extract indexes for the landscaping prediction target and the pest identification target, and obtain a landscaping prediction index set and a pest identification index set; A classification identification training unit, configured to perform classification identification training on the landscaping historical data set and the garden pest historical data set by using the landscaping prediction index set and the pest identification index set respectively, and obtain a landscaping prediction channel and a pest identification channel; A garden analysis dual-channel construction unit, configured to perform parallel integration on the landscaping prediction channel and the pest identification channel to construct the garden analysis dual-channel.

5. The digital intelligent management system for landscaping maintenance according to claim 4, wherein The classification identification training unit further includes: A branch landscaping index predictor acquisition unit, configured to perform classification identification training on the landscaping historical data set by using the landscaping prediction index set, and obtain a branch landscaping index predictor; A landscaping prediction channel generation unit, configured to perform equal-weight fusion and verification evaluation optimization on the branch landscaping index predictor to generate a landscaping prediction channel; A branch pest index identifier acquisition unit, configured to perform classification identification training on the garden pest historical data set by using a convolutional neural network based on the pest identification index set, and obtain a branch pest index identifier; A pest identification channel acquisition unit, configured to perform verification evaluation and fusion coefficient calculation on the branch pest index identifier, obtain a branch identifier fusion coefficient, and perform weighted fusion on the branch pest index identifier based on the branch identifier fusion coefficient to obtain a pest identification channel.

6. The digital intelligent management system for landscaping maintenance according to claim 5, characterized in that, The pest identification channel acquisition unit further includes: A pest index decisive factor determination unit, configured to perform importance evaluation on the pest identification index set based on the garden pest historical data set to determine a pest index decisive factor; A fusion coefficient calculation formula construction unit for constructing a fusion coefficient calculation formula: where D i represents the fusion coefficient of the i-th branch recognizer, identifies the decisive factor of the pest and disease index of the i-th branch recognizer, β i identifies the recognition accuracy of the i-th branch recognizer, D i increases as increases; A fusion coefficient calculation unit, configured to perform performance verification evaluation and fusion coefficient calculation on the branch pest index identifier respectively based on the fusion coefficient calculation formula to obtain the branch identifier fusion coefficient.

7. The digital intelligent management system for landscaping maintenance according to claim 6, characterized in that, The policy analysis module further includes: A landscaping maintenance label policy library acquisition unit, configured to perform classification labeling on the landscaping maintenance policy library according to landscaping maintenance element information to obtain a landscaping maintenance label policy library; A greening maintenance policy parameter threshold acquisition unit, configured to perform policy retrieval and analysis in the landscaping maintenance label policy library based on the M garden growth analysis results to obtain M greening maintenance policy parameter thresholds; A target greening maintenance policy parameter acquisition unit, configured to perform global optimization within the M greening maintenance policy parameter thresholds to obtain M target greening maintenance policy parameters; A landscaping maintenance policy parameter determination unit, configured to perform correlation analysis and balanced correction on the M target greening maintenance policy parameters to determine M landscaping maintenance policy parameters.

8. The digital intelligent management system for landscaping maintenance according to claim 7, characterized in that, The target greening maintenance policy parameter acquisition unit further includes: A space initialization unit, configured to initialize an M-dimensional greening maintenance policy particle swarm space according to the M greening maintenance policy parameter thresholds; A fitness function construction unit, which is used to obtain the landscaping maintenance objectives, analyze the effect indicators of the landscaping maintenance objectives and perform data correlation fitting, and construct a landscaping maintenance effect fitness function; An iterative global optimization unit, which is used to perform iterative global optimization in the M greening maintenance strategy particle swarm spaces by using the landscaping maintenance effect fitness function, and determine M target greening maintenance strategy parameters with the maximum particle fitness.

9. The digital intelligent management system for landscaping maintenance according to claim 8, wherein The landscaping maintenance strategy parameter determination unit further includes: A greening maintenance influence coefficient acquisition unit, which is used to perform associated influence analysis on the M target greening maintenance strategy parameters according to the M garden area information, and obtain M greening maintenance influence coefficients; A Nash equilibrium correction unit, which is used to perform Nash equilibrium correction on the M target greening maintenance strategy parameters based on the M greening maintenance influence coefficients, and determine the M landscaping maintenance strategy parameters.

10. A digital intelligent management method for landscaping maintenance, characterized in that, The method is implemented by a digital intelligent management system for landscaping maintenance according to any one of claims 1-9, and the method includes: Performing area division coding on the target garden to obtain M garden area information, and deploying a sensing network to collect and obtain M garden state data streams of the M garden area information; Constructing a dual-channel garden analysis, where the dual-channel garden analysis includes a landscaping prediction channel and a pest identification channel, and using the dual-channel garden analysis to perform parallel analysis on the M garden state data streams, and outputting M garden growth analysis results; Obtaining a landscaping maintenance strategy library, and using the landscaping maintenance strategy library to perform strategy analysis and associated correction on the M garden growth analysis results, and determining M landscaping maintenance strategy parameters; Based on the matching and activation of the M landscaping maintenance strategy parameters with the list of garden maintenance equipment, obtaining a set of garden linkage maintenance equipment, and performing remote control of greening maintenance on the target garden through the set of garden linkage maintenance equipment.

Images