Garden landscape plant health assessment method

Abstract

The invention discloses a garden landscape plant health assessment method, and belongs to the technical field of resource assessment. Comprising the following steps: step 1, constructing a time sequence database of environmental factors and plant growth states; step 2, according to different plant species and growth stages thereof, dynamically calculating and generating a health reference value adapted to the current environmental condition; step 3, in the optimal evaluation time window, performing preliminary evaluation on the plant health condition, extracting original health indexes, and performing correction in combination with an environment disturbance compensation factor; step 4, performing double comparative analysis on the corrected health assessment result, eliminating false anomalies caused by environmental fluctuation, and calculating an anomaly confirmation credibility index; step 5, grading the plant health conditions, visually displaying the plant health conditions in a thermodynamic diagram form, and marking the environmental interference degree; and step 6, based on an evaluation result, automatically generating a targeted maintenance suggestion, and designing a corresponding maintenance effect verification scheme.

Description

Technical Field

[0001] The present application relates to the technical field of resource assessment, and more specifically, to a method for assessing the health of landscape plants. Background Art

[0002] With the continuous advancement of urbanization, landscaping plays an increasingly important role in the urban ecological environment. Landscape plants not only have functions such as beautifying the environment, improving air quality, and regulating the climate, but also are an important part of improving the quality of the urban ecological environment and promoting the physical and mental health of humans. Therefore, how to ensure the healthy growth of garden plants has become an important task in modern garden management.

[0003] However, most traditional methods for assessing the health of garden plants rely on manual inspections and empirical judgments, which are not only inefficient but also easily affected by subjective factors, making it difficult to achieve efficient and accurate health assessments. At the same time, the growth status of plants is strongly affected by environmental factors such as temperature, humidity, light, and soil conditions. Changes in these factors can cause fluctuations in the health status of plants. Therefore, how to scientifically assess plants under different environmental conditions has become a challenge.

[0004] Currently, with the development of sensing technology, Internet of Things technology, and data analysis technology, the assessment of the health of garden plants is gradually moving towards the direction of intelligence and dataization. By collecting multi-dimensional data on the plant growth environment in real time and combining it with the physiological health indicators of plants, the health status of plants can be more accurately assessed. However, there are still some problems in the existing technology, such as the impact of environmental disturbances on the health assessment results has not been effectively compensated, the accuracy and reliability of the assessment results are difficult to guarantee, and the degree of automation of plant health management is still low.

[0005] In summary, how to accurately and real-time assess the health status of landscape plants based on the dynamic data of environmental changes and plant growth status, and generate scientific maintenance suggestions on this basis, has become a technical problem that urgently needs to be solved. Summary of the Invention

[0006] In order to overcome a series of defects existing in the prior art, the purpose of the present application is to provide a method for assessing the health of landscape plants in view of the above problems, including the following steps:

[0007] Step 1, collect environmental parameters and plant health characterization data in real time, and construct a time series database of environmental factors and plant growth status;

[0008] Step 2, dynamically calculate and generate a health benchmark value adapted to the current environmental conditions according to different plant species and their growth stages;

[0009] Step 3, within the optimal evaluation time window, conduct a preliminary assessment of the plant health status, extract the original health indicators, and correct them by combining the environmental perturbation compensation factors;

[0010] Step 4, conduct a double comparison analysis on the corrected health assessment results, eliminate the false anomalies caused by environmental fluctuations, and calculate the anomaly confirmation credibility index;

[0011] Step 5, classify the plant health status and visually display it in the form of a heat map, while marking the degree of environmental interference;

[0012] Step 6, based on the evaluation results, automatically generate targeted maintenance suggestions and design corresponding maintenance effect verification plans.

[0013] Furthermore, the establishment of the time-correlated database includes the following steps:

[0014] Set up micro-meteorological monitoring points at different spatial heights and soil depths to collect temperature, humidity, light intensity, carbon dioxide concentration, soil pH value, soil moisture and nutrient element content parameters in real time; at the same time, deploy physiological state monitoring devices in key plant areas to collect chlorophyll content, water status, photosynthetic efficiency and growth and development indicators to comprehensively characterize the plant health status;

[0015] Preprocess and screen the original data at the edge computing node, and realize the real-time data transmission back through the low-power wide-area network;

[0016] Formulate the data collection specifications for plant growth performance, combine the automated image collection equipment and manual inspection methods, regularly record the plant morphological characteristics, growth rate, phenological period, photosynthesis efficiency and pest and disease occurrence, and construct a quantitative index system for plant health status;

[0017] Adopt the time series database technology to store the corresponding relationship between environmental parameters and plant growth performance, and establish an efficient data index and query optimization mechanism to achieve fast retrieval and correlation analysis.

[0018] Furthermore, the micro-meteorological monitoring points are set at 3 height levels in the vertical direction, namely the ground surface, 1.5 meters and the top of the canopy, and the soil depth monitoring points are set at 3 levels, namely 10 cm, 30 cm and 50 cm. Each monitoring point is equipped with temperature and humidity sensors, light sensors, gas sensors and multi-parameter soil sensors. The sampling frequency is once every 10 minutes for temperature and humidity, once every 15 minutes for light intensity, once every 30 minutes for gas concentration, and once every 60 minutes for soil parameters;

[0019] The physiological status monitoring device includes: a portable chlorophyll meter with a measurement range of 0 - 99.9 SPAD values and an accuracy of ±1.5 SPAD values; a plant water potential meter with a measurement range of -0.1 to -4.0 MPa and an accuracy of ±0.2 MPa; a photosynthesis meter whose measurement parameters include photosynthetic rate and stomatal conductance; and a plant nutrient status analyzer based on near-infrared spectroscopy technology;

[0020] The automated image acquisition equipment includes a fixed high-definition camera, a mobile multi-spectral imaging device, and a timed cruise drone. Among them: the fixed camera covers the key area with a 120° view angle and takes pictures every 4 hours; the multi-spectral imaging device collects visible light, near-infrared, and red-edge band images and completes a scan of the key area every 24 hours; the timed cruise drone is equipped with a 16-megapixel camera and a multi-spectral sensor, with a flight altitude of 40 - 80 meters. It conducts a comprehensive aerial survey of the entire park every 7 days to form an orthophoto map with a resolution of not less than 8 cm / pixel, and dynamically adjusts the shooting frequency according to weather conditions and seasonal characteristics.

[0021] Further, step 2 includes the following steps:

[0022] Collect and organize the physiological and ecological characteristics, suitable growth conditions, sensitive environmental factors, and growth and development laws of different garden plants, group the plants according to ecological habits, stress resistance, and growth characteristics to form a structured knowledge base;

[0023] Based on accumulated temperature, light duration, and phenological observation data, combined with image recognition technology, automatically determine the current growth and development stage of various plants, and formulate differentiated health assessment criteria for different stages;

[0024] Comprehensively analyze the deviation degree between the recent environmental conditions and the optimal growth conditions of plants, calculate the temperature suitability, water suitability, and light suitability respectively, and form a comprehensive environmental adaptability index through weighted fusion to quantify the impact of the current environment on plant growth;

[0025] Combined with historical data, analyze the performance of similar plants under similar environmental conditions and growth stages, and dynamically adjust the threshold range of health assessment according to the environmental adaptability index to generate a health benchmark value suitable for the current conditions;

[0026] For plant communities in different microclimate regions, terrain conditions, and planting densities, establish a spatially differentiated health benchmark value system, and dynamically adjust the fluctuation range of the benchmark value according to seasonal changes and climate fluctuation factors;

[0027] By continuously comparing the differences between the predicted benchmark value and the actual plant performance, automatically adjust the calculation parameters to achieve the adaptive optimization and accuracy improvement of the health benchmark value calculation method.

[0028] Furthermore, the comprehensive environmental adaptability index calculation takes into account four ability indicators: the expansion rate of temperature tolerance range, the recovery rate of water stress, the adjustment efficiency of light adaptability, and the enhancement degree of pathogen resistance. The weights of each indicator are dynamically adjusted according to plant types and seasons. An environmental adaptability index ranging from 0 to 100 is calculated through weighted average. An index increase of more than 15% is regarded as a significant improvement in adaptability, an increase of 5%-15% is regarded as an improvement in adaptability, and a decrease of less than 5% is regarded as no obvious change in adaptability.

[0029] Furthermore, step 3 includes the following steps:

[0030] Based on the analysis of plant physiological activity laws and environmental stability, identify the time period with the smallest fluctuations in plant physiological indicators that can fully characterize the healthy state, and determine the optimal evaluation time window;

[0031] Within the determined optimal evaluation time window, comprehensively collect data on plant morphological indicators, physiological indicators, and pest and disease conditions, calculate the deviation degree of the dispersion of each indicator from the reference value, and generate an original health index to preliminarily quantify the plant health status;

[0032] Evaluate the environmental fluctuations within 24 - 72 hours before the assessment, quantify their temporary impact on the current physiological state of the plant, and calculate the short-term environmental perturbation compensation coefficient;

[0033] Evaluate the inductive effect of environmental trend changes in the recent 7 - 30 days on the adaptive physiological regulation of the plant, and generate a medium-term environmental compensation factor;

[0034] Quantify the impact of seasonal environmental transitions in the recent 1 - 3 months on the plant growth strategy and resource allocation, distinguish the changes caused by adaptive physiological adjustments of the plant from the degradation of non-healthy states, and calculate the long-term environmental adjustment factor;

[0035] Based on the analysis results of short-term, medium-term, and long-term environmental changes, comprehensively calculate the environmental perturbation compensation factor and apply it to the correction of the original health index to eliminate the interference of environmental factors at different time scales, so as to obtain a more environmentally neutral assessment result of the true health state of the plant.

[0036] Furthermore, step 4 includes the following steps:

[0037] Compare the current corrected plant health assessment result with the historical data of the same plant or plant group, analyze long-term trend changes, seasonal fluctuations, and non-periodic anomalies, judge whether the current state is within the normal change range, and screen out potential abnormal points exceeding the expected change rate;

[0038] Compare the health assessment data of the target plants with the data of plants of the same species at the same time point but different spatial locations, analyze the spatial distribution characteristics of the health status, identify local anomalies and group common changes, determine the significance level of spatial anomalies, and distinguish individual-specific problems from regional changes caused by the environment;

[0039] For the health indicators marked as abnormal, analyze their correlation with environmental factors, calculate the correlation coefficient and lag effect between health anomalies and changes in various environmental factors, and determine whether the anomalies are caused by uncompensated environmental disturbances, so as to filter out false anomalies caused by short-term environmental fluctuations;

[0040] Based on the analysis results of anomaly significance in the time dimension, anomaly significance in the spatial dimension, and environmental relevance, use the multi-factor weighted scoring method to calculate the comprehensive confirmation score of the anomaly, and set multi-level threshold criteria to classify the plant health anomalies, and distinguish serious anomalies that require immediate intervention from minor fluctuations that only need continuous observation;

[0041] Based on factors such as data quality score, sample sufficiency, environmental condition similarity, and historical diagnosis accuracy, assign a credibility index to each confirmed anomaly to provide decision-making reference and mark uncertain anomaly points that need further verification;

[0042] Conduct a classification and attribution analysis of the confirmed real anomalies, and combine plant physiological characteristics, environmental conditions, and historical cases to infer possible causes of the disease, and generate a comprehensive diagnosis report including anomaly description, confirmation credibility, possible causes, and severity.

[0043] Furthermore, step 5 includes the following steps:

[0044] Based on the calibrated health indicators, combined with the physiological and ecological characteristics of plants, seasonal growth patterns, and plant flora classification, formulate health assessment criteria applicable to different species;

[0045] Analyze the potential impact of environmental suitability on the long-term health status of plants, identify the key environmental factors restricting growth, and calculate the environmental support index;

[0046] Connect the plant health assessment results with the geographic information system, establish the spatial mapping relationship between plant location coordinates and health data, and combine with the base map data to achieve multi-scale health management from individual plants to communities;

[0047] Use the spatial interpolation algorithm to generate a continuous and smooth health distribution heat map, and adjust the credibility expression of the heat map according to the observation density to visually present the spatial distribution characteristics of the health status of landscape plants;

[0048] Based on the environmental disturbance compensation factor, calculate the degree of interference of environmental fluctuations in each area on the plant health assessment, so as to overlay environmental disturbance isolines or area identifiers on the heat map.

[0049] Further, step 6 includes the following steps:

[0050] Based on the plant health status grading results, combined with the analysis of plant physiological requirements, pest and disease diagnosis, and environmental limiting factors, call the set of maintenance measures in the expert knowledge base and automatically match the optimal maintenance plan;

[0051] Determine the key evaluation indicators and expected improvement goals for each maintenance measure, formulate sampling strategies and detection frequencies, set the best recheck time window, and establish a control analysis framework before and after maintenance;

[0052] Calculate the improvement amplitude, recovery rate, and health recovery rate of health indicators, analyze the change patterns of various physiological indicators, and evaluate the targeted effects of maintenance measures;

[0053] Track the change in the sensitivity of plants to environmental stress factors after maintenance, calculate the environmental adaptability index, evaluate the benefits of maintenance measures in enhancing plant stress resistance and long-term stability, and identify the key factors affecting the improvement of adaptability;

[0054] Compare the maintenance results with the initial expectations, analyze the reasons for the deviation, identify the successful experiences and deficiencies of the maintenance measures, and update the measure effectiveness scores in the expert knowledge base;

[0055] Apply the maintenance practice feedback data to the calibration and optimization of the health assessment model, adjust the environmental disturbance compensation coefficient, update the health benchmark value calculation method, improve the abnormal confirmation algorithm, regularly evaluate the overall performance and perform iterative upgrades to achieve continuous improvement of the health management of landscape plants.

[0056] Further, the calculation formula for the health index improvement rate is: (post-maintenance index value - pre-maintenance index value) / pre-maintenance index value × 100%; the calculation formula for the recovery rate is: health index improvement rate / intervention days; the stability index is evaluated by the coefficient of variation of the observed values for 5 consecutive days. A coefficient of variation below 20% indicates recovery to stability, above 35% indicates instability, and between 20% - 35% indicates a state to be observed.

[0057] Compared with the prior art, the present application has the following beneficial effects:

[0058] The present application combines multi-dimensional data collection, dynamic health benchmark value generation, environmental disturbance compensation, and double contrast analysis to achieve accurate assessment of the health status of landscape plants and automatic generation of maintenance suggestions, improving the intelligence and adaptability of plant health management. Description of the Drawings

[0059] Figure 1 Schematic flowchart of a method for evaluating the health of landscape plants disclosed in an embodiment of the present application. Detailed implementation manners

[0060] To make the objectives, technical solutions, and advantages of the implementation of the present invention clearer, the technical solutions in the embodiments of the present invention will be described in more detail below with reference to the accompanying drawings in the embodiments of the present invention. In the drawings, the same or similar reference numerals denote the same or similar elements or elements with the same or similar functions from beginning to end. The described embodiments are some, but not all, of the embodiments of the present invention.

[0061] All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0062] The embodiments described below with reference to the accompanying drawings and directional terms are exemplary only and are intended to explain the present invention and should not be construed as a limitation to the present invention.

[0063] As Figure 1 shown, a method for evaluating the health of landscape plants includes the following steps:

[0064] Step 1: Collect environmental parameters and plant health characterization data in real time, and construct a time series database of environmental factors and plant growth states.

[0065] Step 2: Dynamically calculate and generate a health benchmark value adapted to the current environmental conditions according to different plant species and their growth stages.

[0066] Step 3: In the optimal evaluation time window, preliminarily evaluate the plant health status, extract the original health indicators, and correct them in combination with the environmental disturbance compensation factor.

[0067] Step 4: Conduct a double comparative analysis on the corrected health evaluation results, eliminate the false anomalies caused by environmental fluctuations, and calculate the anomaly confirmation credibility index.

[0068] Step 5: Classify the plant health status, visually display it in the form of a heat map, and mark the degree of environmental interference.

[0069] Step 6: Automatically generate targeted maintenance suggestions based on the evaluation results, and design corresponding maintenance effect verification schemes.

[0070] In summary, the method for evaluating the health of landscape plants realizes the accurate perception and scientific evaluation of the plant health status by integrating multi-source data collection, dynamic benchmark modeling, perturbation correction mechanism, dual anomaly recognition and analysis, visualization display, and intelligent decision support. This method can not only reflect the growth status of plants under specific environmental conditions in real time, but also effectively eliminate misjudgments caused by climate or temporary factors, improving the accuracy and credibility of the evaluation results. Through health grading and heat map display, garden managers can intuitively grasp the overall health situation, and based on the personalized maintenance suggestions generated by the system, achieve targeted intervention and scientific maintenance, thereby improving the maintenance efficiency of garden plants and the sustainability of ecological landscapes.

[0071] Further, the establishment of the time-correlated database includes the following steps:

[0072] Set up micro-meteorological monitoring points at different spatial heights and soil depths to collect parameters such as temperature, humidity, light intensity, carbon dioxide concentration, soil pH value, soil moisture, and nutrient element content in real time; at the same time, deploy physiological state monitoring devices in key plant areas to collect chlorophyll content, water status, photosynthetic efficiency, and growth and development indicators to comprehensively characterize the plant health status;

[0073] Preprocess and screen the original data at the edge computing node, and realize the real-time data transmission through the low-power wide-area network;

[0074] Formulate the data collection specifications for plant growth performance, combine automated image collection equipment and manual inspection methods, regularly record plant morphological characteristics, growth rate, phenological period, photosynthesis efficiency, and the occurrence of pests and diseases, and construct a quantitative index system for plant health status;

[0075] Adopt time-series database technology to store the corresponding relationship between environmental parameters and plant growth performance, and establish an efficient data index and query optimization mechanism to achieve rapid retrieval and correlation analysis.

[0076] Further, the micro-meteorological monitoring points are set at 3 height levels in the vertical direction, namely the ground surface, 1.5 meters, and the top of the canopy. The soil depth monitoring points are set at 3 levels, namely 10 cm, 30 cm, and 50 cm. Each monitoring point is equipped with temperature and humidity sensors, light sensors, gas sensors, and multi-parameter soil sensors. The sampling frequency is once every 10 minutes for temperature and humidity, once every 15 minutes for light intensity, once every 30 minutes for gas concentration, and once every 60 minutes for soil parameters;

[0077] The physiological status monitoring device includes: a portable chlorophyll meter with a measurement range of 0 - 99.9 SPAD values and an accuracy of ±1.5 SPAD values; a plant water potential meter with a measurement range of -0.1 to -4.0 MPa and an accuracy of ±0.2 MPa; a photosynthesis meter whose measurement parameters include photosynthetic rate and stomatal conductance; and a plant nutrient status analyzer based on near-infrared spectroscopy technology;

[0078] The automated image acquisition equipment includes a fixed high-definition camera, a mobile multi-spectral imaging device, and a timed cruise drone. Among them: the fixed camera covers the key area with a 120° view angle and takes pictures every 4 hours; the multi-spectral imaging device collects visible light, near-infrared, and red-edge band images and completes a scan of the key area every 24 hours; the timed cruise drone is equipped with a 16-megapixel camera and a multi-spectral sensor, with a flight altitude of 40 - 80 meters. It conducts a comprehensive aerial survey of the entire park every 7 days to form an orthophoto map with a resolution of no less than 8 cm / pixel, and dynamically adjusts the shooting frequency according to weather conditions and seasonal characteristics.

[0079] Further, step 2 includes the following steps:

[0080] Collect and sort out the physiological and ecological characteristics, suitable growth conditions, sensitive environmental factors, and growth and development laws of different garden plants, group the plants according to ecological habits, stress resistance, and growth characteristics to form a structured knowledge base;

[0081] Based on accumulated temperature, light duration, and phenological observation data, combined with image recognition technology, automatically determine the current growth and development stage of various plants, and formulate differentiated health assessment criteria for different stages;

[0082] Comprehensively analyze the deviation degree between the recent environmental conditions and the optimal growth conditions of plants, calculate the temperature suitability, water suitability, and light suitability respectively, and form a comprehensive environmental adaptability index through weighted fusion to quantify the impact of the current environment on plant growth;

[0083] Combined with historical data, analyze the performance of similar plants under similar environmental conditions and growth stages, and dynamically adjust the threshold range of health assessment according to the environmental adaptability index to generate a health benchmark value suitable for the current conditions;

[0084] For plant communities in different microclimate regions, terrain conditions, and planting densities, establish a spatially differentiated health benchmark value system, and dynamically adjust the fluctuation range of the benchmark value according to seasonal changes and climate fluctuation factors;

[0085] By continuously comparing the differences between the predicted benchmark value and the actual plant performance, automatically adjust the calculation parameters to achieve the adaptive optimization and accuracy improvement of the health benchmark value calculation method.

[0086] In summary, by constructing a structured plant knowledge base and combining image recognition with phenological observations, the precise recognition of plant growth stages and the dynamic formulation of differential health assessment criteria have been achieved, significantly enhancing the adaptability of the health assessment model to species differences and growth cycles. By calculating the comprehensive environmental adaptability index using key ecological factors such as accumulated temperature, light, and moisture, the matching degree between plants and current environmental conditions has been quantified, providing a scientific basis for health assessment. At the same time, by combining historical data analysis with spatial zoning characteristics, the assessment thresholds and the fluctuation range of benchmark values are dynamically adjusted to ensure the effectiveness and stability of the assessment criteria under diverse habitats and climate backgrounds. Through continuous comparison between predictions and actual performance and the parameter self-adaptive adjustment mechanism, the intelligent optimization and precision iteration of the health benchmark value calculation process have been realized, improving the environmental perception ability and dynamic response ability of the overall assessment system.

[0087] Furthermore, the calculation of the comprehensive environmental adaptability index considers four ability indicators: the temperature tolerance range expansion rate, the water stress recovery rate, the light adaptability adjustment efficiency, and the pathogen resistance enhancement degree. The weights of each indicator are dynamically adjusted according to plant types and seasons. The environmental adaptability index ranging from 0 to 100 is calculated through weighted average. An increase in the index exceeding 15% is regarded as a significant improvement in adaptability, an increase of 5% - 15% is regarded as an improvement in adaptability, and a decrease below 5% is regarded as no obvious change in adaptability.

[0088] Furthermore, the generation of the health benchmark value suitable for current conditions includes the following steps:

[0089] Revise the historical health range in combination with the environmental adaptability index to reflect the current environmental impact. The adjusted threshold is calculated as follows: where, H adj (t) is the adjusted health assessment threshold, the health range dynamically revised based on historical health data and combined with the current environmental adaptability index; H hist (t) is the historical health assessment value, representing the historical health assessment mean of similar plants under similar environmental conditions and growth stages; β is the adaptability adjustment coefficient, reflecting the influence degree of the environmental adaptability index EAI(t) on the plant health assessment threshold; EAI max is the maximum value of the environmental adaptability index, representing the limit adaptability value of environmental conditions, used to normalize the environmental adaptability index; EAI(t) is the environmental adaptability index at the current time t, representing the influence degree of the current environmental conditions on plant health;

[0090] Based on the adjusted health range, calculate the health benchmark value under current conditions: where, H base (t) is the health benchmark value under current conditions, the standard health value dynamically calculated based on historical data and the environmental adaptability index; Hmax (t) is the maximum health assessment value in historical data, representing the highest threshold of plant health under specific environmental conditions; H min (t) is the minimum health assessment value in historical data, representing the lowest threshold of plant health under specific environmental conditions.

[0091] Furthermore, step 3 includes the following steps:

[0092] Based on the analysis of plant physiological activity laws and environmental stability, identify the time period with the smallest fluctuations in plant physiological indicators and that can fully characterize the health status, and determine the optimal assessment time window;

[0093] Within the determined optimal assessment time window, comprehensively collect data on plant morphological indicators, physiological indicators, and pest and disease conditions, calculate the deviation degree of the dispersion of each indicator from the reference value, and generate the original health index to preliminarily quantify the plant health status;

[0094] Evaluate the environmental fluctuations within the previous 24 - 72 hours, quantify their temporary impact on the current physiological state of the plant, and calculate the short - term environmental perturbation compensation coefficient;

[0095] Evaluate the induction effect of the environmental trend changes in the recent 7 - 30 days on the adaptive physiological regulation of the plant, and generate the medium - term environmental compensation factor;

[0096] Quantify the impact of seasonal environmental changes in the recent 1 - 3 months on the plant growth strategy and resource allocation, distinguish the changes caused by the adaptive physiological adjustment of the plant from the degradation of the non - healthy state, and calculate the long - term environmental adjustment factor;

[0097] Based on the analysis results of short - term, medium - term, and long - term environmental changes, comprehensively calculate the environmental perturbation compensation factor and apply it to the correction of the original health index to eliminate the interference of environmental factors at different time scales, so as to obtain a more environmentally neutral assessment result of the true health status of the plant.

[0098] In summary, by scientifically defining the optimal time window for plant health assessment and ensuring the representativeness and stability of the collected data, the accuracy of health assessment is improved. On this basis, an original health index is constructed by integrating morphological, physiological, and pest and disease indicators to achieve a preliminary quantitative assessment of the plant health status. Further, through hierarchical analysis of environmental changes within short-term (24 - 72 hours), medium-term (7 - 30 days), and long-term (1 - 3 months), the impacts of instantaneous perturbations, physiological adaptation inductions, and seasonal transitions on plant health indicators are respectively identified, and the corresponding perturbation compensation factors are accurately calculated. Finally, by correcting the original health index through comprehensive environmental perturbation compensation, the interference signals caused by non-health changes are effectively eliminated, significantly improving the robustness and neutrality of the assessment results to environmental changes, ensuring that the assessment reflects the plant's own health status rather than the artifacts of external perturbations, thereby providing a more real and reliable data basis for subsequent hierarchical diagnosis and maintenance decision-making.

[0099] Furthermore, based on the analysis of plant physiological activity laws and environmental stability, identify the time period with the smallest fluctuations in plant physiological indicators and that can fully represent the health status, and determine the optimal assessment time window, including the following steps:

[0100] Under the preset window length L, generate multiple candidate time windows W = [t, t + L] by the sliding window method to ensure coverage of the critical periods of plant growth;

[0101] Within each window, calculate the mean μ of each physiological indicator W,i and variance Meanwhile, evaluate the environmental stability and obtain the mean μ of the environmental data W(e) and variance

[0102] Construct an evaluation function to balance the fluctuations of plant indicators and environmental stability and quantify the evaluation effects of each candidate window, where n is the number of plant physiological indicators, i.e., the number of physiological parameters selected during the analysis; α represents the weight coefficient of environmental stability in the comprehensive evaluation function, which is used to balance the impacts of plant physiological indicator fluctuations and environmental factors;

[0103] Select the time window W that minimizes J(W) * = argmin W J(W) as the optimal assessment window, and verify its representativeness of the plant health status by comparing with historical data, where W * represents the finally determined optimal assessment time window, i.e., the time interval that minimizes the comprehensive evaluation function J(W), reflecting the most stable physiological state of the plant and the period that best represents the health status.

[0104] Furthermore, step 4 includes the following steps:

[0105] Compare the currently corrected plant health assessment results with the historical data of the same plant or plant population, analyze long-term trend changes, seasonal fluctuations, and aperiodic anomalies, determine whether the current state is within the normal change range, and screen out potential anomaly points that exceed the expected change rate;

[0106] Compare the health assessment data of the target plant with the data of plants of the same species but different spatial locations at the same time point, analyze the spatial distribution characteristics of the health status, identify local anomalies and common changes in the population, determine the significance level of spatial anomalies, and distinguish individual-specific problems from regional changes caused by the environment;

[0107] For the health indicators marked as abnormal, analyze their correlation with environmental factors, calculate the correlation coefficient and lag effect between health anomalies and changes in each environmental factor, and determine whether the anomaly is caused by uncompensated environmental disturbances, so as to filter out false anomalies caused by short-term environmental fluctuations;

[0108] Based on the analysis results of anomaly significance in the time dimension, anomaly significance in the spatial dimension, and environmental relevance, use the multi-factor weighted scoring method to calculate the comprehensive confirmation score of the anomaly, set multi-level threshold criteria, classify the plant health anomalies, and distinguish serious anomalies that require immediate intervention from minor fluctuations that only require continuous observation;

[0109] Based on factors such as data quality score, sample sufficiency, environmental condition similarity, and historical diagnosis accuracy, assign a credibility index to each confirmed anomaly to provide decision-making reference and mark uncertain anomaly points that need further verification;

[0110] Conduct a classification and attribution analysis of the confirmed real anomalies, combine plant physiological characteristics, environmental conditions, and historical cases, infer possible causes of the disease, and generate a comprehensive diagnosis report including anomaly description, confirmation credibility, possible causes, and severity.

[0111] In summary, by introducing a dual contrast analysis mechanism of time and space, the accuracy and reliability of plant health anomaly identification have been significantly improved. First, the current evaluation results are compared with historical data in terms of trends to identify non-periodic anomalies and the rate of anomaly change, enabling a rational judgment of the long-term health dynamics of individual plants. Second, horizontal comparison is carried out by combining the health data of similar plants at the same time point but different spatial positions to reveal the spatial differences in the health status of the local and the population, and to clarify whether the anomalies have regional commonalities. Subsequently, by analyzing the correlation and lag effect between health indicators and various environmental factors, pseudo-anomalies caused by short-term environmental fluctuations are further excluded, effectively improving the accuracy of anomaly screening. Based on the multi-dimensional analysis results of time, space, and environment, the multi-factor weighting method is used to comprehensively score and classify each anomaly situation to clarify the severity of the anomaly and the intervention priority. At the same time, a credibility index mechanism is introduced, considering data quality and historical diagnosis accuracy, providing a quantitative reference basis for management decisions. Finally, a detailed diagnostic report is output based on the classification attribution model, covering the causes of anomalies, risk levels, and credibility assessments, providing scientific support for subsequent precise maintenance, and significantly enhancing the intelligent diagnosis and intervention capabilities of garden health monitoring.

[0112] Furthermore, potential anomaly points exceeding the expected rate of change are screened out through the following steps:

[0113] Using historical plant health data H history (t), the long-term trend T history (t) is extracted through regression analysis or moving average method;

[0114] The seasonal component S history (t) = Seasonality(H history (t)) is extracted from the historical data using the seasonal decomposition method to clarify the impact of periodic changes on health indicators, where S history (t) is the historical seasonal component, representing the seasonal fluctuation part in the historical plant health assessment data before time t; Seasonality represents the seasonal fluctuation component in the historical health data, that is, the fluctuation pattern within a specific period;

[0115] The residual in the historical data is obtained through the formula R history (t) = H history (t) - T history (t) - S history (t) to clarify the random anomalies other than trends and seasonality. Among them, R history (t) represents the residual of the historical data at time point t, the part after removing trends and seasonal fluctuations, reflecting non-periodic anomalies;

[0116] Calculate the deviation of the current corrected health index from the long-term trend and seasonal fluctuations, ΔH(t) = H current (t) - T history (t) - S history (t), and determine whether it exceeds the preset normal range threshold δ threshold , where ΔH(t) represents the deviation of the current corrected plant health assessment result from the historical trend and seasonal fluctuations; H current (t) represents the plant health assessment result at the current time t, which has been corrected and reflects the latest plant state;

[0117] By comparing the change rate of the health index at the current and previous times with the change rate of the historical trend, calculate Screen out potential outliers exceeding the threshold δ rate , where Rate of Change(t) represents the difference in the change rate of the plant health assessment result between the current time and the previous time; Δt represents the time interval; H current (t - 1) represents the current plant health assessment result at the previous time t - 1; T history (t - 1) represents the historical health trend value at the previous time t - 1, that is, the historical health data after trend extraction.

[0118] Furthermore, step 5 includes the following steps:

[0119] Based on the calibrated health index, combined with the physiological and ecological characteristics of the plant, seasonal growth patterns, and plant flora classification, formulate health assessment criteria applicable to different species;

[0120] Analyze the potential impact of environmental suitability on the long-term health status of plants, identify key environmental factors limiting growth, and calculate the environmental support index;

[0121] Connect the plant health assessment result with the geographic information system, establish the spatial mapping relationship between the plant location coordinates and health data, and combine with the base map data to achieve multi-scale health management from individual plants to communities;

[0122] Use the spatial interpolation algorithm to generate a continuous and smooth health distribution heat map, and adjust the credibility expression of the heat map according to the observation density to visually present the spatial distribution characteristics of the health status of landscape plants;

[0123] Based on the environmental disturbance compensation factor, calculate the interference degree of environmental fluctuations in each region on the plant health assessment, so as to overlay environmental interference contour lines or regional identifiers on the heat map.

[0124] In summary, by combining the calibrated plant health indicators with plant types, physiological and ecological characteristics, and seasonal growth patterns, a multi-species adapted health assessment standard system was constructed, achieving individualized and grouped accurate health identification. On this basis, the long-term impact of environmental suitability on plant growth was further analyzed, the constraints were identified and their intervention intensity was quantified, and an environmental support index was formed to reveal potential ecological restriction pressures. Subsequently, by deeply integrating the health assessment results with the geographic information system (GIS), a geographical mapping relationship between plant spatial distribution and health status was established, realizing multi-scale management capabilities from single plant positioning to community monitoring. With the help of the health heat map generated by the spatial interpolation algorithm, the health distribution situation can be intuitively displayed at the scale of the entire garden, and its credibility expression can be dynamically adjusted, making the assessment results more visual, continuous, and decision-making reference value. At the same time, the degree of interference calculated by the environmental disturbance compensation factor is visualized and superimposed on the heat map, making the spatial distribution of environmental interference clear at a glance, greatly enhancing the accuracy of health anomaly tracing and intervention area determination, and providing an efficient, intelligent, and visual support tool for garden maintenance and resource allocation.

[0125] Further, step 6 includes the following steps:

[0126] Based on the plant health status classification results, combined with plant physiological needs, pest and disease diagnosis and environmental limiting factor analysis, the maintenance measure set in the expert knowledge base is called and the optimal maintenance plan is automatically matched;

[0127] Determine key evaluation indicators and expected improvement goals for each maintenance measure, develop sampling strategies and testing frequencies, set optimal review time windows, and establish a pre- and post-maintenance comparison analysis framework;

[0128] Calculate the improvement range, recovery rate and health recovery rate of health indicators, analyze the change patterns of various physiological indicators, and evaluate the targeted effects of maintenance measures;

[0129] Track changes in plant sensitivity to environmental stress factors after maintenance, calculate environmental adaptability index, evaluate the effectiveness of maintenance measures in enhancing plant stress resistance and long-term stability, and identify key factors affecting adaptation improvement;

[0130] Compare maintenance results with initial expectations, analyze causes of deviations, identify successful experiences and shortcomings of maintenance measures, and update measure effectiveness scores in the expert knowledge base;

[0131] Apply maintenance practice feedback data to the calibration and optimization of the health assessment model, adjust the environmental disturbance compensation coefficient, update the health benchmark value calculation method, improve the anomaly confirmation algorithm, regularly evaluate the overall performance and perform iterative upgrades to achieve continuous improvement in garden plant health management.

[0132] In summary, by combining the results of plant health status grading with the physiological needs of plants, pest and disease diagnosis, and environmental limiting factors, an intelligent matching of automated maintenance plans is achieved, significantly enhancing the pertinence and scientificity of maintenance measures. Based on the expert knowledge base, the best maintenance measures are retrieved, and clear key evaluation indicators, expected improvement goals, and review mechanisms are set for each measure to construct a control analysis framework before and after maintenance, making it possible to quantitatively evaluate and dynamically monitor the maintenance effect. During the implementation process, the improvement amplitude and recovery rate of health indicators are accurately tracked, and the changing trends of various physiological indicators are evaluated to scientifically quantify the maintenance effectiveness. At the same time, the response sensitivity of plants to environmental stress after maintenance is dynamically monitored, and the environmental adaptability index is calculated to comprehensively evaluate the actual effect of maintenance on enhancing plant stress resistance and ecological adaptability, and to identify the key factors restricting the improvement of adaptability. In addition, the actual maintenance effect is compared and analyzed with the expected goals to identify the deficiencies in the measures, and the results are fed back to the expert knowledge base to continuously improve the effect scoring system of the maintenance measures, realizing the continuous accumulation and optimization of knowledge. Finally, all maintenance feedback data will be used in reverse to evaluate the optimization of the model, including the dynamic update of the health benchmark value calculation method, the adjustment of the environmental disturbance compensation coefficient, and the iterative optimization of the anomaly identification algorithm, promoting the adaptive evolution and continuous performance improvement of the entire plant health assessment and management system, and constructing a closed-loop and intelligent garden plant health management process.

[0133] Further, the calculation formula for the health indicator improvement rate is: (indicator value after maintenance - indicator value before maintenance) / indicator value before maintenance × 100%; the calculation formula for the recovery rate is: health indicator improvement rate / number of intervention days; the stability index is evaluated by the coefficient of variation of the observed values for 5 consecutive days. A coefficient of variation lower than 20% indicates stable recovery, higher than 35% indicates instability, and between 20% - 35% indicates a state to be observed.

[0134] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for evaluating the health of garden landscape plants, characterized in that, It includes the following steps: Step 1: Collect environmental parameters and plant health characterization data in real time, and construct a time series database of environmental factors and plant growth status; Step 2: Dynamically calculate and generate health benchmark values adapted to the current environmental conditions according to different plant species and their growth stages; Step 3: Conduct a preliminary assessment of the plant health status within the optimal evaluation time window, extract the original health indicators, and correct them in combination with the environmental perturbation compensation factor; Step 4: Conduct a double comparative analysis on the corrected health assessment results, eliminate the false anomalies caused by environmental fluctuations, and calculate the anomaly confirmation credibility index; Step 5: Classify the plant health status, visually display it in the form of a heat map, and mark the degree of environmental interference; Step 6: Automatically generate targeted maintenance suggestions based on the evaluation results, and design corresponding maintenance effect verification schemes.

2. The method for evaluating the health of a garden landscape plant according to claim 1, characterized in that The establishment of the time-correlated database includes the following steps: Set up micro-meteorological monitoring points at different spatial heights and soil depths to collect parameters such as temperature, humidity, light intensity, carbon dioxide concentration, soil pH value, soil moisture and nutrient element content in real time; at the same time, deploy physiological state monitoring devices in key plant areas to collect chlorophyll content, water status, photosynthetic efficiency and growth and development indicators to comprehensively characterize the plant health status; Preprocess and screen the original data at the edge computing node, and realize the real-time data transmission through the low-power wide area network; Formulate the data collection specifications for plant growth performance, combine the automated image collection equipment and manual inspection methods, regularly record the plant morphological characteristics, growth rate, phenological period, photosynthesis efficiency and pest and disease occurrence conditions, and construct a quantitative index system for plant health status; Adopt the time series database technology to store the corresponding relationship between environmental parameters and plant growth performance, and establish an efficient data index and query optimization mechanism to achieve fast retrieval and correlation analysis.

3. The method for evaluating the health of a landscape plant according to claim 2, characterized in that, The micro-meteorological monitoring points are set at 3 height levels in the vertical direction, namely the ground surface, 1.5 meters and the top of the canopy, and the soil depth monitoring points are set at 3 levels, namely 10 cm, 30 cm and 50 cm. Each monitoring point is equipped with temperature and humidity sensors, light sensors, gas sensors and multi-parameter soil sensors. The sampling frequency is once every 10 minutes for temperature and humidity, once every 15 minutes for light intensity, once every 30 minutes for gas concentration, and once every 60 minutes for soil parameters; The physiological state monitoring device includes: a portable chlorophyll meter with a measurement range of 0 - 99.9 SPAD value and an accuracy of ±1.5 SPAD value; a plant water potential meter with a measurement range of -0.1 to -4.0 MPa and an accuracy of ±0.2 MPa; a photosynthesis meter with measured parameters including photosynthetic rate and stomatal conductance; and a plant nutrient status analyzer based on near-infrared spectroscopy technology; The automated image acquisition device includes a fixed high-definition camera, a mobile multi-spectral imaging device, and a timed cruise drone, where: The fixed camera covers key areas with a 120° viewing angle and takes pictures every 4 hours; The multi-spectral imaging device collects visible light, near-infrared, and red-edge band images and completes a scan of key areas every 24 hours; The timed cruise drone is equipped with a 16-megapixel camera and a multi-spectral sensor, flies at an altitude of 40 - 80 meters, conducts a comprehensive aerial survey of the entire park every 7 days, forms an orthophoto map with a resolution of not less than 8 cm / pixel, and dynamically adjusts the shooting frequency according to weather conditions and seasonal characteristics.

4. The method for evaluating the health of a garden landscape plant according to claim 1, wherein Step 2 includes the following steps: Collect and organize the physiological and ecological characteristics, suitable growth conditions, sensitive environmental factors, and growth and development laws of different garden plants, group the plants according to ecological habits, stress resistance, and growth characteristics to form a structured knowledge base; Based on accumulated temperature, light duration, and phenological observation data, combined with image recognition technology, automatically determine the current growth and development stage of various plants, and formulate differentiated health assessment criteria for different stages; Comprehensively analyze the deviation degree between the recent environmental conditions and the optimal growth conditions of plants, calculate the temperature suitability, water suitability, and light suitability respectively, and form a comprehensive environmental adaptability index through weighted fusion to quantify the impact of the current environment on plant growth; Combined with historical data, analyze the performance of similar plants under similar environmental conditions and growth stages, and dynamically adjust the threshold range of health assessment based on the environmental adaptability index to generate a health benchmark value suitable for the current conditions; Establish a spatially differentiated health benchmark value system for plant communities in different microclimate regions, terrain conditions, and planting densities, and dynamically adjust the fluctuation range of the benchmark value according to seasonal changes and climate fluctuation factors; By continuously comparing the differences between the predicted benchmark value and the actual plant performance, automatically adjust the calculation parameters to achieve the adaptive optimization and accuracy improvement of the health benchmark value calculation method.

5. The method for evaluating the health of a garden landscape plant according to claim 4, wherein The calculation of the comprehensive environmental adaptability index considers four ability indicators: the expansion rate of temperature tolerance range, the recovery rate of water stress, the adjustment efficiency of light adaptability, and the enhancement degree of pathogen resistance. The weights of each indicator are dynamically adjusted according to plant types and seasons. The environmental adaptability index of 0 - 100 is calculated through weighted average. An index increase of more than 15% is regarded as a significant improvement in adaptability, an increase of 5% - 15% is regarded as an improvement in adaptability, and a decrease of less than 5% is regarded as no obvious change in adaptability.

6. The method for evaluating the health of garden landscape plants according to claim 1, wherein Step 3 includes the following steps: Based on the analysis of plant physiological activity laws and environmental stability, identify the time period with the smallest fluctuation of plant physiological indicators and that can fully represent the health status, and determine the optimal assessment time window; Within the determined optimal assessment time window, comprehensively collect data on plant morphological indicators, physiological indicators, and pest and disease conditions, calculate the dispersion degree of each indicator and the deviation degree from the benchmark value, and generate an original health index to preliminarily quantify the plant health status; Evaluate the environmental fluctuation situation within 24 - 72 hours before the assessment, quantify its temporary impact on the current physiological state of the plant, and calculate the short-term environmental perturbation compensation coefficient; Evaluate the induction effect of environmental trend changes in the recent 7 - 30 days on the adaptive physiological regulation of plants, and generate medium - term environmental compensation factors; Quantify the impact of seasonal environmental transitions in the recent 1 - 3 months on plant growth strategies and resource allocation, distinguish the changes caused by adaptive physiological adjustments from the degradation of non - healthy states in plants, and calculate long - term environmental adjustment factors; Based on the analysis results of short - term, medium - term, and long - term environmental changes, comprehensively calculate environmental perturbation compensation factors, and apply them to the correction of the original health index to eliminate the interference of environmental factors at different time scales, so as to obtain a more environmentally neutral assessment result of the true health status of plants.

7. The method for evaluating the health of garden landscape plants according to claim 1, wherein, Step 4 includes the following steps: Compare the currently corrected plant health assessment results with the historical data of the same plant or plant population, analyze long - term trend changes, seasonal fluctuations, and aperiodic anomalies, judge whether the current state is within the normal change range, and screen out potential abnormal points that exceed the expected change rate; Compare the health assessment data of the target plant with the data of plants of the same species but different spatial positions at the same time point, analyze the spatial distribution characteristics of the health status, identify local anomalies and common changes in the population, determine the significance level of spatial anomalies, and distinguish individual - specific problems from regional changes caused by the environment; For the health indicators marked as abnormal, analyze their correlation with environmental factors, calculate the correlation coefficient and lag effect between health anomalies and changes in each environmental factor, and judge whether the anomaly is caused by uncompensated environmental perturbations, so as to filter out false anomalies caused by short - term environmental fluctuations; Based on the analysis results of anomaly significance in the time dimension, anomaly significance in the spatial dimension, and environmental relevance, use the multi - factor weighted scoring method to calculate the comprehensive confirmation score of the anomaly, and set multi - level threshold criteria to classify plant health anomalies, distinguishing serious anomalies that require immediate intervention from minor fluctuations that only need continuous observation; Based on factors such as data quality score, sample sufficiency, environmental condition similarity, and historical diagnosis accuracy, assign a credibility index to each confirmed anomaly to provide decision - making reference, and mark uncertain anomaly points that need further verification; Conduct a classification and attribution analysis of the confirmed true anomalies, combine plant physiological characteristics, environmental conditions, and historical cases to infer possible causes of the disease, and generate a comprehensive diagnosis report including anomaly description, confirmation credibility, possible reasons, and severity.

8. The method for evaluating the health of a garden landscape plant according to claim 1, wherein, Step 5 includes the following steps: Based on the calibrated health indicators, combined with the physiological and ecological characteristics of plants, seasonal growth patterns, and plant flora classification, formulate health assessment criteria applicable to different species; Analyze the potential impact of environmental suitability on the long - term health status of plants, identify key environmental factors limiting growth, and calculate the environmental support index; Connect the plant health assessment results with the geographic information system, establish a spatial mapping relationship between plant location coordinates and health data, and combine with base map data to achieve multi - scale health management from individual plants to communities; Use spatial interpolation algorithms to generate a continuous and smooth health distribution heat map, and adjust the credibility expression of the heat map according to the observation density to visually present the spatial distribution characteristics of the health status of landscape plants. Based on the environmental disturbance compensation factor, calculate the degree of interference of environmental fluctuations in each region on the plant health assessment, so as to overlay environmental disturbance isopleths or regional identifiers on the heat map.

9. The method for evaluating the health of a garden landscape plant according to claim 1, characterized in that Step 6 includes the following steps: Based on the classification results of plant health status, combined with the analysis of plant physiological requirements, pest and disease diagnosis, and environmental limiting factors, call the set of maintenance measures in the expert knowledge base and automatically match the optimal maintenance plan; Determine the key evaluation indicators and expected improvement goals for each maintenance measure, formulate sampling strategies and detection frequencies, set the best recheck time window, and establish a control analysis framework before and after maintenance; Calculate the improvement range, recovery rate, and health recovery rate of health indicators, analyze the change patterns of various physiological indicators, and evaluate the targeted effects of maintenance measures; Track the changes in the sensitivity of plants to environmental stress factors after maintenance, calculate the environmental adaptability index, evaluate the benefits of maintenance measures in enhancing plant stress resistance and long-term stability, and identify the key factors affecting the improvement of adaptability; Compare the maintenance results with the initial expectations, analyze the reasons for the deviation, identify the successful experiences and deficiencies of maintenance measures, and update the measure effectiveness scores in the expert knowledge base; Apply the maintenance practice feedback data to the calibration and optimization of the health assessment model, adjust the environmental disturbance compensation coefficient, update the calculation method of the health benchmark value, improve the anomaly confirmation algorithm, regularly evaluate the overall performance and perform iterative upgrades to achieve continuous improvement of garden plant health management.

10. A method for evaluating the health of garden landscape plants according to claim 9, characterized in that, The calculation formula for the health indicator improvement rate is: (indicator value after maintenance - indicator value before maintenance) / indicator value before maintenance × 100%; the calculation formula for the recovery rate is: health indicator improvement rate / intervention days; the stability index is evaluated by the coefficient of variation of the observed values for 5 consecutive days. A coefficient of variation lower than 20% indicates recovery to stability, higher than 35% indicates instability, and between 20% - 35% indicates a state to be observed.

Images