Garden plant environment ecological assessment method and system

Abstract

The invention provides a garden plant environment ecological assessment method and system. According to the method, self-adaptive layered scanning of vegetation canopies is executed through a flight device, multi-spectral reflection data and three-dimensional morphological parameters are synchronously obtained, and a reflection database for space-time calibration is constructed. High-reflection interference is suppressed by adopting a polarization compensation technology, layered reflectivity weighted calculation is performed based on penetration level identification, and a three-dimensional photosynthetic efficiency map is generated by combining vertical section attenuation gradient and photosynthetic parameter mapping. Secondary scanning is started for the abnormal area, the root system absorption delay error is corrected by fusing soil moisture parameters, and finally an accurate ecological evaluation result is output. According to the technical scheme provided by the invention, three-dimensional dynamic monitoring of the photosynthetic efficiency of the plant community is realized through an air-ground cooperative scanning and dynamic compensation algorithm, and the precision and reliability of ecological assessment in a complex canopy environment are effectively improved.

Description

Technical Field

[0001] This application relates to the technical field of environmental monitoring, and particularly to a method and system for ecological assessment of the environment of garden plants. Background Art

[0002] With the advancement of refined urban ecological management, the health assessment of garden plants needs to achieve dynamic monitoring with high spatio-temporal resolution. Especially under complex canopy structures, it is necessary to synchronously obtain multi-dimensional data such as photosynthetic efficiency, canopy light transmittance, and root water absorption. Traditional methods are difficult to meet the quantitative analysis of ecological parameters of the vertical profile of plant communities, and there is an urgent need for an intelligent monitoring technology that can integrate air-ground data, suppress environmental interference, and dynamically correct assessment errors.

[0003] Currently, the mainstream solution uses a fixed multi-spectral sensor network. By deploying a multi-node sensor array in the garden, the reflection spectral data of the vegetation canopy is continuously collected, and the photosynthesis intensity is estimated by combining a preset canopy model. This solution realizes centralized data processing through wireless transmission and uses machine learning algorithms to preliminarily identify abnormal areas.

[0004] The coverage range and resolution of the fixed sensor network are limited by the hardware deployment density, and it is difficult to dynamically adapt to the three-dimensional morphological changes of the vegetation canopy. Its reflection data is only collected from a single perspective, unable to eliminate spectral interference in high-reflection areas (such as waxy leaves), and does not consider the influence of the canopy penetration level difference on the reflectance weight. In addition, the model relies on static soil parameter assumptions and cannot correct the misjudgment of photosynthetic efficiency caused by the delay of root water absorption in real time, resulting in insufficient accuracy of vertical profile ecological assessment. Summary of the Invention

[0005] This application provides a method and system for ecological assessment of the environment of garden plants to solve the problems of insufficient coverage and lack of dynamic adaptation caused by the static deployment of the sensor network in the prior art.

[0006] In a first aspect, this application provides a method for ecological assessment of the environment of garden plants, including: Implementing stratified scanning of the garden plant community by a flying device, generating a flight trajectory based on the three-dimensional density distribution of the vegetation canopy, and synchronously collecting multi-spectral reflection data including photosynthesis characteristic parameters; Deploying multiple ground verification points in the scanned area, collecting three-dimensional morphological parameters of the vegetation canopy in combination with the spatio-temporal coverage characteristics of the flight trajectory, performing spatio-temporal calibration and matching on the three-dimensional morphological parameters and the multi-spectral reflection data, and constructing a reflection database; Identifying high-reflection interference areas according to the detected reflection characteristics of the vegetation canopy surface, and performing multi-angle polarization compensation on the high-reflection interference areas by adjusting the polarization angle configuration to generate multi-spectral reflection data after specular reflection suppression; Based on the penetration level identification in the reflection database, perform hierarchical reflectance weighting calculation on the multi-spectral reflection data after reflection suppression, and associate the attenuation gradient of different multi-spectral reflection data in the vertical profile of the vegetation canopy with the mapping relationship of photosynthesis characteristic parameters to generate a three-dimensional distribution map of the photosynthetic efficiency of the plant community; According to the photosynthetic efficiency abnormal area in the three-dimensional distribution map, control the flying device to perform secondary scanning. By fusing the multi-spectral reflection data of the secondary scanning with the soil moisture parameters of the ground verification points, correct the evaluation error caused by the delay of root water absorption in the three-dimensional distribution map, and output the ecological evaluation result.

[0007] Optionally, based on the penetration level identification in the reflection database, perform hierarchical reflectance weighting calculation on the spectral data after reflection suppression, including: Extract the hierarchical segmentation information corresponding to the penetration level identification from the reflection database, and the hierarchical segmentation information includes the penetration ability value and the hierarchical thickness value corresponding to each level in the vegetation canopy; According to the penetration ability value, assign weights to the spectral signals of each level in the spectral data after reflection suppression; Based on the hierarchical thickness value, perform superposition calculation on the weighted spectral signals in the order of levels to obtain the reflectance weighting result of each level in the vertical direction.

[0008] Optionally, based on the reflectance weighting result, associate the attenuation gradient of different spectral data in the vertical profile of the vegetation canopy with the mapping relationship of photosynthesis characteristic parameters to generate a three-dimensional distribution map of the photosynthetic efficiency of the plant community, including: Extract the signal intensity change rate of the spectral signals of different levels in the vertical profile changing with the penetration level in the reflection database, and establish the corresponding relationship between the signal intensity change rate and the light energy absorption value in the photosynthesis characteristic parameters; According to the corresponding relationship, convert the spectral signals of each level in the reflectance weighting result into the corresponding light energy absorption values; Based on the spatial position of the hierarchical segmentation information, map the light energy absorption values in the order of levels into three-dimensional space to form a three-dimensional distribution map representing the distribution of the photosynthetic efficiency of the plant community.

[0009] Optionally, according to the photosynthetic efficiency abnormal area in the three-dimensional distribution map, control the flying device to perform secondary scanning. By fusing the multi-spectral reflection data of the secondary scanning with the soil moisture parameters of the ground verification points, correct the evaluation error caused by the delay of root water absorption in the three-dimensional distribution map, and output the ecological evaluation result, including: Extract continuous spatial regions with photosynthetic efficiency values lower than or higher than a preset normal range from the three-dimensional distribution map, and mark them as photosynthetic efficiency abnormal regions; According to the spatial distribution range of the photosynthetic efficiency abnormal region, adjust the flight height and scanning path density of the flying device, generate a secondary scanning trajectory covering the photosynthetic efficiency abnormal region, and control the flying device to perform secondary scanning according to the secondary scanning trajectory; Extract the multi-spectral reflection data of the photosynthetic efficiency abnormal region in the secondary scanning, and obtain the spectral reflection value of each spatial coordinate point in the abnormal region; Extract the soil moisture parameters corresponding to the spatial coordinates of the abnormal region from the ground verification points, where the soil moisture parameters are the measured values of the soil moisture content at the ground verification points within a preset time window; Bind the measured values of the soil moisture content and the spectral reflection values of the abnormal region one-to-one according to the spatial coordinates; According to the measured value of the soil moisture content, perform moisture absorption delay compensation on the spectral reflection value, and replace the photosynthetic efficiency value of the corresponding photosynthetic efficiency abnormal region in the three-dimensional distribution map with the compensated spectral reflection value. Based on the replaced photosynthetic efficiency value, recalculate the photosynthetic efficiency distribution of all spatial coordinate points in the three-dimensional distribution map, and generate an ecological assessment result.

[0010] Optionally, according to the detected reflection characteristics of the vegetation canopy surface, identify high-reflection interference regions, and through adjusting the polarization angle configuration, perform multi-angle polarization compensation on the high-reflection interference regions to generate multi-spectral reflection data after suppressing specular reflection, including: Based on the spectral reflection values of each spatial coordinate point in the multi-spectral reflection data, extract the reflection intensity of the vegetation canopy surface within a preset visible light band, and mark the spatial coordinate points with reflection intensity exceeding the preset threshold as high-reflection interference regions; For the high-reflection interference regions, control the polarization filter component of the flying device to sequentially switch to multiple different polarization angle combinations; At each polarization angle combination, re-collect the spectral reflection values of the high-reflection interference regions to obtain multiple sets of spectral reflection data with different polarization angles; Perform superposition processing on the spectral reflection values of the same spatial coordinate point in the multiple sets of spectral reflection data with different polarization angles, and retain the minimum value of the intensity of the superposed spectral reflection value as the spectral reflection value after suppressing specular reflection of the spatial coordinate point; Merge the spectral reflection value after suppressing specular reflection and the original spectral reflection value that is not marked as a high-reflection interference region according to the spatial coordinates to generate multi-spectral reflection data after suppressing specular reflection.

[0011] Optionally, the spatio-temporal coverage feature includes a time stamp sequence and a spatial coordinate sequence; Deploy multiple ground verification points within the scanning area. Combining the spatio-temporal coverage characteristics of the flight trajectory, collect the three-dimensional morphological parameters of the vegetation canopy. Perform spatio-temporal calibration and matching on the three-dimensional morphological parameters and the multi-spectral reflection data to construct a reflection database, including: Deploy multiple ground verification points within the scanning area according to a preset density. Each ground verification point is equipped with a three-dimensional laser scanning device. The three-dimensional laser scanning device measures the depth detection data of the vegetation canopy by emitting laser pulses and receiving reflected signals; calculate the height values, gap densities, and surface curvature values of each measurement point based on the depth detection data to form a set of three-dimensional morphological parameters of the ground verification points; Extract the acquisition timestamps and spatial coordinates of each spectral sampling point in the multi-spectral reflection data. Based on the timestamp sequence and spatial coordinate sequence of the flight trajectory, establish the corresponding position markers between the spatial coordinates of the spectral sampling points and the canopy height values in the three-dimensional morphological parameters; According to the corresponding position markers, bind the canopy height values, canopy gap density values, and canopy surface curvature values in the three-dimensional morphological parameters to the spectral reflection values in the multi-spectral reflection data to obtain the bound data units; Based on the preset binding rules, store the bound data units in the order of the grid distribution of spatial coordinates to generate a reflection database.

[0012] Optionally, extract the signal intensity change rates of spectral signals at different levels in the vertical section with the change of penetration levels in the reflection database, and establish the corresponding relationship between the signal intensity change rates and the light energy absorption values in the photosynthesis characteristic parameters, including: Extract the spectral signal intensity values at adjacent penetration levels at the same spatial coordinate point from the reflection database, and calculate the ratio of the spectral signal intensity value of the latter penetration level to the previous level as the signal intensity change rate between levels; Based on the signal intensity change rate between levels, calculate the continuous product of all upper-level signal intensity change rates from the top to the bottom of the canopy layer by layer in the order of penetration levels to obtain the cumulative signal intensity change rate; Through a preset experimental calibration method, in the vegetation canopy corresponding to the ground verification points, obtain the light energy absorption value per unit area actually absorbed at different penetration levels through a light intensity attenuation measurement device; Fit the cumulative signal intensity change rate and the measured light energy absorption value in the order of penetration levels with an exponential relationship, and output the exponential relationship fitting result; Based on the exponential relationship fitting result, generate the corresponding mapping table between the cumulative signal intensity change rate and the light energy absorption value as the corresponding relationship between the signal intensity change rate and the light energy absorption value in the photosynthesis characteristic parameters.

[0013] In a second aspect, the present application provides a garden plant environmental ecological assessment system, including: A collection module, configured to perform hierarchical scanning on a garden plant community through a flying device, generate a flight trajectory based on the three-dimensional density distribution of the vegetation canopy, and synchronously collect multi-spectral reflection data including photosynthesis characteristic parameters; The collection module is further configured to arrange a plurality of ground verification points within the scanning area, collect three-dimensional morphological parameters of the vegetation canopy in combination with the spatio-temporal coverage characteristics of the flight trajectory, perform spatio-temporal calibration and matching on the three-dimensional morphological parameters and the multi-spectral reflection data, and construct a reflection database; A generation module, configured to identify a high-reflection interference area according to the detected surface reflection characteristics of the vegetation canopy, perform multi-angle polarization compensation on the high-reflection interference area by adjusting the polarization angle configuration, and generate multi-spectral reflection data after reflection suppression; A calculation module, configured to perform hierarchical reflectance weighted calculation on the multi-spectral reflection data after reflection suppression based on the penetration level identifier in the reflection database, and associate the attenuation gradient of different multi-spectral reflection data in the vertical profile of the vegetation canopy with the mapping relationship of the photosynthesis characteristic parameters, and generate a three-dimensional distribution map of the photosynthetic efficiency of the plant community; A correction module, configured to control the flying device to perform secondary scanning according to the photosynthetic efficiency abnormal area in the three-dimensional distribution map, correct the evaluation error caused by the root water absorption delay in the three-dimensional distribution map by fusing the multi-spectral reflection data of the secondary scanning and the soil moisture parameters of the ground verification points, and output an ecological assessment result.

[0014] In a third aspect, an embodiment of the present application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a garden plant environmental ecological assessment method as described in the first aspect above.

[0015] In a fourth aspect, an embodiment of the present application provides a computer storage medium, storing a computer program, and when the computer program is executed by a computer, it implements a garden plant environmental ecological assessment method as described in the first aspect.

[0016] In the embodiments of the present application, a flight device is used to perform hierarchical scanning on a garden plant community, generate a flight trajectory based on the three-dimensional density distribution of the vegetation canopy, and synchronously collect multi-spectral reflection data including photosynthesis characteristic parameters; a plurality of ground verification points are arranged in the scanning area, and in combination with the spatio-temporal coverage characteristics of the flight trajectory, three-dimensional morphological parameters of the vegetation canopy are collected, and the three-dimensional morphological parameters are calibrated and matched with the multi-spectral reflection data in space and time to construct a reflection database; according to the detected surface reflection characteristics of the vegetation canopy, a high-reflection interference area is identified, and by adjusting the polarization angle configuration, multi-angle polarization compensation is applied to the high-reflection interference area to generate multi-spectral reflection data after reflection suppression; based on the penetration level identifier in the reflection database, hierarchical reflectance weighting calculation is performed on the multi-spectral reflection data after reflection suppression, and the mapping relationship between the attenuation gradient of different multi-spectral reflection data in the vertical profile of the vegetation canopy and the photosynthesis characteristic parameters is associated to generate a three-dimensional distribution map of the photosynthetic efficiency of the plant community; according to the photosynthetic efficiency abnormal area in the three-dimensional distribution map, the flight device is controlled to perform secondary scanning, and by fusing the multi-spectral reflection data of the secondary scanning with the soil moisture parameters of the ground verification points, the evaluation error caused by the root water absorption delay in the three-dimensional distribution map is corrected, and an ecological evaluation result is output. Through dynamic flight scanning, polarization interference suppression, hierarchical reflectance weighting, and secondary data fusion correction, three-dimensional accurate evaluation of plant photosynthetic efficiency under complex canopy structures is achieved, and problems such as insufficient coverage of traditional static sensor networks, lack of reflection interference suppression, deviation in vertical profile weight allocation, and cumulative errors in root water absorption delay are solved.

[0017] Further, based on the penetration level identifier, such as the penetration ability value and thickness value in the level segmentation information, hierarchical weight allocation and superposition calculation are performed on the spectral data after reflection suppression to quantify the reflectance contribution of each level, solving the problem of distortion of the vertical profile reflectance weight caused by unstratified calibration in the traditional method. The technical effect is that through the hierarchical dynamic weight allocation and superposition model, the physical rationality of the reflectance calculation in the vertical direction of the canopy is significantly improved, the error of single reflectance averaging is avoided, and the vertical resolution accuracy of the three-dimensional distribution map of photosynthetic efficiency is enhanced.

[0018] These aspects or other aspects of the present application will be more clearly understood in the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0020] Figure 1 The flowchart of a method for ecological assessment of garden plant environment provided by this application is shown; Figure 2 The schematic structural diagram of a system for ecological assessment of garden plant environment provided by this application is shown; Figure 3 The schematic structural diagram of a computing device provided by this application is shown. Detailed implementation manners

[0021] In order to enable the personnel in the technical field to better understand the solution of this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application.

[0022] In some processes described in the specification, claims and the above-mentioned drawings of this application, a plurality of operations that appear in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order in which they appear in this article or may be executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish each different operation, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions such as "first" and "second" in this article are used to distinguish different messages, devices, modules, etc., do not represent the sequence, and do not limit that "first" and "second" are different types.

[0023] Researchers have found that existing garden ecological monitoring technologies have insufficient three-dimensional coverage due to the static layout of the sensor network, spectral data distortion caused by the lack of suppression of high-reflection interference, deviation in the assessment of photosynthetic efficiency due to the unstratified calibration of the vertical profile reflectivity weights, and difficulty in correcting the root water absorption delay error in the static soil parameter modeling. Based on this, a method for ecological assessment of garden plant environment is provided. This method can achieve three-dimensional dynamic monitoring of the photosynthetic efficiency of the vegetation canopy by means of dynamic flight scanning and polarization compensation technology, fuse air-ground multi-source data, and correct the root water absorption error in real time, significantly improving the assessment accuracy. The technical solution of this application is applicable to scenarios such as urban garden three-dimensional green belts, complex canopy communities in ecological restoration areas, and high-density planting areas in agricultural parks that require refined analysis of vertical profile ecological parameters.

[0024] The entire R & D process reflects the core advantages of three-dimensional dynamic perception and multi-source data closed-loop feedback. It generates dynamic coverage capabilities through adaptive hierarchical scanning of the flight device, combines a real-time adjustment mechanism for suppressing high-reflection interference with polarization compensation, realizes hierarchical modeling of the vertical profile reflectivity based on penetration layer identification, and forms an error self-correction closed-loop through secondary scanning and soil parameter fusion, ultimately breaking through the bottleneck of multi-dimensional analysis of complex canopy structures by static monitoring technologies.

[0025] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present application.

[0026] Figure 1 The following is a flowchart of a method for ecological assessment of the garden plant environment provided by an embodiment of the present application. As Figure 1 shown, the method includes: 101. Implement hierarchical scanning of the garden plant community through a flying device, generate a flight trajectory based on the three-dimensional density distribution of the vegetation canopy, and synchronously collect multi-spectral reflection data including photosynthesis characteristic parameters; Hierarchical scanning refers to dividing the monitoring area of the flying device into multiple levels for layer-by-layer data collection according to the vertical structure density characteristics of the vegetation canopy. Among them, the flying device can be a drone, etc. The three-dimensional density distribution refers to a three-dimensional structure model generated by quantifying parameters such as leaf density and branch distribution at different heights of the vegetation canopy. The photosynthesis characteristic parameters include spectral indicators directly related to the plant light energy conversion efficiency, such as chlorophyll fluorescence intensity and photosynthetically active radiation absorption rate.

[0027] In the embodiment of the present application, first, the original point cloud data of the vegetation canopy is obtained through lidar, and the density clustering algorithm is used to identify the leaf aggregation areas in different height layers, and multiple vertical levels are divided. Subsequently, based on the layer density difference, hierarchical scanning of the garden plant community is implemented through a flying device, and the adaptive ant colony algorithm is used to dynamically plan the flight trajectory, so that the drone stays longer and reduces the flight height in the high-density area to ensure the accuracy of spectral data collection. During the flight, the onboard hyperspectral imager synchronously records the multi-band reflection data of each layer and binds it to the trajectory coordinates in real time to form a spectral data set with spatial position marks.

[0028] For example, in the monitoring of the canopy of the banyan tree in the wetland park, the drone is equipped with lidar and hyperspectral imager to perform hierarchical scanning. The lidar point cloud data shows that there are three significant density layers in the vertical height of the canopy: the top layer has a leaf density of 110 per cubic meter, the middle layer reaches 300, and the bottom layer is 85. The system divides three levels accordingly and plans an adaptive flight trajectory. The middle layer adopts a spiral progressive path, and the single-layer scanning time is extended to 18 minutes. The top layer and the bottom layer adopt a broken line round-trip path, and the total coverage area reaches 3 hectares. During the flight, the chlorophyll fluorescence intensity and photosynthetically active radiation absorption rate data are synchronously collected to generate a spectral data set with geographical coordinate marks.

[0029] 102. Deploy multiple ground verification points within the scanning area. Combining the spatio-temporal coverage characteristics of the flight trajectory, collect the three-dimensional morphological parameters of the vegetation canopy, perform spatio-temporal calibration and matching on the three-dimensional morphological parameters and the multi-spectral reflection data, and construct a reflection database. The ground verification points refer to fixed sensor nodes deployed below the vegetation canopy, which are used to collect three-dimensional morphological parameters such as branch inclination angle and leaf gap ratio. The spatio-temporal coverage characteristics refer to the distribution characteristics of the scanning path of the flying device in the time and space dimensions, and need to be aligned with the ground data to achieve air-ground collaborative analysis.

[0030] In the embodiment of the present application, ground verification points are arranged in a grid within the scanning area. A lidar scanner with differential GPS positioning and a full-frame multi-spectral camera are deployed at the ground verification points. By taking multi-angle hemisphere images of the vegetation canopy and synchronously collecting three-dimensional point cloud data with millimeter-level accuracy, use the improved Canny edge detection algorithm to extract the leaf contour features, and combine the Monte Carlo ray tracing technology to generate three-dimensional morphological parameters including leaf inclination distribution, gap ratio, and branch projection area ratio; at the same time, start the spatio-temporal calibration engine, perform four-dimensional encoding on the time stamp of the data collected by the flying device and the spatial coordinates of the ground station, and establish a spatio-temporal mapping relationship based on the improved R-tree index algorithm; then use the accelerated robust feature matching algorithm to register the feature points of the aerial survey point cloud and the ground-based scanning model, and achieve sub-centimeter-level spatial alignment through the iterative closest point optimization algorithm; store the calibrated three-dimensional morphological parameters and the multi-spectral reflectance data corresponding to the coordinates in the distributed database in spatio-temporal sequence to form a reflection database including longitude, latitude, elevation, reflection band, and structural parameters.

[0031] Continuing with the above example, arrange ground verification sites according to a predetermined grid, and configure a lidar scanner with differential GPS positioning and a full-frame multi-spectral camera at each site. When the flying device performs layered scanning and passes through the northeast region, the ground verification system is started synchronously. The lidar at Site 3 captures dense point cloud data in the middle layer of the canopy, accurately extracts the leaf contour features through the improved Canny edge detection algorithm, and combines the Monte Carlo ray tracing technology to calculate three-dimensional morphological parameters such as the average leaf inclination angle and canopy gap ratio. The spatio-temporal calibration engine real-time analyzes the time stamp of the three-dimensional morphological parameters of the flying device and the spatial coordinates of the ground station, generates a unique hash code and establishes a four-dimensional mapping relationship, and matches the aerial survey point cloud data in the adjacent period through the improved R-tree index. In the feature matching stage, the accelerated robust feature algorithm is used to extract the key points of the aerial survey and ground-based point clouds, and the sub-millimeter-level spatial alignment is completed through the iterative closest point optimization algorithm. Finally, the calibrated leaf area index is dynamically associated with the multi-spectral reflectance data corresponding to the coordinates to form an incrementally updated reflection database, providing accurate canopy structure basic data for subsequent anti-reflection processing.

[0032] 103. Identify the high-reflection interference regions based on the detected reflection characteristics of the vegetation canopy surface, and adopt multi-angle polarization compensation for the high-reflection interference regions by adjusting the polarization angle configuration to generate multi-spectral reflection data after specular reflection suppression. The high-reflection interference regions refer to the abnormal specular reflection regions caused by the leaf wax layer or water accumulation, and their reflectivity is significantly higher than that of normal vegetation. Multi-angle polarization compensation means suppressing the specular reflection noise from different directions by adjusting the angle combination of polarization filters and retaining the effective diffuse reflection signals of the vegetation.

[0033] In the embodiment of the present application, based on the reflection characteristics of the vegetation canopy surface recorded in the reflection database, a dynamic threshold segmentation algorithm is used to identify the high-reflection interference regions where the reflectivity exceeds the preset threshold. For the high-reflection interference regions, the polarization camera carried by the flying device automatically adjusts the filter angle to collect polarization spectral data in three directions of 0 degree, 45 degrees, and 90 degrees respectively. Through the weighted fusion algorithm, the specular reflection components are removed from the collected polarization spectral data, and the diffuse reflection spectral characteristics are retained. A multi-spectral data set is constructed based on the remaining polarization spectral data, that is, the multi-spectral data set contains multiple polarization spectral data after specular reflection suppression.

[0034] Continuing with the above example, the scanning data shows that the reflectivity of the sunny side at the top of the canopy has increased abnormally to 78% due to the leaf wax layer on the leaf surface. The drone polarization camera collects data of the target area at three angles of 45 degrees, 90 degrees, and 135 degrees, and separates the specular reflection and diffuse reflection components through the polarization difference algorithm. After processing, the spectral noise in the high-reflection region is reduced by 70%, and the signal-to-noise ratio of the chlorophyll fluorescence signal is increased from 3.5 to 9.1, effectively supporting the accurate extraction of photosynthetic parameters.

[0035] 104. Based on the penetration level identifier in the reflection database, perform hierarchical reflectivity weighted calculation on the multi-spectral reflection data after specular reflection suppression, and associate the attenuation gradient of different multi-spectral reflection data in the vertical profile of the vegetation canopy with the mapping relationship of photosynthesis characteristic parameters to generate a three-dimensional distribution map of the photosynthetic efficiency of the plant community. The penetration level identifier refers to the vertical level number divided according to the lidar data, which is used to mark the spatial position of the spectral data in the canopy. The attenuation gradient refers to the signal attenuation law caused by leaf occlusion when light waves penetrate different levels.

[0036] In the embodiments of the present application, the vegetation canopy is vertically divided into a number of optical penetration units based on the penetration level identifiers in the reflection database. The discrete ordinate method is used to calculate the light radiation transmission path within each unit, and the stratified reflectance weighted calculation is performed on the multi-spectral reflection data after reflection suppression, where the weight coefficient is dynamically adjusted according to the voxel density and leaf area index of the corresponding level. Subsequently, a depth residual network model is constructed. The input layer includes multi-spectral reflectance, ambient temperature and humidity, and canopy structure parameters. The light attenuation gradient features of the vertical profile are extracted through an adaptive feature fusion module, and the mapping relationship of the light energy utilization efficiency related to the photosynthesis characteristic parameters is associated. Finally, the three-dimensional Kriging interpolation algorithm is used to spatially reconstruct the photosynthetic efficiency index of the discrete units, and the three-dimensional distribution map of photosynthetic efficiency with millimeter-level resolution is generated by superimposing the canopy digital surface model.

[0037] Continuing with the above example, in the vertical dissection of the canopy of the banyan tree forest in the wetland park, the banyan tree canopy is divided into optical penetration units with a thickness of half a meter per layer. Based on the penetration level identifiers in the reflection database, the discrete ordinate method is used to simulate the solar radiation transmission path at noon. When performing the stratified weighted calculation on the multi-spectral reflection data after reflection suppression in step 103, the voxel density in the middle layer of the canopy is at a high level, and the weight coefficient is dynamically assigned in combination with the leaf area index measured at the ground verification points, so that the weighted value of the reflectance in the 710 nm band is corrected from the initial 0.38 to 0.42. The input of the constructed depth residual network model includes the corrected five-band reflectance, canopy temperature, and air humidity parameters. The light attenuation gradient features in the vertical direction are identified through the feature fusion module and associated with the mapping relationship of the light energy utilization efficiency in the photosynthesis characteristic parameters. The three-dimensional distribution map of photosynthetic efficiency in the northeast region of the canopy is reconstructed using the three-dimensional Kriging interpolation algorithm. After superimposing the high-precision canopy surface model, it clearly shows that there is an elliptical low-value area of photosynthetic efficiency in the middle layer region. In the simultaneously output heat map, the energy conversion rate in the corresponding region is significantly lower than that of the surrounding area, which is marked as an abnormal target area, triggering the subsequent step 105 to perform a special verification scan on the root water absorption delay effect in this area.

[0038] 105. According to the photosynthetic efficiency abnormal area in the three-dimensional distribution map, control the flight device to perform a secondary scan. By fusing the multi-spectral reflection data of the secondary scan with the soil moisture parameters of the ground verification points, correct the evaluation error caused by the root water absorption delay in the three-dimensional distribution map, and output the ecological evaluation result.

[0039] Root water absorption delay refers to the time difference between the water absorption rate of plant roots and the transpiration demand of the canopy, which will cause an evaluation error in photosynthetic efficiency. Soil moisture parameters refer to the real-time soil moisture content, hydraulic conductivity and other data collected at the ground verification points.

[0040] In the embodiments of the present application, the system automatically detects the photosynthetic efficiency abnormal area in the three-dimensional distribution map and triggers the secondary scanning instruction of the drone. During the secondary scanning, the soil moisture content data of the ground verification points is obtained synchronously, and the root water absorption delay time is calculated through the time-delay compensation model to obtain relevant delay parameters. The delay parameters are substituted into the photosynthesis model to dynamically correct the evaluation error caused by the root water absorption delay in the three-dimensional distribution map, and finally the ecological evaluation result without water interference is output.

[0041] In the monitoring of the banyan forest in the wetland park, based on the three-dimensional distribution map, an elliptical low-value area of photosynthetic efficiency in the middle layer of the northeast region is automatically identified, and the drone is triggered to perform spiral secondary scanning with meter-level accuracy during the stable afternoon light period. The flight device is equipped with a high-sensitivity multi-spectral imager to collect sub-meter resolution data of the target area, and the soil moisture sensors at the ground verification points are synchronously activated to obtain the dynamic data of the soil moisture content and hydraulic conductivity at the centimeter depth in real time. By analyzing the phase difference between the canopy transpiration rate and the root water absorption rate through the time-delay compensation model, and combining the change curve of the soil hydraulic conductivity, the root water absorption delay time parameter is calculated and input into the photosynthesis model to perform reverse compensation and correction on the original light energy utilization efficiency. The corrected three-dimensional distribution map shows that the photosynthetic efficiency value in the original abnormal area has recovered to the normal fluctuation range. The system automatically generates an evaluation report on the root water transport capacity, marks this area as a water stress warning area on the digital twin platform, and pushes a precise irrigation optimization plan to complete the closed-loop ecological evaluation process from data collection to decision support.

[0042] This solution realizes the three-dimensional perception of the vegetation canopy through dynamic hierarchical scanning and adaptive trajectory planning, and constructs a high-precision reflection database by calibrating the time and space of the ground verification points. The multi-angle polarization compensation technology is used to effectively suppress the high-reflection interference, and the hierarchical weighted model based on the penetration layer identification accurately quantifies the photosynthetic efficiency distribution of the vertical profile. Finally, through the secondary scanning and soil parameter fusion, the limitation of the traditional method in modeling the root water absorption delay effect is broken through, and the full-link accurate evaluation of ecological parameters from the canopy to the roots is realized, providing the three-dimensional dynamic analysis ability for the monitoring of complex vegetation environments.

[0043] In some embodiments, based on the penetration layer identification in the reflection database, the hierarchical reflectance weighting calculation is performed on the spectral data after reflection suppression, including: 201. Extract the hierarchical segmentation information corresponding to the penetration layer identification from the reflection database, and the hierarchical segmentation information includes the penetration ability value and the layer thickness value corresponding to each layer in the vegetation canopy; The penetration layer identifier refers to the vertical layer number of the vegetation canopy divided by lidar data, which is used to mark the spatial positions of different height layers. The layer segmentation information includes a penetration ability value and a layer thickness value. The penetration ability value refers to the proportion of light waves that are not blocked by leaves when penetrating a certain layer, and the layer thickness value refers to the physical thickness of the layer in the vertical direction.

[0044] In the embodiments of the present application, the layer segmentation information of the data table corresponding to the penetration layer identifier is called from the reflection database, and the penetration ability value and the layer thickness value of each layer are extracted through a structured query statement. The penetration ability value is calculated from the leaf gap rate within the layer, and the formula is: penetration ability value = 1 - (leaf projection area / layer cross-sectional area). The layer thickness value is directly obtained through the vertical height difference of the lidar point cloud data.

[0045] 202. According to the penetration ability value, weight distribution is performed on the spectral signals of each layer in the spectrally-reflection-suppressed data. Weight distribution means assigning a correction coefficient to the spectral signal of each layer according to the penetration ability value. The higher the weight of the spectral signal of the layer with a lower penetration ability value and more serious occlusion is, to compensate for signal attenuation.

[0046] In the embodiments of the present application, the spectrally-reflection-suppressed data is segmented by layer, and the spectral signal intensity of each layer is normalized. A weight function is constructed based on the penetration ability value, and the formula is: weight coefficient = 1 / (penetration ability value + 0.1), where 0.1 is a smoothing factor. The weight coefficient is multiplied by the normalized spectral signal to generate a weighted layer spectral data set, realizing weight distribution for the spectral signals of each layer in the spectrally-reflection-suppressed data.

[0047] 203. Based on the layer thickness value, the weighted spectral signals are superimposed and calculated in the layer order to obtain the weighted result of the reflectance in the vertical direction for each layer.

[0048] The layer-order superposition calculation means combining the weighted spectral signals with the layer thickness value in the order from the top layer to the bottom layer in the vertical direction, and accumulating the reflectance contribution values layer by layer.

[0049] In the embodiments of the present application, based on the layer thickness value, the weighted spectral signals are superimposed and calculated in the layer order to construct a vertical superposition model, and the formula is: layer reflectance contribution value = weighted spectral signal × layer thickness value. The contribution values are accumulated layer by layer in the layer number order (from the top layer to the bottom layer) to generate a weighted distribution curve of the reflectance in the vertical direction, and finally the quantified reflectance results of each layer are output.

[0050] The following is a specific example: In the monitoring scenario of the canopy layer of the banyan tree in the wetland park, the system executes a hierarchical reflectance quantification process. In step 201, the vertical layer segmentation information is extracted from the reflection database. The penetration ability value of the top layer is 0.15 at 8 to 10 meters, and the proportion of the leaf projection area of the layer calculated from the lidar point cloud is 85%. The penetration ability value of the middle layer is 0.38 at 5 to 8 meters, and the layer thickness is 3 meters. The penetration ability value of the bottom layer is 0.72 at 2 to 5 meters, and the layer thickness is 3 meters. In step 202, weight distribution is performed on the 710-nm band spectral signal after reflection suppression. The original normalized signals are 0.25 for the top layer, 0.40 for the middle layer, and 0.55 for the bottom layer. The initial weight coefficients are calculated as 4.0, 2.08, and 1.22 according to the reciprocal of the penetration ability value plus 0.1, and are corrected to 0.548, 0.285, and 0.167 after normalization. In step 203, hierarchical superposition calculation is performed. The contribution value of the top layer is 0.25 multiplied by 0.548 multiplied by 2 meters to get 0.274. The middle layer is 0.40 multiplied by 0.285 multiplied by 3 meters to get 0.342. The bottom layer is 0.55 multiplied by 0.167 multiplied by 3 meters to get 0.276. The independent quantification results of the reflectance of each layer are output: the reflectance contribution of the top layer of 0.274 characterizes the high reflectance characteristics of the leaf wax layer, the middle layer of 0.342 indicates abnormal stomatal conductance, and the bottom layer of 0.276 reflects the surface water transpiration effect.

[0051] This solution solves the reflectance averaging error caused by unstratified calibration in traditional methods through the dynamic weight distribution of the penetration ability value and the layer thickness value and vertical superposition calculation. The hierarchical weight model quantifies the true reflectance contribution of each layer. The superposition calculation combined with the physical thickness improves the vertical resolution accuracy, and finally provides a reliable quantification basis for the three-dimensional evaluation of the photosynthetic efficiency of the vegetation canopy.

[0052] In some embodiments, based on the reflectance weighting result, the mapping relationship between the attenuation gradient of different spectral data in the vertical profile of the vegetation canopy and the photosynthesis characteristic parameters is associated to generate a three-dimensional distribution map of the photosynthetic efficiency of the plant community, including: 301. Extract the signal intensity change rate of the spectral signals of different layers in the reflection database changing with the penetration layer in the vertical profile, and establish the corresponding relationship between the signal intensity change rate and the light energy absorption value in the photosynthesis characteristic parameters; The signal intensity change rate refers to the rate at which the spectral signal attenuates with the increase of the penetration layer in the vertical profile, usually expressed as the percentage of attenuation per meter or the absolute value. The light energy absorption value is a key index in the photosynthesis characteristic parameters, reflecting the efficiency of the plant leaves in converting light energy into chemical energy.

[0053] In the embodiments of the present application, the spectral signal intensity of each level varying with the penetration level in the vertical section is extracted from the reflection database, and the signal intensity change rate between adjacent levels is calculated in the vertical order. For example, the signal intensity change rate from the top layer to the middle layer = (top layer signal intensity / middle layer signal intensity) x 100%. The corresponding relationship between the signal intensity change rate and the light energy absorption value in the photosynthesis characteristic parameters is established through a linear regression model. For every 10% increase in the signal intensity change rate, the light energy absorption value decreases by 3%.

[0054] 302. According to the corresponding relationship, convert the spectral signal of each level in the reflectance weighting result into the corresponding light energy absorption value; The conversion of the spectral signal refers to the physical quantity of converting the spectral signal value in the reflectance weighting result into the light energy absorption value based on the corresponding relationship in step 301.

[0055] In the embodiments of the present application, for the reflectance weighting result of each level, substitute it into the corresponding relationship between the signal intensity change rate and the light energy absorption value in the photosynthesis characteristic parameters established in step 301. The light energy absorption value = level reflectance weighting value x (1 - attenuation loss coefficient x signal intensity change rate). For example, if the level reflectance weighting value is 1500 and the corresponding signal intensity change rate is 20%, then the light energy absorption value = 1500×(1 - 20%×0.3) = 1500×0.94 = 1410. After completing the conversion level by level, a level dataset containing the light energy absorption values is generated.

[0056] 303. Based on the spatial position of the level segmentation information, map the light energy absorption values to the three-dimensional space in the level order to form a three-dimensional distribution map representing the photosynthetic efficiency distribution of the plant community.

[0057] The three-dimensional space mapping refers to projecting the light energy absorption values to the three-dimensional grid in the level order according to the spatial coordinates (such as longitude, latitude, altitude) in the level segmentation information to form a three-dimensional distribution model.

[0058] In the embodiments of the present application, a three-dimensional space grid is constructed based on the spatial position data of the level segmentation information, and the light energy absorption values of each level are filled into the corresponding grid cells in the level order through a spatial interpolation algorithm, such as Kriging interpolation. For example, the middle layer light energy absorption value of 1410 is assigned to the grid area with an altitude of 2.1 - 3.6 meters, and finally a three-dimensional distribution map representing the photosynthetic efficiency distribution of the plant community including the light energy absorption intensity and spatial position is generated.

[0059] The following is a specific example: In a monitoring scenario of a pine forest canopy, the system executes a three-dimensional map generation process for photosynthetic efficiency: Step 301 extracts the signal intensity change rates of three vertical levels. The signal from the top layer to the middle layer drops from 2000 to 1200, with an attenuation gradient of 40%. The signal from the middle layer to the bottom layer drops from 1200 to 840, with an attenuation gradient of 30%. A quantitative relationship between the attenuation gradient and the light energy absorption value is established through measured data. For every 10% attenuation gradient, the light energy absorption value decreases by 2.5%. Step 302 converts the weighted reflectance results. The weighted value of the top layer is 1800 without attenuation, and the light energy absorption value remains 1800. The weighted value of the middle layer is 1300 corresponding to 40% attenuation, and the absorption value is calculated as 1300 multiplied by 0.9 to get 1170. The weighted value of the bottom layer is 900 corresponding to 70% attenuation, and the absorption value is adjusted to 742.5. Step 303, based on the hierarchical space coordinates, maps the light energy absorption value of 1800 in the top layer to the area with an elevation of 15 - 20 meters, the middle layer of 1170 to 10 - 15 meters, and the bottom layer of 742.5 to 5 - 10 meters, and generates a three-dimensional distribution map through Kriging interpolation. The three-dimensional distribution map shows that the light energy absorption is the strongest at the top of the canopy. In some local areas of the middle layer, due to the interlacing of branches and leaves, a high-efficiency hotspot with an absorption value of 1170 is formed. The bottom layer is significantly reduced due to occlusion, accurately revealing the vertical photosynthetic efficiency distribution characteristics of the pine forest.

[0060] This solution quantifies the dynamic relationship between the vertical attenuation gradient and light energy absorption, converts the reflectance data into photosynthetic efficiency parameters with clear physical meanings, and combines three-dimensional space mapping technology to achieve a three-dimensional presentation of the photosynthetic efficiency of the vegetation canopy. Compared with the traditional two-dimensional plane evaluation method, this technology reveals the differences in light energy utilization in the vertical direction of the canopy, providing a reliable basis for accurately identifying high-efficiency photosynthesis areas and ecological restoration target areas.

[0061] In some embodiments, according to the photosynthetic efficiency abnormal regions in the three-dimensional distribution map, the flight device is controlled to perform a secondary scan. By fusing the multi-spectral reflectance data of the secondary scan with the soil moisture parameters of the ground verification points, the evaluation error caused by the delayed water absorption of the root system in the three-dimensional distribution map is corrected, and an ecological evaluation result is output, including: 401. Extract continuous spatial regions in the three-dimensional distribution map where the photosynthetic efficiency value is lower than or higher than the preset normal range, and mark them as photosynthetic efficiency abnormal regions; The photosynthetic efficiency abnormal region refers to a three-dimensional spatial region in the three-dimensional distribution map where the photosynthetic efficiency value continuously deviates from the preset normal threshold range, usually caused by delayed water absorption of the root system or abnormal canopy structure. The preset normal range is set through historical data statistics or baseline values of the same type of vegetation.

[0062] In the embodiments of the present application, first, the photosynthetic efficiency values of all units are extracted from the three-dimensional distribution map, and preliminary screening is performed based on a preset normal threshold range. For example, when the normal range is 800 to 1500 units, the values lower than 600 or higher than 1600 are marked as candidate abnormal points. Then, the three-dimensional region growing algorithm is used, with the candidate abnormal points as seed points, to spread in the 26-neighborhood direction in three-dimensional space, detect and merge all continuously adjacent abnormal voxels, and form a spatially continuous photosynthetic efficiency abnormal region. Finally, the minimum circumscribed cube spatial coordinates of the photosynthetic efficiency abnormal region are output, including the minimum and maximum values of the X, Y, and Z axes, to provide a spatial range definition for subsequent secondary scanning.

[0063] 402. According to the spatial distribution range of the photosynthetic efficiency abnormal region, adjust the flight height and scanning path density of the flying device, generate a secondary scanning trajectory covering the photosynthetic efficiency abnormal region, and control the flying device to perform secondary scanning according to the secondary scanning trajectory; The secondary scanning trajectory refers to a flight path dynamically adjusted according to the spatial distribution of the abnormal region, and high-precision data supplementation is achieved by increasing the scanning density and decreasing the flight height.

[0064] In the embodiments of the present application, based on the spatial distribution range of the photosynthetic efficiency abnormal region, the target region is divided into three-dimensional grids with a precision of 0.3 meters, and an adaptive ant colony algorithm is used to generate a snake-shaped round-trip path covering all grids. The path interval in the horizontal direction is set to 0.3 meters, and the flight height is hierarchically planned according to the layer thickness in the vertical direction. The unmanned aerial vehicle realizes high-precision path tracking through a PID controller and RTK positioning technology, generates a secondary scanning trajectory covering the photosynthetic efficiency abnormal region, and controls the flying device to perform secondary scanning according to the secondary scanning trajectory. The flight height is reduced from 50 meters to 20 meters, and the horizontal positioning error is controlled within 0.1 meter, ensuring that the scanning data resolution is improved to the centimeter level.

[0065] 403. Extract the multi-spectral reflection data of the photosynthetic efficiency abnormal region in the secondary scanning, and obtain the spectral reflection value of each spatial coordinate point in the abnormal region; The spectral reflection value of the spatial coordinate point refers to the multi-spectral reflectance data corresponding to each three-dimensional coordinate point (longitude, latitude, altitude) in the secondary scanning, covering the visible light to near-infrared bands.

[0066] In the embodiments of the present application, the drone is equipped with a hyperspectral imager and flies along an encrypted trajectory to collect image data of the multispectral reflection data of the photosynthetic efficiency abnormal area in the secondary scan at a rate of 100 frames per second. The image pixels are matched with the three-dimensional space coordinates in real time through the SLAM algorithm to generate a reflectance matrix with geographical tags. Subsequently, the empirical line correction method is used to eliminate the influence of the change of light intensity. For example, the data collected at different times are uniformly corrected to the reflectance values under the standard light conditions, and finally the accurate spectral reflectance values of each spatial coordinate point are obtained.

[0067] 404. Extract the soil moisture parameters corresponding to the spatial coordinates of the abnormal area from the ground verification points. The soil moisture parameters are the measured values of the soil moisture content at the ground verification points within a preset time window; bind the measured values of the soil moisture content and the spectral reflectance values of the abnormal area one-to-one according to the spatial coordinates. The measured value of the soil moisture content refers to the volumetric water content of the soil measured at the ground verification point within a preset time window, such as within 1 hour before and after the scan, and is used to characterize the available water for root water absorption. The spatial coordinate binding refers to associating the soil parameters to the corresponding abnormal area voxel units according to the geographical location.

[0068] In the embodiments of the present application, the measured values of the soil moisture content at the ground verification points within a preset time window corresponding to the spatial coordinates of the abnormal area are extracted from the ground verification point network, and the Kriging interpolation algorithm is used to generate a soil moisture spatial distribution surface with the same resolution as the abnormal area grid. The central coordinates of each voxel unit in the abnormal area are projected onto the ground plane, and the soil moisture content value at the corresponding position is extracted. For example, when the central coordinates of a voxel unit are X = 102, Y = 205, and Z = 5 meters, the moisture content of 12% at the ground coordinates X = 102 and Y = 205 is associated, and the one-to-one binding of the spatial coordinates of the air-ground data is completed.

[0069] 405. According to the measured values of the soil moisture content, perform moisture absorption delay compensation on the spectral reflectance values, and replace the photosynthetic efficiency values in the corresponding photosynthetic efficiency abnormal area in the three-dimensional distribution map with the compensated spectral reflectance values. Based on the replaced photosynthetic efficiency values, recalculate the photosynthetic efficiency distribution of all spatial coordinate points in the three-dimensional distribution map and generate an ecological assessment result.

[0070] The moisture absorption delay compensation refers to performing time-delay correction on the spectral reflectance values according to the soil moisture content to eliminate the interference of the lag of root water absorption on the calculation of photosynthetic efficiency.

[0071] In the embodiments of the present application, a preset delay time mapping table is queried according to the soil moisture content value. For example, a moisture content of 9% corresponds to a 4-hour delay. The spectral reflectance values of the photosynthetic efficiency abnormal area are corrected with time lag. When the measured value of the soil moisture content is lower than the preset standard, the reflectance value of the near-infrared band in the spectral reflectance value is increased proportionally; when the measured value of the soil moisture content is higher than the preset standard, the reflectance value of the near-infrared band in the spectral reflectance value is decreased proportionally, and the reflectance is recalculated using historical light data. The corrected reflectance is substituted into the photosynthetic efficiency model to regenerate a three-dimensional distribution map, and the spatial data is smoothed by a cubic spline interpolation algorithm, and finally an ecological assessment report eliminating the moisture absorption delay error is output.

[0072] The following is a specific example: In the monitoring of the banyan forest in the wetland park, the three-dimensional distribution map detected an abnormal area of photosynthetic efficiency in the northwest area where the photosynthetic efficiency value was continuously lower than 500. The system planned an encrypted scanning path with a precision of 0.3 meters, and the drone completed a secondary scan at a height of 20 meters, collecting 1200 high-precision reflectance points. The ground data showed that the soil moisture content in the photosynthetic efficiency abnormal area was 9%, corresponding to a 4-hour delay. After correcting the reflectance value with time lag, the photosynthetic efficiency value increased from 480 to 650, and the updated three-dimensional distribution map showed that the deviation decreased from 35% to 12%, generating an ecological assessment result, and suggesting to implement drip irrigation and shading measures.

[0073] This solution effectively eliminates the photosynthetic efficiency evaluation error caused by the root water absorption delay through high-precision secondary scanning and soil moisture dynamic compensation. Abnormal area detection and adaptive path planning improve the local data resolution, and the time lag correction model quantifies the impact of environmental interference, ultimately achieving the spatial consistency and time reliability of the ecological assessment results, providing accurate decision-making support for vegetation health management.

[0074] In some embodiments, according to the detected reflection characteristics of the vegetation canopy surface, a high-reflection interference area is identified, and by adjusting the polarization angle configuration, multi-angle polarization compensation is performed on the high-reflection interference area to generate multi-spectral reflection data after suppressing specular reflection, including: 501. Based on the spectral reflectance values of each spatial coordinate point in the multi-spectral reflection data, the reflection intensity on the vegetation canopy surface within the preset visible light band is extracted, and the spatial coordinate points with the reflection intensity exceeding the preset threshold are marked as high-reflection interference areas; The high-reflection interference area refers to an area where the reflectance is significantly higher than the normal range of the vegetation due to the leaf wax layer, water accumulation, or specific materials. The preset visible light band is usually selected as the band sensitive to specular reflection, and the preset threshold is dynamically set according to the vegetation type.

[0075] In the embodiments of the present application, based on the spectral reflection values of each spatial coordinate point in the multispectral reflection data, the spectral intensity of each spatial coordinate point in the preset visible green light band is extracted, and the optimal segmentation threshold of the current scene is automatically calculated through a dynamic threshold segmentation algorithm. The Otsu algorithm is used to maximize the between-class variance, and the spatial coordinate points with reflection intensity exceeding the threshold are marked as candidate interference points. Subsequently, adjacent candidate points are merged through morphological closing operations to generate a continuous high-reflection interference region.

[0076] 502. For the high-reflection interference region, control the polarization filter assembly of the flying device to sequentially switch to multiple different polarization angle combinations; at each polarization angle combination, re-collect the spectral reflection values of the high-reflection interference region to obtain multiple sets of spectral reflection data with different polarization angles. The polarization angle combination refers to the rotation angle setting of the polarization filter. Usually, four groups of angles, 0°, 45°, 90°, and 135°, cover the full polarization direction, which is used to suppress specular reflection noise from multiple angles.

[0077] In the embodiments of the present application, for the high-reflection interference region, after the flying device receives the coordinates of the high-reflection region, control the polarization filter assembly to sequentially switch to a preset number of different polarization angle combinations. At each polarization angle combination, for each angle, the unmanned aerial vehicle hovers directly above the target area, and a stable attitude is maintained through a PID controller and RTK positioning to ensure that the positioning error ≤ 0.1 m. The hyperspectral imager collects spectral reflection values at a rate of 50 frames per second. After each angle switch, a complete area scan is completed to generate multiple sets of spectral reflection data sets with different polarization angles.

[0078] 503. Perform superposition processing on the spectral reflection values of the same spatial coordinate point in the multiple sets of spectral reflection data with different polarization angles, and retain the minimum value of the intensity of the spectral reflection value after superposition as the spectral reflection value of the spatial coordinate point after specular reflection suppression. Superposition processing refers to the fusion calculation of the spectral reflection values of the same spatial coordinate point at multiple polarization angles. Retaining the minimum value can effectively suppress the residual signal of specular reflection.

[0079] In the embodiments of the present application, the multiple sets of spectral reflection data with different polarization angles are aligned and superposed according to spatial coordinates, and the spectral reflection values after superposition are retained. The reflection value sequences for each coordinate point are sorted, and the minimum value of the intensity is extracted as the spectral reflection value of the spatial coordinate point after specular reflection suppression. For example, if the reflection values of a certain coordinate point at 0°, 45°, 90°, and 135° are 70%, 45%, 60%, and 50% respectively, then 45% is selected as the reflection value after specular reflection suppression. This process traverses all coordinate points to generate a minimum reflection value matrix for the high-reflection region.

[0080] 504. Merge the spectrally reflectance values after anti-reflection with the original spectrally reflectance values in areas not marked as high-reflection interference areas according to spatial coordinates to generate multi-spectral reflectance data after anti-reflection.

[0081] Spatial coordinate merging means seamlessly stitching the processed high-reflection area data with the original data to generate a complete and noise-suppressed multi-spectral reflectance data set.

[0082] In the embodiments of the present application, based on the georegistration algorithm, the spectrally reflectance values after anti-reflection are merged with the original spectrally reflectance values in areas not marked as high-reflection interference areas according to spatial coordinates. For the coordinate points at the edge of the high-reflection area, a bilinear interpolation algorithm is used for smooth transition to avoid data jumps. For example, the reflectance value of the edge point is calculated by distance weighting of the reflectance values of four adjacent non-interference points. The finally generated global reflectance database contains multi-spectral reflectance data after unified calibration.

[0083] The following is a specific example: In the coniferous forest canopy monitoring scenario, in step 501, a high-reflection interference area caused by the wax layer of pine needles is detected, and the reflectance in the green light band reaches 68%, exceeding the preset threshold of 55%. In step 502, the polarization filter is controlled to re-collect the target area at four groups of angles of 0°, 60°, 120°, and 180°. The hovering accuracy of the unmanned aerial vehicle is controlled within 0.1 m, and the acquisition duration for each group of angles is 30 s. In step 503, the minimum value of 50% is taken from the four groups of reflectance values of 75%, 50%, 65%, and 55% at a certain coordinate point. After anti-reflection, the signal-to-noise ratio of the chlorophyll fluorescence signal in this area is increased from 4.2 to 9.5. In step 504, the data is merged through bilinear interpolation, and the generated global reflectance database shows that the noise in the high-reflection area is reduced by 82%, and the extraction error of photosynthetic parameters is reduced from 18% to 5%.

[0084] This solution accurately locates high-reflection interference areas through dynamic threshold segmentation, suppresses specular reflection noise through multi-angle polarization acquisition, retains the effective diffuse reflection signal of vegetation through the minimum value superposition algorithm, and generates a global high-precision reflectance database by combining interpolation fusion technology. Finally, it significantly improves the inversion accuracy of photosynthesis characteristic parameters in complex canopy environments, providing an anti-interference data basis for ecological assessment.

[0085] In some embodiments, the spatio-temporal coverage characteristics include a time-stamp sequence and a spatial coordinate sequence; Deploy multiple ground verification points in the scanning area, and in combination with the spatio-temporal coverage characteristics of the flight trajectory, collect the three-dimensional morphological parameters of the vegetation canopy, and perform spatio-temporal calibration and matching on the three-dimensional morphological parameters and the multi-spectral reflectance data to construct a reflectance database, including: 601. Arrange multiple ground verification points within the scanning area according to a preset density. Each ground verification point is equipped with a three-dimensional laser scanning device. The three-dimensional laser scanning device measures the depth detection data of the vegetation canopy by emitting laser pulses and receiving reflected signals; based on the depth detection data, calculate the height values, gap densities, and surface curvature values of each measurement point to form a three-dimensional morphological parameter set of the ground verification points. A ground verification point refers to a fixed monitoring node evenly arranged under the vegetation canopy according to a preset density. Each node is equipped with a three-dimensional laser scanning device. Depth detection data refers to the physical parameters of the vegetation canopy calculated by the time difference between laser pulse emission and reflected signal reception, including height values, gap densities, and surface curvature values. The height value is the vertical distance from the ground surface to the bottom of the canopy, the gap density is the proportion of leaf gaps in the unit volume, and the surface curvature value is a quantitative index of the curvature of the leaf or branch surface.

[0086] In the embodiment of the present application, multiple ground verification points are arranged in a 10-meter by 10-meter grid within the scanning area, and a three-dimensional laser scanner is deployed at each ground verification point. The three-dimensional laser scanner emits 905-nanometer laser at a frequency of 1000 pulses per second, and measures the depth detection data of the vegetation canopy by the time-of-flight method after receiving the reflected signal. Denoise and interpolate the original depth detection data, extract the height values of each measurement point, and control the accuracy within plus or minus 2 centimeters. When calculating the gap density, divide the leaf projection area by the scanning area and take the complement. Calculate the surface curvature value through a local surface fitting algorithm, and form a data table containing three-dimensional coordinates and morphological parameters.

[0087] 602. Extract the acquisition timestamps and spatial coordinates of each spectral sampling point in the multi-spectral reflection data. Based on the timestamp sequence and spatial coordinate sequence of the flight trajectory, establish the corresponding position markers between the spatial coordinates of the spectral sampling points and the canopy height values in the three-dimensional morphological parameters. The timestamp sequence refers to the time marker sequence during the acquisition of multi-spectral reflection data, and the spatial coordinate sequence refers to the longitude, latitude, and elevation coordinate sequence of the flight device trajectory. The corresponding position marker refers to associating the spectral sampling points and the ground morphological parameters to the same spatial coordinate through spatio-temporal alignment.

[0088] In the embodiments of the present application, the acquisition time and spatial coordinates of each spectral sampling point are extracted from the multispectral reflection data. Based on the timestamp sequence and spatial coordinate sequence of the flight trajectory, a linear interpolation algorithm is used to align the timestamps. For example, if the timestamp of a spectral sampling point is 10:00:30, the data with a scanning time of 10:00:28, which is the closest to the ground verification point in terms of matching distance, is selected. The longitude and latitude of the flight trajectory are converted into planar coordinates through coordinate transformation, and the matching error is controlled within 0.1 meter. The corresponding position markers between the spatial coordinates of the spectral sampling points and the canopy height values in the three-dimensional morphological parameters are established to ensure the spatial consistency between the spectral sampling points and the ground morphological parameters.

[0089] 603. According to the corresponding position markers, the canopy height values, canopy gap density values, and canopy surface curvature values in the three-dimensional morphological parameters are bound to the spectral reflection values in the multispectral reflection data to obtain bound data units. Position binding means establishing a one-to-one correspondence between the canopy morphological parameters and the spectral reflection values under the same spatial coordinates to form multi-dimensional data units.

[0090] In the embodiments of the present application, a hash table storage relationship is established with the corresponding position markers as the unique identifiers. The keys are the three-dimensional coordinates X, Y, and Z, and the values are the morphological parameter height H, gap density D, curvature C, and spectral reflection values R1, R2 to Rn. For example, the coordinates X102.3, Y45.6, and Z5.2 correspond to a height value of 5.2 meters, a gap density of 0.15, a curvature of 0.08, and spectral reflection values of 0.63 and 0.55. The canopy height values, canopy gap density values, and canopy surface curvature values in the three-dimensional morphological parameters are spatially bound to the spectral reflection values in the multispectral reflection data to form a one-to-one correspondence. All coordinate points are traversed to complete the binding, generating a set of data units containing air-ground multi-source attributes.

[0091] 604. Based on the preset binding rules, the bound data units are stored in the order of the grid distribution of the spatial coordinates to generate a reflection database.

[0092] The grid distribution order means dividing the spatial coordinates into regular grids according to a preset resolution and storing the data in the row-column order. The binding rules include data compression format, redundancy check, and index table construction strategy.

[0093] In the embodiments of the present application, based on binding rules such as a preset data compression format, redundancy check, and index table construction strategy, the bound data units are stored in the order of the grid distribution of spatial coordinates. A 0.1-meter resolution is divided into three-dimensional grids, and each grid stores the morphological parameters and spectral reflectance values of the corresponding coordinate points. The Z-order curve is used to spatially encode the grids to generate a global index table. The data is compressed and saved in a columnar storage format, and a CRC check code is added at the same time, and finally a reflection database that supports fast spatial queries is generated.

[0094] The following is a specific example: In a monitoring scenario of a wetland willow forest, the system arranges 15 ground verification points at 20-meter intervals. The three-dimensional laser scanner equipped at each node scans the canopy at a pulse frequency of 1,200 times per second, and measures the height value range from 2.8 meters to 6.5 meters, the gap density from 0.10 to 0.28, and the surface curvature from 0.03 to 0.12. The unmanned aerial vehicle synchronously collects multispectral data. Through timestamp alignment and UTM coordinate projection, the spatial matching error between the spectral sampling points and the ground morphological parameters is controlled within 0.05 meters. At coordinates X120.5, Y85.3, and Z4.2, a height of 4.2 meters, a gap density of 0.15, a curvature of 0.08, and spectral reflectance values of 0.62 and 0.58 are bound to form a multi-dimensional data unit. Finally, a reflection database is generated by grid storage at a resolution of 0.05 meters, which supports fast retrieval of middle canopy data within the input elevation range of 4.0 to 4.5 meters, shows that the light energy absorption value in the area with a gap density of 0.15 to 0.20 increases by 25%, and accurately locates the photosynthetic efficiency hotspots.

[0095] Through the high-precision three-dimensional morphological parameter acquisition of the ground verification points and the spatio-temporal calibration of the air-ground data in this solution, the precise binding of the canopy physical attributes and spectral reflectance values is realized. Grid storage and index optimization improve the data retrieval efficiency, provide a highly consistent multi-source database for the ecological analysis of the vertical profile of the vegetation canopy, and significantly enhance the reliability of the inversion of photosynthesis characteristic parameters in complex environments.

[0096] In some embodiments, the signal intensity change rate of the spectral signals at different levels in the vertical profile in the reflection database is extracted, and the corresponding relationship between the signal intensity change rate and the light energy absorption value in the photosynthesis characteristic parameters is established, including: 701. Extract the spectral signal intensity values of adjacent penetration levels at the same spatial coordinate point from the reflection database, and calculate the ratio of the spectral signal intensity value of the latter penetration level to the previous level as the signal intensity change rate between levels; The signal intensity change rate between levels refers to the ratio of the spectral signal intensities of adjacent penetration levels at the same spatial coordinate point, and is used to quantify the attenuation degree of light waves during vertical penetration. For example, if the middle layer signal intensity is 800 and the top layer is 1000, the change rate is 0.8.

[0097] In the embodiments of the present application, the spectral signal intensity values of adjacent penetration levels are queried from the reflection database according to spatial coordinates. For example, for a coordinate point, the spectral signal intensity value of the top penetration level is extracted as 1200 units, the middle penetration level is 900 units, and the bottom penetration level is 720 units. Calculate the signal intensity ratio of the middle penetration level to the top level as 900 / 1200 = 0.75, and the ratio of the bottom level to the middle level as 720 / 900 = 0.8. Traverse all spatial coordinate points, calculate the ratio of the spectral signal intensity value of the latter penetration level to the previous level, and generate a signal intensity change rate matrix between levels. This process optimizes the query efficiency through database indexing to ensure that more than 100,000 coordinate points of data can be processed per second.

[0098] 702. Based on the signal intensity change rate between levels, calculate the product of the signal intensity change rates between all upper levels from the top of the canopy to the bottom layer by layer in the order of penetration levels to obtain the cumulative signal intensity change rate; The cumulative signal intensity change rate refers to the product of the signal intensity change rates between all levels from the top of the canopy to the bottom layer, which reflects the total attenuation degree of light waves penetrating the entire canopy. For example, the change rate from the top layer to the middle layer is 0.8, and the change rate from the middle layer to the bottom layer is 0.7, then the cumulative change rate is 0.8×0.7 = 0.56.

[0099] In the embodiments of the present application, the signal intensity change rate data between levels are read layer by layer from the top layer to the bottom layer in the order of penetration levels. For each spatial coordinate point, calculate the product of the signal intensity change rates between all upper levels from the top of the canopy to the bottom layer layer by layer in the order of penetration levels to generate the cumulative signal intensity change rate. For example, the change rate from the top layer to the middle layer of a certain coordinate point is 0.75, and the change rate from the middle layer to the bottom layer is 0.8, then the cumulative change rate is 0.75×0.8 = 0.6. A parallel computing framework is used to accelerate large-scale data operations. For example, a GPU is used to calculate millions of coordinate points simultaneously to generate a three-dimensional distribution map of the cumulative change rate. This distribution map reflects the overall attenuation characteristics of light waves when penetrating the canopy.

[0100] 703. Through a preset experimental calibration method, in the vegetation canopy corresponding to the ground verification point, obtain the light energy absorption value per unit area actually absorbed at different penetration levels through a light intensity attenuation measurement device; The light intensity attenuation measurement device is a device deployed at the ground verification point, which measures the actual light energy absorption value at different penetration levels through a quantum sensor, and the unit is μmol / m² / s.

[0101] In the embodiments of the present application, through a preset experimental calibration method, an optical intensity attenuation measurement device, such as a quantum sensor array, is deployed in the vegetation canopy corresponding to the ground verification point. The incident light intensity at the top of the canopy, the transmitted light intensity in the middle layer, and the transmitted light intensity at the bottom layer are measured in the calibration area respectively. For example, the incident light intensity at the top layer is 2000 μmol / m² / s, the transmitted light intensity in the middle layer is 1200 μmol / m² / s, and the transmitted light intensity at the bottom layer is 960 μmol / m² / s. According to the obtained light energy absorption value per unit area actually absorbed at different penetration levels, the light energy absorption value per unit area is calculated as the difference between the incident and transmitted light intensities. The absorption value in the middle layer is 2000 - 1200 = 800 μmol / m² / s, and the bottom layer is 1200 - 960 = 240 μmol / m² / s. By repeating the measurement multiple times and taking the average value, the statistical significance of the experimental data is ensured.

[0102] 704. Fit the cumulative signal intensity change rate and the measured light energy absorption value in the order of penetration levels with an exponential relationship, and output the exponential relationship fitting result. Exponential relationship fitting refers to establishing a mathematical relationship between the cumulative change rate X and the measured light energy absorption value Y through a non - linear regression model, usually in the form of Y = a×e^(bX).

[0103] In the embodiments of the present application, the cumulative signal intensity change rate and the measured light energy absorption value are aligned in the order of penetration levels and input into a non - linear regression model. The Levenberg - Marquardt algorithm is used to fit the exponential function Y = a×e^(bX), where X is the cumulative change rate and Y is the light energy absorption value. For example, inputting the data pairs 0.6, 800 and 0.48, 1000, the fitting result is Y = 2500×e^(-1.5X), and the determination coefficient R² = 0.97. Output the exponential relationship fitting result, and evaluate the generalization ability of the model through a cross - validation method. For example, divide 80% of the data for training and 20% for testing to ensure that the prediction error ≤ 5%.

[0104] 705. Based on the exponential relationship fitting result, generate a corresponding mapping table of the cumulative signal intensity change rate and the light energy absorption value as the corresponding relationship between the signal intensity change rate and the light energy absorption value in the photosynthesis characteristic parameters.

[0105] The corresponding mapping table is a look - up table storing the relationship between the cumulative signal intensity change rate and the light energy absorption value, which is used to quickly convert the spectral signal into a physical absorption value.

[0106] In the embodiments of the present application, based on the fitting result of the exponential relationship, a corresponding mapping table between the cumulative signal intensity change rate and the light energy absorption value is generated. For example, the cumulative change rate is divided at intervals of 0.01, and the corresponding light energy absorption values are calculated. When the cumulative signal intensity change rate is 0.6, the light energy absorption value is 2500×e^(-1.5×0.6)=1015 μmol / m² / s. The mapping table is stored using a hash table data structure, with the key being the change rate (precision 0.01) and the value being the absorption value. For change rates that do not match exactly, a linear interpolation algorithm is used to estimate the absorption value. For example, for a change rate of 0.605, interpolation is calculated as 0.60 corresponding to 1015, 0.61 corresponding to 990, and 0.605 corresponding to 1002.5. The query efficiency is optimized to support millions of real-time conversions per second, serving as the corresponding relationship between the signal intensity change rate and the light energy absorption value in the photosynthesis characteristic parameters.

[0107] The following is a specific example: In a tropical rainforest ecological restoration project, the system monitors the canopy layer of Ficus trees in the degraded area: The signal intensities of three penetration levels at a certain coordinate point are extracted through step 701. The top layer is 1800 units, the middle layer is 1260 units, and the bottom layer is 882 units. The change rate of the middle layer / top layer is calculated as 0.7, and the change rate of the bottom layer / middle layer is 0.7. The cumulative signal intensity change rate is 0.7×0.7 = 0.49. In step 703, the incident light intensity measured by the light intensity attenuation measurement device at the top of the canopy is 2200 μmol / m² / s, the transmitted light intensity in the middle layer is 1320 μmol / m² / s, and the transmitted light intensity in the bottom layer is 924 μmol / m² / s. The corresponding absorption value in the middle layer is 880 μmol / m² / s, and the absorption value in the bottom layer is 396 μmol / m² / s. In step 704, a non-linear regression fit is used to obtain the exponential relationship formula Y = 2400×e^(-1.8X), and the determination coefficient R² = 0.98, indicating that the model is highly reliable. In step 705, a mapping table is generated. The cumulative change rate of 0.49 corresponds to a light energy absorption value of 2400×e^(-1.8×0.49)=2400×0.402≈965 μmol / m² / s. Substituting this value into the three-dimensional map analysis shows that the actual light energy utilization rate in the middle layer of the degraded area is increased by 28% compared with the traditional linear model, and the evaluation error is reduced from 25% to 4%. Based on this, the project team accurately locates 3 areas with insufficient photosynthetic efficiency and implements targeted forest gap thinning and soil improvement. After 6 months of monitoring, it shows that the average light energy absorption value in the target area is increased by 35%, verifying the effectiveness of the present solution in the ecological restoration of complex canopies.

[0108] Through physical modeling and experimental calibration of signal attenuation between layers, this solution establishes an exponential mapping relationship for vertical light energy absorption, breaking through the accuracy limitations of traditional linear models. The product operation of the cumulative signal intensity change rate quantifies the overall attenuation law of light waves penetrating the canopy. Combining with the exponential relationship fitted from ground measured data significantly improves the physical consistency between light energy absorption values and spectral signals. The generated dynamic mapping table supports real-time and efficient conversion, enabling three-dimensional precise analysis of the photosynthetic efficiency of the vegetation canopy, with the error rate reduced compared to traditional methods. This technology provides targeted decision-making basis for scenarios such as degraded ecological restoration and high-density agricultural planting. For example, it can accurately identify weak photosynthesis areas and guide intervention measures such as thinning and irrigation, ultimately promoting the scientific and refined management of ecology and agriculture and contributing to the achievement of sustainable development goals.

[0109] Figure 2 The following is a schematic structural diagram of a garden plant environmental ecological assessment system provided by an embodiment of this application, as Figure 2 shown. This system includes: A collection module 21, configured to perform hierarchical scanning on the garden plant community through a flight device, generate a flight trajectory based on the three-dimensional density distribution of the vegetation canopy, and synchronously collect multi-spectral reflection data including photosynthesis characteristic parameters; The collection module 21 is further configured to arrange multiple ground verification points within the scanning area, collect three-dimensional morphological parameters of the vegetation canopy in combination with the spatio-temporal coverage characteristics of the flight trajectory, perform spatio-temporal calibration and matching on the three-dimensional morphological parameters and the multi-spectral reflection data, and construct a reflection database; A generation module 22, configured to identify a high-reflection interference area according to the detected surface reflection characteristics of the vegetation canopy, perform multi-angle polarization compensation on the high-reflection interference area by adjusting the polarization angle configuration, and generate multi-spectral reflection data after suppressing the reflected light; A calculation module 23, configured to perform hierarchical reflectance weighted calculation on the multi-spectral reflection data after suppressing the reflected light based on the penetration layer identifier in the reflection database, and associate the attenuation gradient of different multi-spectral reflection data in the vertical profile of the vegetation canopy with the mapping relationship of photosynthesis characteristic parameters, and generate a three-dimensional distribution map of the photosynthetic efficiency of the plant community; A correction module 24, configured to control the flight device to perform secondary scanning according to the photosynthetic efficiency abnormal area in the three-dimensional distribution map, correct the evaluation error caused by the delay of root water absorption in the three-dimensional distribution map by fusing the multi-spectral reflection data of the secondary scanning and the soil moisture parameters of the ground verification points, and output an ecological evaluation result.

[0110] Figure 2 The described garden plant environmental ecological assessment system can execute Figure 1For a method for ecological assessment of garden plant environment described in the illustrated embodiment, its implementation principle and technical effects will not be elaborated further. For a system for ecological assessment of garden plant environment in the above embodiment, the specific manners in which each module and unit perform operations have been described in detail in the embodiment related to this method, and will not be elaborated here.

[0111] In a possible design, Figure 2 A system for ecological assessment of garden plant environment in the illustrated embodiment can be implemented as a computing device, such as Figure 3 shown, the computing device may include a storage component 31 and a processing component 32; The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32.

[0112] The processing component 32 is used for the Figure 1 method for ecological assessment of garden plant environment in the above

[0113] embodiment. Among them, the processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components for executing the above method.

[0114] The storage component 31 is configured to store various types of data to support operations on the terminal. The storage component may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0115] Of course, the computing device may also necessarily include other components, such as an input / output interface, a display component, a communication component, etc.

[0116] The input / output interface provides an interface between the processing component and the peripheral interface module, and the above peripheral interface module may be an output device, an input device, etc.

[0117] The communication component is configured to facilitate wired or wireless communication between the computing device and other devices, etc.

[0118] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server. The above-mentioned processing components, storage components, etc. can be basic server resources leased or purchased from a cloud computing platform.

[0119] The embodiments of the present application also provide a computer storage medium storing a computer program, and when the computer program is executed by a computer, it can implement the above-mentioned Figure 1 a method for ecological evaluation of the environment of garden plants shown in the embodiments.

[0120] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the foregoing method embodiments, and will not be described herein again.

[0121] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0122] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0123] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit them; although the present application 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 make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for evaluating the environmental ecology of garden plants, characterized in that, Including: Implementing hierarchical scanning of the garden plant community through a flying device, generating a flight trajectory based on the three-dimensional density distribution of the vegetation canopy, and synchronously collecting multi-spectral reflection data including photosynthesis characteristic parameters; Deploying multiple ground verification points within the scanning area, collecting three-dimensional morphological parameters of the vegetation canopy in combination with the spatio-temporal coverage characteristics of the flight trajectory, performing spatio-temporal calibration and matching on the three-dimensional morphological parameters and the multi-spectral reflection data, and constructing a reflection database; Identifying high-reflection interference regions according to the detected surface reflection characteristics of the vegetation canopy, and adopting multi-angle polarization compensation for the high-reflection interference regions by adjusting the polarization angle configuration to generate multi-spectral reflection data after reflection suppression; Based on the penetration level identification in the reflection database, performing hierarchical reflectance weighting calculation on the multi-spectral reflection data after reflection suppression, and associating the attenuation gradient of different multi-spectral reflection data in the vertical profile of the vegetation canopy with the mapping relationship of photosynthesis characteristic parameters to generate a three-dimensional distribution map of the photosynthetic efficiency of the plant community; According to the photosynthetic efficiency abnormal regions in the three-dimensional distribution map, controlling the flying device to perform secondary scanning, and correcting the evaluation error caused by the delayed water absorption of the root system in the three-dimensional distribution map by fusing the multi-spectral reflection data of the secondary scanning and the soil moisture parameters of the ground verification points, and outputting an ecological evaluation result.

2. The method according to claim 1, wherein Based on the penetration level identification in the reflection database, performing hierarchical reflectance weighting calculation on the spectrum data after reflection suppression, including: Extracting the hierarchical segmentation information corresponding to the penetration level identification from the reflection database, where the hierarchical segmentation information includes the penetration ability value and the hierarchical thickness value corresponding to each level in the vegetation canopy; According to the penetration ability value, performing weight assignment on the spectrum signals of each level in the spectrum data after reflection suppression; Based on the hierarchical thickness value, performing superposition calculation on the spectrum signals with assigned weights in the hierarchical order to obtain the reflectance weighting result of each level in the vertical direction.

3. The method according to claim 2, characterized in that, Based on the reflectance weighting result, associating the attenuation gradient of different spectrum data in the vertical profile of the vegetation canopy with the mapping relationship of photosynthesis characteristic parameters to generate a three-dimensional distribution map of the photosynthetic efficiency of the plant community, including: Extracting the signal intensity change rate of the spectrum signals of different levels in the vertical profile in the reflection database as the penetration level changes, and establishing the corresponding relationship between the signal intensity change rate and the light energy absorption value in the photosynthesis characteristic parameters; According to the corresponding relationship, converting the spectrum signals of each level in the reflectance weighting result into corresponding light energy absorption values; Based on the spatial position of the hierarchical segmentation information, mapping the light energy absorption values in the hierarchical order into three-dimensional space to form a three-dimensional distribution map representing the photosynthetic efficiency distribution of the plant community.

4. The method according to claim 1, characterized in that, According to the photosynthetic efficiency abnormal regions in the three-dimensional distribution map, controlling the flying device to perform secondary scanning, and correcting the evaluation error caused by the delayed water absorption of the root system in the three-dimensional distribution map by fusing the multi-spectral reflection data of the secondary scanning and the soil moisture parameters of the ground verification points, and outputting an ecological evaluation result, including: Extract continuous spatial regions in the three-dimensional distribution map where the photosynthetic efficiency value is lower than or higher than the preset normal range, and mark them as photosynthetic efficiency abnormal regions; According to the spatial distribution range of the photosynthetic efficiency abnormal regions, adjust the flight height and scanning path density of the flying device, generate a secondary scanning trajectory covering the photosynthetic efficiency abnormal regions, and control the flying device to perform secondary scanning according to the secondary scanning trajectory; Extract the multi-spectral reflection data of the photosynthetic efficiency abnormal regions in the secondary scanning, and obtain the spectral reflection values of each spatial coordinate point in the abnormal regions; Extract the soil moisture parameters corresponding to the spatial coordinates of the abnormal regions from the ground verification points, where the soil moisture parameters are the measured values of the soil moisture content at the ground verification points within a preset time window; Bind the measured values of the soil moisture content and the spectral reflection values of the abnormal regions one-to-one according to the spatial coordinates; According to the measured values of the soil moisture content, perform moisture absorption delay compensation on the spectral reflection values, and replace the photosynthetic efficiency values corresponding to the photosynthetic efficiency abnormal regions in the three-dimensional distribution map with the compensated spectral reflection values. Based on the replaced photosynthetic efficiency values, recalculate the photosynthetic efficiency distribution of all spatial coordinate points in the three-dimensional distribution map, and generate an ecological assessment result.

5. The method according to claim 1, characterized in that, According to the detected reflection characteristics of the vegetation canopy surface, identify high-reflection interference regions, and through adjusting the polarization angle configuration, perform multi-angle polarization compensation on the high-reflection interference regions to generate multi-spectral reflection data after suppressing the specular reflection, including: Based on the spectral reflection values of each spatial coordinate point in the multi-spectral reflection data, extract the reflection intensity of the vegetation canopy surface within a preset visible light band, and mark the spatial coordinate points with the reflection intensity exceeding the preset threshold as high-reflection interference regions; For the high-reflection interference regions, control the polarization filter component of the flying device to sequentially switch to multiple different polarization angle combinations; At each polarization angle combination, re-collect the spectral reflection values in the high-reflection interference regions to obtain multiple sets of spectral reflection data with different polarization angles; Perform superposition processing on the spectral reflection values of the same spatial coordinate point in the multiple sets of spectral reflection data with different polarization angles, and retain the minimum value of the intensity of the superposed spectral reflection values as the spectral reflection value after suppressing the specular reflection of the spatial coordinate point; Merge the spectral reflection values after suppressing the specular reflection and the original spectral reflection values that are not marked as high-reflection interference regions according to the spatial coordinates to generate multi-spectral reflection data after suppressing the specular reflection.

6. The method according to claim 1, wherein The spatio-temporal coverage characteristics include a time stamp sequence and a spatial coordinate sequence; Arrange multiple ground verification points in the scanning area, combine the spatio-temporal coverage characteristics of the flight trajectory, collect the three-dimensional morphological parameters of the vegetation canopy, and perform spatio-temporal calibration and matching on the three-dimensional morphological parameters and the multi-spectral reflection data to construct a reflection database, including: A plurality of ground verification points are arranged in the scanning area according to a preset density. Each ground verification point is equipped with a three-dimensional laser scanning device. The three-dimensional laser scanning device measures the depth detection data of the vegetation canopy by emitting laser pulses and receiving reflected signals; based on the depth detection data, the height values, gap densities, and surface curvature values of each measurement point are calculated to form a three-dimensional morphological parameter set of the ground verification points; The acquisition timestamp and spatial coordinates of each spectral sampling point in the multi-spectral reflection data are extracted. Based on the timestamp sequence and spatial coordinate sequence of the flight trajectory, a corresponding position mark between the spatial coordinates of the spectral sampling point and the canopy height value in the three-dimensional morphological parameters is established; According to the corresponding position mark, the canopy height value, canopy gap density value, and canopy surface curvature value in the three-dimensional morphological parameters are positionally bound to the spectral reflection value in the multi-spectral reflection data to obtain a bound data unit; Based on a preset binding rule, the bound data units are stored in the order of the grid distribution of the spatial coordinates to generate a reflection database.

7. The method according to claim 3, wherein Extract the signal intensity change rate of the spectral signals at different levels in the vertical section with the change of the penetration level in the reflection database, and establish the corresponding relationship between the signal intensity change rate and the light energy absorption value in the photosynthesis characteristic parameters, including: Extract the spectral signal intensity values of adjacent penetration levels at the same spatial coordinate point from the reflection database, and calculate the ratio of the spectral signal intensity value of the latter penetration level to the previous level as the signal intensity change rate between levels; Based on the signal intensity change rate between levels, calculate the continuous product of the signal intensity change rates between all upper levels from the top to the bottom of the canopy layer by layer in the order of the penetration level to obtain the cumulative signal intensity change rate; Through a preset experimental calibration method, in the vegetation canopy corresponding to the ground verification point, the light energy absorption value per unit area actually absorbed at different penetration levels is obtained through a light intensity attenuation measurement device; Perform an exponential relationship fitting on the cumulative signal intensity change rate and the measured light energy absorption value in the order of the penetration level, and output the exponential relationship fitting result; Based on the exponential relationship fitting result, generate a corresponding mapping table between the cumulative signal intensity change rate and the light energy absorption value as the corresponding relationship between the signal intensity change rate and the light energy absorption value in the photosynthesis characteristic parameters.

8. An environmental ecological evaluation system for garden plants, characterized in that, Including: An acquisition module for performing layered scanning on the garden plant community through a flight device, generating a flight trajectory based on the three-dimensional density distribution of the vegetation canopy, and synchronously acquiring multi-spectral reflection data containing photosynthesis characteristic parameters; The acquisition module is further configured to arrange a plurality of ground verification points in the scanning area, collect the three-dimensional morphological parameters of the vegetation canopy in combination with the spatio-temporal coverage characteristics of the flight trajectory, perform spatio-temporal calibration and matching on the three-dimensional morphological parameters and the multi-spectral reflection data, and construct a reflection database; A generation module for identifying a high-reflection interference area according to the surface reflection characteristics of the detected vegetation canopy, and performing multi-angle polarization compensation on the high-reflection interference area by adjusting the polarization angle configuration to generate multi-spectral reflection data after reflection suppression; A calculation module, configured to perform hierarchical reflectance weighted calculation on the multi-spectral reflection data after specular reflection suppression based on the penetration layer identification in the reflection database, and associate the attenuation gradient of different multi-spectral reflection data in the vertical profile of the vegetation canopy with the mapping relationship of photosynthesis characteristic parameters, so as to generate a three-dimensional distribution map of the photosynthetic efficiency of the plant community; A correction module, configured to control the flying device to perform secondary scanning according to the photosynthetic efficiency abnormal area in the three-dimensional distribution map, correct the evaluation error caused by the delay of root water absorption in the three-dimensional distribution map by fusing the multi-spectral reflection data of the secondary scanning and the soil moisture parameters of the ground verification points, and output an ecological evaluation result.

9. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a method for ecological evaluation of the environment of garden plants according to any one of claims 1 to 7.

10. A computer storage medium, characterized in that, A computer program is stored, and when the computer program is executed by a computer, a method for ecological evaluation of the environment of garden plants according to any one of claims 1 to 7 is implemented.

Images