Garden environment informatization management system based on diversified analysis

Abstract

The invention discloses a garden environment informatization management system based on diversified analysis, which belongs to the technical field of garden environment management and comprises a management platform, a parameter monitoring module, a vegetation growth evaluation module, an environment influence analysis module, an early warning unit and a decision execution module. According to the method, a partition management mode is adopted, the vegetation growth state of a local area is comprehensively analyzed based on soil data and vegetation apparent images of a single area, the vegetation abnormal growth type of an abnormal management area is obtained through abnormal recognition, and historical data tracing analysis is combined, so that the vegetation abnormal growth state of the abnormal management area is obtained. The method specifically comprises the steps of performing root cause judgment based on historical meteorological data and facility data so as to generate different strategy information according to a root cause judgment result, constructing a closed-loop management system from anomaly recognition to data attribution to decision execution, and performing decision execution by using diversified data and an intelligent analysis technology and taking data as driving. And comprehensive monitoring, multi-dimensional analysis and accurate decision making of environmental elements are realized.

Description

Technical Field

[0001] This invention relates to the field of landscape environment management technology, and more specifically, to a landscape environment information management system based on multivariate analysis. Background Technology

[0002] The garden area is planted with a wide variety of plants. Garden management involves multiple aspects such as plants, soil, water bodies and facilities, and is a systematic project that covers multiple dimensions such as ecology, function, space and humanities.

[0003] In traditional garden environment management, the main approach is to regulate the garden vegetation environment by collecting production data such as environmental temperature and humidity, as well as through regular inspections and personal experience of garden workers. However, the plants growing in different parks not only have different growth habits but also different maintenance needs. This often only allows for a rough maintenance of the overall state of the garden, making it difficult to achieve zoned supervision of garden areas, or to conduct targeted diagnosis of abnormal root causes by combining historical meteorological and facility data, and to generate detailed management methods.

[0004] To address the aforementioned issues, we propose a landscape environment information management system based on diversified analysis. Summary of the Invention

[0005] The purpose of this invention is to solve existing problems and provide a garden environment information management system based on diversified analysis compared with existing technologies.

[0006] The objective of this invention can be achieved through the following technical solution: a garden environment information management system based on diversified analysis, including a management platform, a parameter monitoring module, a vegetation growth assessment module, an environmental impact analysis module, an early warning unit, and a decision execution module;

[0007] The management platform divides the overall garden area into multiple sub-management areas based on vegetation characteristics, and assigns each sub-area a number;

[0008] The parameter monitoring module is used to collect vegetation growth information and environmental impact information within the sub-management area, and send the vegetation growth information and environmental impact information to the vegetation growth assessment module and the environmental impact analysis module, respectively.

[0009] The vegetation growth assessment module acquires vegetation growth information, including soil data and vegetation appearance images. Based on the fusion analysis of soil data and vegetation appearance images, it determines whether the vegetation growth status of the sub-management area is qualified. Sub-management areas that are not qualified are marked as abnormal management areas. The abnormal vegetation growth type of the abnormal management area is obtained, and the number of the abnormal management area is sent to the environmental impact analysis module.

[0010] The environmental impact analysis module obtains environmental impact information of the abnormal management area through the parameter monitoring module. The environmental impact information includes historical meteorological data and facility data. Based on the historical meteorological data and facility data, the module performs a fusion analysis on the abnormal vegetation growth status of the abnormal management area, generates meteorological impact signals, facility impact signals, or composite impact signals, and performs multi-dimensional root cause determination based on the impact signals. The impact signals are then sent to the early warning unit, and the root cause determination results are sent to the management platform.

[0011] The management platform generates different strategy information based on the root cause determination results and sends it to the decision execution module.

[0012] Furthermore, the process by which the vegetation growth assessment module determines whether the vegetation growth status is satisfactory includes:

[0013] The vegetation appearance image of the sub-management area is acquired. After image preprocessing, the vegetation area in the garden image is separated from the image background and its features are extracted. An abnormal feature recognition model that has been trained and optimized in advance is used to identify abnormal appearance features of the vegetation appearance image. The matching similarity between each abnormal appearance feature and the corresponding set abnormal appearance feature is obtained. When the matching similarity of any item is not within the preset matching similarity range, a vegetation appearance abnormal signal is generated; otherwise, a vegetation appearance normal signal is generated.

[0014] Soil data of sub-management areas are obtained with the same timestamp. Multiple data points in the soil data are compared one by one with the preset standard range corresponding to the current growth stage of vegetation. When any soil data item exceeds the preset standard range, an abnormal soil index signal is generated; otherwise, a normal soil index signal is generated.

[0015] When both vegetation appearance and soil indicators are normal, the vegetation growth status of the sub-management area is deemed acceptable; otherwise, the vegetation growth status of the sub-management area is deemed unacceptable.

[0016] Furthermore, abnormal appearance features include color abnormality texture features, surface lesion features, and surface pest features. The abnormal feature recognition model is a deep learning model. During the training process, the model continuously learns different types of abnormal features to accurately identify abnormal features in new vegetation appearance images and performs phenological period calibration for different seasons.

[0017] Soil data include soil surface moisture, deep soil salinity, soil organic matter content, soil pH, and soil heavy metal content in the sub-management areas.

[0018] Furthermore, when the vegetation growth status of a sub-management area is determined to be unqualified, the unqualified sub-management area is marked as an abnormal management area. The abnormal appearance characteristics of the abnormal management area and soil data are input into a multi-parameter fusion verification model to obtain the abnormal vegetation growth type. The multi-parameter fusion verification model is also a deep learning model. The multi-parameter fusion verification model integrates deep learning and domain knowledge, and achieves accurate diagnosis through spatiotemporally aligned multi-source data collaborative training.

[0019] Furthermore, the process of the environmental impact analysis module to perform integrated analysis on the abnormal growth status of vegetation in the abnormal management area includes: obtaining historical meteorological data of the abnormal management area with the same timestamp, including average rainfall, average solar radiation, average temperature, and average humidity; retrieving the preset suitable meteorological parameters for the current growth stage of vegetation in the abnormal management area stored in the management platform; and generating a meteorological impact signal when any item in the historical meteorological data is not within the range of the corresponding preset suitable meteorological parameters.

[0020] The facility data of the abnormal management area is obtained with the same timestamp. The facility data includes average irrigation amount, average fertilizer amount, and average pesticide application amount. The preset suitable facility parameters for the current growth stage of the vegetation in the abnormal management area are retrieved and stored in the management platform. When any one of the facility data is not within the corresponding preset suitable facility parameters, a facility impact signal is generated.

[0021] A composite influence signal is generated when any one of the historical meteorological data is outside the corresponding preset suitable meteorological parameter range, or when any one of the facility data is outside the corresponding preset suitable facility parameter range.

[0022] Furthermore, when generating a meteorological impact signal, historical meteorological data that is not within the corresponding preset suitable meteorological parameter range is acquired and marked as abnormal meteorological data. The abnormal meteorological data is compared with the suitable meteorological parameter range. When the abnormal meteorological data is greater than the maximum value of the preset suitable meteorological parameter range, the meteorological abnormality fluctuation value of the abnormal meteorological data relative to the maximum value of the preset suitable meteorological parameter range is acquired. When the abnormal meteorological data is less than the minimum value of the preset suitable meteorological parameter range, the meteorological abnormality fluctuation value of the abnormal meteorological data relative to the minimum value of the preset suitable meteorological parameter range is acquired.

[0023] When the abnormal meteorological fluctuation value is greater than the corresponding preset abnormal meteorological fluctuation threshold, the abnormal meteorological data is determined to be the cause of the abnormal meteorological data. When the abnormal meteorological fluctuation value is less than the corresponding preset abnormal meteorological fluctuation threshold, the abnormal meteorological data is determined to be the cause of the abnormal meteorological data.

[0024] Furthermore, when generating a facility impact signal, facility data that is not within the corresponding preset suitable facility parameters is acquired and marked as abnormal facility data. The abnormal facility data is compared with the suitable facility parameter range, and the abnormally large or small facility fluctuation value is obtained. When the abnormally large facility fluctuation value is greater than the corresponding preset abnormally large facility fluctuation threshold, the abnormal facility data is determined to be the cause of the abnormality. When the abnormally small facility fluctuation value is less than the corresponding preset abnormally small facility fluctuation threshold, the abnormal facility data is determined to be the cause of the abnormality.

[0025] Furthermore, the management platform generates corresponding levels of strategy information based on the causes of meteorological anomalies (both excessive and insufficient) and facility anomalies (both excessive and insufficient).

[0026] Compared with the prior art, the advantages of this invention are:

[0027] 1. This solution divides the overall garden area into multiple sub-management areas based on vegetation characteristics. It comprehensively analyzes the vegetation growth status of local areas based on soil data and vegetation appearance images of individual areas. It identifies abnormal vegetation growth types in abnormal management areas through anomaly identification and combines historical data for retrospective analysis. Specifically, it determines the root causes based on historical meteorological data and facility data, so as to generate different strategy information based on the root cause determination results. It constructs a closed-loop management system from anomaly identification → data attribution → decision execution. It utilizes diversified data and intelligent analysis technology to achieve comprehensive monitoring, multi-dimensional analysis and accurate decision-making of environmental elements with data as the driving force.

[0028] 2. Based on the above, in the process of determining the root cause of abnormal vegetation growth based on historical meteorological data and facility data, historical meteorological data and facility data of the abnormal management area are obtained with the same timestamp. Historical meteorological data that is not within the corresponding preset suitable meteorological parameter range is obtained and marked as abnormal meteorological data. Facility data that is not within the corresponding preset suitable facility parameters is obtained and marked as abnormal facility data. The abnormal meteorological data and the abnormal facility data are compared with the suitable meteorological parameter range and the abnormal facility data are compared with the suitable facility parameter range to obtain the meteorological abnormality fluctuation value and the facility abnormality fluctuation value respectively. Threshold comparisons are then performed to determine the specific abnormal cause. Attached Figure Description

[0029] Figure 1 This is a system principle block diagram of the present invention;

[0030] Figure 2 This is a flowchart of the method of the present invention. Detailed Implementation

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0032] Example 1: This invention discloses a garden environment information management system based on multivariate analysis. Please refer to [link / reference]. Figure 1 , Figure 2 It includes a management platform, a parameter monitoring module, a vegetation growth assessment module, an environmental impact analysis module, an early warning unit, and a decision execution module;

[0033] The management platform divides the overall garden area into multiple sub-management areas based on vegetation characteristics, and assigns a number to each of them.

[0034] The parameter monitoring module is used to collect vegetation growth information and environmental impact information within the sub-management area, and send the vegetation growth information and environmental impact information to the vegetation growth assessment module and the environmental impact analysis module, respectively.

[0035] The vegetation growth assessment module acquires vegetation growth information, including soil data and vegetation appearance images. Based on the fusion analysis of soil data and vegetation appearance images, it determines whether the vegetation growth status of the sub-management area is qualified. The specific analysis and judgment process is as follows:

[0036] The vegetation appearance images of the sub-management areas are acquired. After image preprocessing (using threshold segmentation, edge detection, region growing, or deep learning-based semantic segmentation methods), the vegetation areas in the garden images are separated from the image background, and feature extraction is performed. An abnormal feature recognition model that has been trained and optimized in advance is used to identify abnormal appearance features of the vegetation appearance images. Abnormal appearance features include color abnormality texture features, surface lesion features, and surface pest features. The abnormal feature recognition model is a deep learning model. During the training process, the model continuously learns different types of abnormal features to accurately identify abnormal features of new vegetation appearance images. Phenological calibration is performed for different seasons. For example, yellow leaves in autumn are not judged as nutrient deficiency and are not considered as abnormal appearance features.

[0037] Obtain the matching similarity between each abnormal appearance feature and the corresponding set abnormal appearance feature. If the matching similarity of any item is not within the preset matching similarity range, generate a vegetation appearance abnormal signal; otherwise, generate a vegetation appearance normal signal.

[0038] When acquiring soil data for sub-management areas using the same timestamp, special attention must be paid to spatiotemporal matching. Vegetation images reflect the cumulative state over the past 7–15 days, while soil data is collected in real time; therefore, time window alignment is necessary during data acquisition and analysis.

[0039] Soil data includes soil surface moisture, deep soil salinity, soil organic matter content, soil pH, and soil heavy metal content in the sub-management area. Multiple data points in the soil data are compared one by one with the preset standard range corresponding to the current growth stage of vegetation. When any soil data item exceeds the preset standard range, an abnormal soil index signal is generated; otherwise, a normal soil index signal is generated.

[0040] When both vegetation appearance and soil indicators are normal, the vegetation growth status of the sub-management area is deemed acceptable, and the acceptable sub-management area is marked as a normal management area. Conversely, when both are not normal, the vegetation growth status of the sub-management area is deemed unacceptable, and the unacceptable sub-management area is marked as an abnormal management area. Based on vegetation characteristics, the entire garden area is divided into multiple sub-management areas, and a zoned management model is adopted. For example, the soil moisture standard ranges for lawns and trees are different, and the acceptable standard requires a dynamic threshold rather than a fixed value, which enables targeted management.

[0041] The abnormal appearance features of the abnormal management area and soil data are input into a multi-parameter fusion verification model to obtain the abnormal vegetation growth type. The multi-parameter fusion verification model is also a deep learning model. It integrates deep learning and domain knowledge and achieves accurate diagnosis through spatiotemporally aligned multi-source data collaborative training. Each time a newly collected abnormal sample is injected (incremental learning keeps the model evolving), the accuracy of judging the abnormal vegetation growth type is further improved through three-dimensional monitoring and intelligent cross-validation of soil-vegetation appearance data.

[0042] The vegetation growth assessment module sends the number of the abnormal management area to the environmental impact analysis module.

[0043] Example 2: The environmental impact analysis module obtains environmental impact information of the abnormal management area through the parameter monitoring module. The environmental impact information includes historical meteorological data and facility data. Based on the historical meteorological data and facility data, the abnormal vegetation growth status of the abnormal management area is fused and analyzed to generate meteorological impact signals, facility impact signals or composite impact signals, and the impact signals are sent to the early warning unit.

[0044] The specific analysis process includes:

[0045] Historical meteorological data of the abnormal management area is obtained with the same timestamp. The historical meteorological data includes average rainfall, average solar radiation, average temperature, and average humidity. The preset suitable meteorological parameters for the current growth stage of vegetation in the abnormal management area are retrieved and stored in the management platform. When any item in the historical meteorological data is not within the range of the corresponding preset suitable meteorological parameters, a meteorological impact signal is generated.

[0046] The facility data of the abnormal management area is obtained with the same timestamp. The facility data includes average irrigation amount, average fertilizer amount, and average pesticide application amount. The preset suitable facility parameters for the current growth stage of the vegetation in the abnormal management area are retrieved and stored in the management platform. When any one of the facility data is not within the corresponding preset suitable facility parameters, a facility impact signal is generated.

[0047] When any one of the historical meteorological data is outside the corresponding preset suitable meteorological parameter range, and any one of the facility data is outside the corresponding preset suitable facility parameter range, a composite influence signal is generated.

[0048] Multi-dimensional root cause analysis is performed using spatiotemporal correlation. When a meteorological impact signal is generated, historical meteorological data that is outside the corresponding preset suitable meteorological parameter range is acquired and marked as abnormal meteorological data. The abnormal meteorological data is compared with the suitable meteorological parameter range. When the abnormal meteorological data is greater than the maximum value of the preset suitable meteorological parameter range, the meteorological abnormality fluctuation value of the abnormal meteorological data compared with the maximum value of the preset suitable meteorological parameter range is acquired. When the abnormal meteorological data is less than the minimum value of the preset suitable meteorological parameter range, the meteorological abnormality fluctuation value of the abnormal meteorological data compared with the minimum value of the preset suitable meteorological parameter range is acquired. For example, if the average rainfall is outside the preset suitable meteorological parameter range, the average rainfall is marked as abnormal meteorological data. When the average rainfall is greater than the maximum value of the preset suitable meteorological parameter range, it indicates that the rainfall is too large, and the meteorological abnormality fluctuation value corresponding to the rainfall is acquired. Conversely, when the rainfall is too small, the meteorological abnormality fluctuation value corresponding to the rainfall is acquired.

[0049] When the abnormal meteorological fluctuation value is greater than the corresponding preset abnormal meteorological fluctuation threshold, the abnormal meteorological data is determined to be the cause of the abnormal meteorological data, that is, the excessive average rainfall (rainstorm factor) is determined to be the cause of the abnormal meteorological data. When the abnormal meteorological fluctuation value is less than the corresponding preset abnormal meteorological fluctuation threshold, the abnormal meteorological data is determined to be the cause of the abnormal meteorological data, that is, the excessive average rainfall (drought factor) is determined to be the cause of the abnormal meteorological data.

[0050] When a facility impact signal is generated, facility data that is not within the corresponding preset suitable facility parameters is acquired and marked as abnormal facility data. The abnormal facility data is compared with the suitable facility parameter range, and the abnormally large or small facility fluctuation value is obtained. When the abnormally large facility fluctuation value is greater than the corresponding preset abnormally large facility fluctuation threshold, the abnormal facility data is determined to be the cause of the abnormality. When the abnormally small facility fluctuation value is less than the corresponding preset abnormally small facility fluctuation threshold, the abnormal facility data is determined to be the cause of the abnormality. The root cause determination result is sent to the management platform.

[0051] The management platform performs a weighted analysis of dominant factors based on the causes of abnormal weather conditions (both excessive and insufficient) and abnormal facility conditions (both excessive and insufficient). It generates corresponding level strategy information and transforms discrete events into causal chains through deep spatiotemporal correlation of meteorological and facility data. This enables a transformation in garden management from "passive disaster relief" to "proactive loss prevention." The corresponding level strategy information is then sent to the decision-making and execution module, which implements measures based on the strategy information.

[0052] In summary: Based on vegetation characteristics, the overall garden area is divided into multiple sub-management areas. Based on soil data and vegetation appearance images of a single area, the vegetation growth status of a local area is comprehensively analyzed to determine whether the vegetation growth status of the sub-management area is qualified. The abnormal appearance characteristics of abnormal management areas and soil data are input into a multi-parameter fusion verification model to obtain the abnormal vegetation growth type of abnormal management areas.

[0053] By combining historical data retrospective analysis, specifically: root cause determination is carried out based on historical meteorological data and facility data, so as to generate different strategy information according to the root cause determination results, and a closed-loop management system is constructed from anomaly identification → data attribution → decision execution. By utilizing diversified data and intelligent analysis technology, data-driven approach is used to achieve comprehensive monitoring, multi-dimensional analysis and accurate decision-making of environmental elements.

[0054] The above description is merely a preferred embodiment of the present invention; however, the scope of protection of the present invention is not limited thereto; any equivalent substitutions or modifications made by those skilled in the art within the technical scope disclosed in the present invention, based on the technical solution and its improved concept, should be covered within the scope of protection of the present invention.

Claims

1. A garden environment information management system based on diversified analysis, characterized by: It includes a management platform, a parameter monitoring module, a vegetation growth assessment module, an environmental impact analysis module, an early warning unit, and a decision execution module; The management platform divides the overall garden area into multiple sub-management areas based on vegetation characteristics, and assigns each sub-area a number; The parameter monitoring module is used to collect vegetation growth information and environmental impact information within the sub-management area, and send them to the vegetation growth assessment module and the environmental impact analysis module respectively. The vegetation growth assessment module acquires vegetation growth information, including soil data and vegetation appearance images. Based on the fusion analysis of soil data and vegetation appearance images, it determines whether the vegetation growth status of the sub-management area is qualified. Sub-management areas that are not qualified are marked as abnormal management areas. The abnormal vegetation growth type of the abnormal management area is obtained, and the number of the abnormal management area is sent to the environmental impact analysis module. The environmental impact analysis module obtains environmental impact information of the abnormal management area through the parameter monitoring module. The environmental impact information includes historical meteorological data and facility data. It performs integrated analysis on the abnormal growth status of vegetation to generate meteorological impact signals, facility impact signals, or composite impact signals. Based on the impact signals, it performs multi-dimensional root cause determination and sends the impact signals to the early warning unit and the root cause determination results to the management platform. The management platform generates strategy information based on the root cause determination results and sends it to the decision execution module.

2. The garden environment information management system based on multi-dimensional analysis according to claim 1, characterized in that: The vegetation growth assessment module determines whether the vegetation growth status is qualified by acquiring vegetation appearance images of the sub-management area, performing feature extraction after image preprocessing, using a pre-trained and optimized abnormal feature recognition model to identify abnormal appearance features of the vegetation appearance images, obtaining the matching similarity between each abnormal appearance feature and the corresponding set abnormal appearance feature, generating a vegetation appearance abnormal signal when the matching similarity of any item is not within the preset matching similarity range, otherwise generating a vegetation appearance normal signal. Soil data for sub-management areas are acquired using the same timestamp. Multiple data points in the soil data are compared one by one with preset standard ranges. When any soil data item exceeds the preset standard range, an abnormal soil index signal is generated; otherwise, a normal soil index signal is generated. When both vegetation appearance and soil indicators are normal, the vegetation growth status of the sub-management area is deemed acceptable; otherwise, the vegetation growth status of the sub-management area is deemed unacceptable.

3. The garden environment information management system based on multi-dimensional analysis according to claim 2, characterized in that: Abnormal appearance features include color abnormality texture features, surface lesion features, and surface pest features. The abnormal feature recognition model is a deep learning model. During the training process, the model continuously learns different types of abnormal features to accurately identify abnormal features in new vegetation appearance images. Soil data include soil surface moisture, deep soil salinity, soil organic matter content, soil pH, and soil heavy metal content in the sub-management areas.

4. The garden environment information management system based on multivariate analysis according to claim 3, characterized in that: When the vegetation growth status of a sub-management area is determined to be unqualified, the unqualified sub-management area is marked as an abnormal management area. The abnormal appearance characteristics of the abnormal management area and the soil data are input into a multi-parameter fusion verification model to obtain the abnormal vegetation growth type.

5. The garden environment information management system based on multi-dimensional analysis according to claim 4, characterized in that: The process of the environmental impact analysis module to perform integrated analysis on the abnormal growth status of vegetation in the abnormal management area includes: acquiring historical meteorological data of the abnormal management area with the same timestamp, retrieving the preset suitable meteorological parameters of the current growth stage of vegetation in the abnormal management area stored in the management platform, and generating a meteorological impact signal when any one of the historical meteorological data is not within the range of the corresponding preset suitable meteorological parameters. The system acquires facility data for abnormal management areas using the same timestamp, retrieves preset suitable facility parameters for the current growth stage of vegetation in the abnormal management areas stored in the management platform, and generates a facility impact signal when any one of the facility data is outside the corresponding preset suitable facility parameters; it also generates a composite impact signal when any one of the historical meteorological data is outside the corresponding preset suitable meteorological parameters and any one of the facility data is outside the corresponding preset suitable facility parameters.

6. The garden environment information management system based on multivariate analysis according to claim 5, characterized in that: When a meteorological impact signal is generated, historical meteorological data that is outside the corresponding preset suitable meteorological parameter range is acquired and marked as abnormal meteorological data. The abnormal meteorological data is compared with the suitable meteorological parameter range, and the abnormally large or small meteorological fluctuation value is obtained respectively. When the abnormally large meteorological fluctuation value is greater than the corresponding preset abnormally large meteorological fluctuation threshold, the abnormal meteorological data is determined to be the cause of the abnormally large meteorological data. When the abnormally small meteorological fluctuation value is less than the corresponding preset abnormally small meteorological fluctuation threshold, the abnormal meteorological data is determined to be the cause of the abnormally small meteorological data.

7. The garden environment information management system based on multivariate analysis according to claim 5, characterized in that: When a facility impact signal is generated, facility data that is not within the corresponding preset suitable facility parameters is acquired and marked as abnormal facility data. The abnormal facility data is compared with the suitable facility parameter range, and the abnormally large or small facility fluctuation value is obtained. When the abnormally large facility fluctuation value is greater than the corresponding preset abnormally large facility fluctuation threshold, the abnormal facility data is determined to be the cause of the abnormality. When the abnormally small facility fluctuation value is less than the corresponding preset abnormally small facility fluctuation threshold, the abnormal facility data is determined to be the cause of the abnormality.

8. The garden environment information management system based on multivariate analysis according to claim 7, characterized in that: The management platform generates corresponding levels of strategy information based on the causes of meteorological anomalies (both excessive and insufficient) and facility anomalies (both excessive and insufficient).

Images