Ecological restoration project acceptance and evaluation method based on unmanned aerial vehicle three-dimensional model interpretation

Abstract

The application relates to the technical field of data processing, in particular to an ecological restoration project acceptance and evaluation method based on unmanned aerial vehicle three-dimensional model interpretation, which comprises the following steps: obtaining data changes of each characteristic parameter of each scene unit through unmanned aerial vehicle three-dimensional model data of an ecological restoration area, determining a deterioration trend degree of each characteristic parameter by means of a data change speed of each characteristic parameter before restoration; evaluating response intensities of each characteristic parameter at each moment, calculating a quick response stability index and a gradual response of the characteristic parameter respectively; judging quick response conditions of each characteristic parameter, setting different weights for the quick response stability index and the gradual response, combining the deterioration trend degree to obtain an overall evaluation index of the scene unit, so as to perform acceptance evaluation; different modes of evaluation are performed on characteristic parameters with different response characteristics, comprehensive judgment and evaluation are performed, accurate evaluation of the ecological restoration project is realized, and the accuracy of the ecological restoration project evaluation is improved.

Description

Technical Field

[0001] This application relates to the field of data processing technology, specifically to a method for acceptance and evaluation of ecological restoration projects based on the interpretation of UAV 3D models. Background Technology

[0002] With the continuous advancement of ecological civilization construction and land spatial governance, ecological restoration projects involving mountains, rivers, forests, fields, lakes, grasslands, and deserts are expanding in terms of spatial scope, investment scale, and implementation cycle. These projects are increasingly characterized by multi-scenario, multi-stage, and long-term operation. The acceptance and evaluation of ecological restoration projects are no longer limited to confirming project completion; they have become crucial bases for project management, financial performance evaluation, and subsequent maintenance decisions. Against this backdrop, 3D models acquired by drones can comprehensively reflect the spatial morphology and ecological state changes of the restoration area. Through the interpretation and information processing of this data, continuous and objective state descriptions can be provided for ecological restoration projects, offering a data foundation and information support for project acceptance, evaluation, and management decisions, thereby meeting the needs of refined management and process supervision of ecological restoration projects.

[0003] The acceptance and evaluation of existing ecological restoration projects mainly rely on the summarization of phased results and manual judgment. Information sources are scattered, lacking unified time-series comprehensive analysis, making it difficult to conduct continuous and objective management and supervision of the restoration process and its effectiveness. In multi-scenario, multi-phase ecological restoration projects, the characteristic parameters of different types of ecological elements vary significantly. Existing management methods typically rely on static indicators or comparisons of single results, which fail to reflect the trends before and after restoration and the restoration response process. This leads to discrepancies between acceptance conclusions and the actual restoration status, reducing the accuracy of project performance evaluation and subsequent management decisions. Summary of the Invention

[0004] To address the aforementioned technical issues, this application provides a method for the acceptance and evaluation of ecological restoration projects based on the interpretation of 3D models from unmanned aerial vehicles (UAVs), thereby resolving existing problems.

[0005] The method for acceptance and evaluation of ecological restoration projects based on UAV 3D model interpretation in this application adopts the following technical solution:

[0006] One embodiment of this application provides a method for acceptance and evaluation of ecological restoration projects based on the interpretation of 3D models from unmanned aerial vehicles (UAVs). The method includes the following steps:

[0007] The system acquires 3D drone models of the ecological restoration area at each sampling time and identifies various feature parameter values ​​of each scene unit in the 3D drone model. The sampling time includes the sampling time before ecological restoration and the sampling time after restoration begins.

[0008] Based on the data change characteristics of each feature parameter within a preset time period before repair, the degree of degradation trend of each feature parameter is evaluated.

[0009] Based on the differences between the data change trends of each feature parameter before repair and the data change trends at each time point after repair begins, the response degree of each feature parameter at each time point after repair begins is constructed to analyze the rapid response of each feature parameter and obtain the rapid response stability index of each feature parameter; based on the cumulative characteristics of the differences of each feature parameter after repair begins and the response degree at the first time point after repair begins, the asymptotic responsiveness of each feature parameter is determined.

[0010] The overall evaluation index of each scenario unit is determined based on the degree of degradation trend of all characteristic parameters, the rapid response stability index, and the progressive responsiveness, in order to conduct acceptance evaluation of the ecological restoration area.

[0011] In one embodiment, the process of obtaining the degree of degradation trend is as follows:

[0012] Calculate the range of values ​​of each feature parameter at all times within a preset time period before repair, obtain the fitted straight line of values ​​at all times through a linear fitting algorithm, and use the positive fusion result of the slope absolute value of the fitted straight line of each feature parameter and the range as the degree of degradation trend of each feature parameter of each scene unit.

[0013] In one embodiment, the process of obtaining the response level of each feature parameter at each time point after the repair begins is as follows:

[0014] Calculate the numerical difference between each feature parameter at each time after the start of repair and the previous time, and record the ratio of the difference to the time interval between two adjacent time points as the instantaneous recovery slope of each feature parameter at each time after the start of repair.

[0015] The instantaneous recovery slope of the repair is compared with the slope of the fitted line to determine the recovery indicator value of each feature parameter at each time point after the repair begins;

[0016] The response level of each feature parameter at each time point after the repair begins is determined based on the instantaneous recovery slope and the recovery identifier value.

[0017] In one embodiment, the process of obtaining the recovery identifier value is as follows:

[0018] If the instantaneous recovery slope has the same sign as the slope of the fitted line, then the recovery flag value is set to 0; otherwise, it is set to 1.

[0019] In one embodiment, the process of obtaining the response level is as follows:

[0020] The difference between the instantaneous recovery slope and the slope of the fitted line is mapped to a positive number; the product of the positive number and the recovery identifier value is used as the response degree of each feature parameter at each time point after the start of repair.

[0021] In one embodiment, the step of analyzing the rapid response of each characteristic parameter to obtain the rapid response stability index of each characteristic parameter specifically involves:

[0022] Calculate the difference between the data value of each feature parameter at each time point after the repair begins and the mean of the data values ​​at all times, and record it as the first difference; obtain the minimum response degree of each feature parameter in the historical local time period after the repair begins.

[0023] The fast response stability index of each characteristic parameter is negatively correlated with the first difference and the minimum response degree, respectively.

[0024] In one embodiment, the process of obtaining the asymptotic responsiveness of each feature parameter value is as follows:

[0025] The minimum recovery identifier value of each feature parameter in the historical local time period after the repair begins is used as the weight of the response degree of each feature parameter at each time point;

[0026] The rapid response degree of each feature parameter is determined based on the aforementioned response degree;

[0027] Obtain the weighted sum of the response levels at all times after the repair begins, and combine it with the fast response level to determine the asymptotic responsiveness of each feature parameter; the asymptotic responsiveness is directly proportional to the sum and inversely proportional to the fast response level.

[0028] In one embodiment, the fast response rate is: the response level of each feature parameter at the first moment after the repair begins.

[0029] In one embodiment, the process of obtaining the overall evaluation index is as follows:

[0030] The forward and reverse mapping values ​​of the fast response degree are used as the weights of the fast response stability index and the progressive responsiveness, respectively. The weighted sum of the fast response stability index and the progressive responsiveness is used as the repair evaluation index of each feature parameter.

[0031] The fusion value of the product of the degradation trend of all characteristic parameters of each scene unit and the repair evaluation index is used as the overall evaluation index of each scene unit.

[0032] In one embodiment, the acceptance assessment of the ecological restoration area specifically includes:

[0033] If the normalized value of the overall evaluation index of each scenario unit is greater than the preset evaluation threshold, then each scenario unit is judged to have passed the acceptance test; otherwise, each scenario unit is judged to have failed the acceptance test.

[0034] This application has at least the following beneficial effects:

[0035] When evaluating the restoration status of ecological restoration areas, this application acquires data fluctuation data of different characteristic parameters of different scene units through continuously captured UAV 3D model data, and conducts multi-dimensional data analysis for the scene. First, the degree of scene degradation trend reflected by each characteristic parameter is judged by the rate of change of data before restoration. Since the evaluation methods for different characteristic parameters differ during the evaluation process, this application evaluates the response intensity of each characteristic parameter at each time point, calculates the rapid response stability index and progressive responsiveness of each characteristic parameter, and judges the rapid response characteristics and progressive response characteristics of the characteristic parameters separately. The rapid response of each characteristic parameter is judged by the response intensity, and the rapid response stability index and progressive responsiveness of the characteristic parameters are calculated with different weights. Combined with the degree of degradation trend, an overall evaluation index of the scene unit is obtained. Different evaluation methods can be used for characteristic parameters with different response characteristics, and comprehensive judgment and evaluation can be carried out to achieve accurate assessment of ecological restoration projects. This improves the accuracy of ecological restoration project assessment by allowing for different processing methods. Attached Figure Description

[0036] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 A flowchart of the ecological restoration project acceptance and evaluation method based on UAV 3D model interpretation provided for this application;

[0038] Figure 2 This is a schematic diagram illustrating the process of obtaining the response level. Detailed Implementation

[0039] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, characteristic parameters, and effects of the ecological restoration project acceptance and evaluation method based on UAV 3D model interpretation proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific characteristic parameters, structures, or features in one or more embodiments can be combined in any suitable form.

[0040] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0041] The following, in conjunction with the accompanying drawings, details the specific scheme of the ecological restoration project acceptance and evaluation method based on UAV 3D model interpretation provided in this application.

[0042] One embodiment of this application provides a method for acceptance and evaluation of ecological restoration projects based on the interpretation of 3D models from unmanned aerial vehicles.

[0043] Specifically, the following methods for acceptance and evaluation of ecological restoration projects based on UAV 3D model interpretation are provided. Please refer to [link / reference]. Figure 1 The method includes the following steps:

[0044] Step S1: Obtain the UAV 3D model of the ecological restoration area at each sampling time, and identify the various feature parameter values ​​of each scene unit in the UAV 3D model; wherein, the sampling time includes the sampling time before ecological restoration and the sampling time after restoration begins.

[0045] For the restoration areas corresponding to ecological restoration projects, the ecological scene units within the restoration areas include mountain slopes, rivers and lakes, woodlands, grasslands, and desertified land. During the project initiation phase, based on the management scope and acceptance requirements of the ecological restoration project, the areas requiring data collection are determined, and these areas serve as the unified object for subsequent data collection phases.

[0046] To support the acceptance and evaluation of ecological restoration projects, the data collection time is divided into a pre-restoration stage and a post-restoration stage. In this embodiment, UAV 3D model data of the restoration area is acquired monthly. A neural network model is used to divide the collected UAV 3D model data into scene units, specifically into mountain slopes, rivers and lakes, woodlands, grasslands, and desertified areas. The scene unit recognition network adopts an encoder-decoder structure. The neural network model is used to extract different feature parameters for each scene unit: the feature parameters for mountain slopes include: average slope, slope standard deviation, surface roughness index, elevation undulation, bare land ratio, and vegetation coverage; the feature parameters for rivers and lakes include: water boundary curvature, river width variation rate, shoreline vegetation coverage, and riverbed elevation variation; the feature parameters for woodlands include: average canopy height, canopy height distribution width, canopy coverage, and forest gap ratio; the feature parameters for grasslands and desertified areas include: surface roughness, dune height or undulation, grassland coverage, and bare land ratio. In this embodiment, the neural network adopts a CNN + global pooling + multi-branch output structure. The implementer may also use other neural networks to identify the UAV 3D model data. This application does not impose any specific restrictions.

[0047] In other embodiments of this application, the implementer may set the acquisition frequency of the UAV 3D model data of the repair area according to the actual situation.

[0048] The time-series data of each feature parameter under each scene unit are normalized using the minimum-maximum normalization method to avoid differences in the units and dimensions of different feature parameters affecting subsequent calculations. Since the data acquisition time span for the repair area is relatively long, Seasonal and Trend Decomposition (STL) using Loess is employed to remove the seasonal term of the same feature parameter in the same scene unit over time. A moving average filtering algorithm is then used to smooth the time-series data of various feature parameters after removing the seasonal term to eliminate high-frequency noise interference. Both the STL decomposition algorithm and the moving average filtering algorithm are well-known, and their specific processes are not detailed here. In other embodiments of this application, implementers may also use other trend decomposition algorithms to remove the seasonal term of the feature parameters over time, and other filtering algorithms to filter the time-series data of the feature parameters.

[0049] Step S2: Based on the data change characteristics of each feature parameter within a preset time period before repair, assess the degree of degradation trend of each feature parameter.

[0050] Because ecological restoration projects involve multiple types of scene units and multidimensional characteristics, the sensitivity of different characteristic parameters to restoration measures varies significantly. The numerical changes of some characteristic parameters before and after restoration mainly exhibit minor fluctuations or long-term stability, resulting in low decision-making reference value for project acceptance and performance evaluation. Therefore, it is necessary to screen scene unit characteristic parameters based on their variation range, thereby excluding characteristic parameters that remain basically stable before restoration or whose variation range is approximately unchanged. Thus, a judgment on the significance of fluctuations is made for the restoration reference window.

[0051] In this embodiment, the most recent 6 months prior to the repair time point are used as the repair reference window. In other embodiments of this application, the implementer can set the length of the repair reference window according to the actual situation.

[0052] For the time-series data of feature parameters after removing seasonality, calculate the range of all data for each feature parameter of each scene unit within the repair reference window.

[0053] The smaller the range, the less obvious the change in the feature parameter before repair, and the less likely the feature parameter will reflect the repair effect when repair is performed later.

[0054] Since ecological degradation is usually not a random fluctuation, but rather a directional process of change influenced by long-term human activities, natural erosion, and cumulative disturbances, it is necessary to assess the degree of degradation trend in the pre-remediation stage in order to determine the true state of the object to be remediated before the implementation of remediation measures.

[0055] The time-series data of each feature parameter of each scene unit within the repair reference window are fitted with a straight line using the least squares method to obtain the fitted straight line. The least squares method for fitting the straight line is a well-known technique, and the specific process will not be elaborated further.

[0056] It should be noted that this application provides only one fitting algorithm for linear fitting of time series data of feature parameters. There are many existing fitting algorithms, and implementers may also use other fitting algorithms to perform linear fitting of time series data of feature parameters. This application does not impose any specific restrictions.

[0057] The absolute value of the slope of the fitted line for each feature parameter of each scene unit and the range are positively fused together to determine the degree of degradation trend for each feature parameter of each scene unit. Here, "positive fusion" refers to combining two or more indicators by addition or multiplication. Preferably, in this embodiment, the expression for the degree of degradation trend is:

[0058]

[0059] In the formula, For the first The degree of degradation trend of the b-th feature parameter of each scene unit; To repair the first reference window The range of the b-th feature parameter of a scene unit; For the first The slope of the fitted line for the b-th feature parameter of a scene unit; This is the normalization function.

[0060] In other embodiments of this application, the degree of degradation trend can also be the sum of the normalized value of the absolute value of the slope of each feature parameter of each scene unit and the normalized value of the range. The normalization method used is the maximum-minimum normalization method, which takes the same index of all feature parameters as input to the maximum-minimum normalization method and outputs the normalized value of each index of each feature parameter. Implementers may also use other methods for normalization, and this application does not impose specific limitations.

[0061] , The larger the value, the more pronounced the overall upward or downward trend of the characteristic parameter before restoration, and the faster the rate of data change. Therefore, this characteristic parameter better reflects the degree of ecological degradation. The larger.

[0062] Step S3: Based on the difference between the data change trends of each feature parameter before repair and the data change trends at each time after repair begins, construct the response degree of each feature parameter at each time after repair begins, so as to analyze the rapid response of each feature parameter and obtain the rapid response stability index of each feature parameter; based on the cumulative characteristics of the differences of each feature parameter after repair begins, and the response degree at the first time after repair begins, determine the asymptotic responsiveness of each feature parameter.

[0063] Some characteristic parameters change significantly within a short period after remediation measures are implemented, making them suitable for reflecting phased remediation progress and short-term management effectiveness. Other characteristic parameters, however, are influenced by ecological succession and structural reconstruction processes, and their changes typically take longer to manifest gradually, making them more suitable for assessing the sustained effects and long-term performance of remediation measures. Using a uniform evaluation method for both types of characteristic parameters can easily underestimate the management significance of gradual changes during project acceptance and monitoring, or misjudge the evaluation weight of short-term fluctuations. Therefore, it is necessary to analyze the response characteristics of different characteristic parameters to match their response rates to remediation measures.

[0064] For each feature parameter of each scene unit obtained after the repair time point, calculate the difference between the value of the feature parameter at each sampling time and the value at the previous sampling time, and record the ratio of the difference to the time interval between the two sampling times as the repair instantaneous recovery slope of the feature parameter at each sampling time.

[0065] If the instantaneous recovery slope of the feature parameter at the sampling time has the same sign as the slope of the fitted line of the feature parameter in the repair reference window, that is, both slopes are either positive or both are negative, then the recovery flag value of the feature parameter at the sampling time is set to 0; otherwise, it is set to 1.

[0066] The purpose of restoring the label value is to ensure that the restored ecological characteristic parameters have a trend that is the opposite of the changes within the previous restoration reference window, that is, a trend of improvement from deterioration.

[0067] The essential difference between fast response feature parameters and asymptotic response feature parameters lies mainly in the different speeds at which the repair measures change. Therefore, by comparing the instantaneous recovery slope of each feature parameter at each sampling time after the repair time point with the slope of the corresponding fitted line of each feature parameter in the repair reference window, the degree of response of the feature parameter at each sampling time can be determined.

[0068] Preferably, in this embodiment, the expression for the response degree of each feature parameter at each sampling time after the repair time point is:

[0069]

[0070] In the formula, For the first The response level of the b-th feature parameter of a scene unit at the c-th sampling time after the repair time point; , The first The recovery identifier value and instantaneous recovery slope of the b-th feature parameter of each scene unit at the c-th sampling time after the repair time point; For the first The slope of the fitted line of the repair reference window for the b-th feature parameter of each scene unit. It is a normalization function for maximum and minimum values; The preset minimum positive number is used to prevent the denominator from being 0. In this embodiment, the minimum positive number is set to 0. The value is set to 0.0001. In other embodiments of this application, the implementer may set the value according to the actual situation. The value of .

[0071] Since rapid response characteristic parameters typically change significantly within a short period after the implementation of ecological restoration measures, their management value lies primarily in the timely reflection of the initial restoration effects. Therefore, the response degree of each characteristic parameter in each scenario unit at the first sampling time after the restoration time point is recorded as the rapid response degree of each characteristic parameter in each scenario unit. The higher the rapid response degree, the more pronounced the rapid response characteristic of that feature parameter.

[0072] When using rapid response characteristic parameters to evaluate the entire ecological restoration project, these parameters often show significant changes in the early stages after ecological restoration. Subsequently, as the effects of restoration measures stabilize, the magnitude of these characteristic parameters should gradually decrease and enter a relatively stable state. Therefore, a rapid response stability index is constructed to quantify them.

[0073] Preferably, in this embodiment, the expression for the fast response stability index is:

[0074]

[0075] In the formula, For the first The fast response stability index of the b-th feature parameter of a scene unit; This refers to the number of sampling times after the repair time point; For the first The minimum response of the b-th feature parameter of a scene unit between the repair time point and the c-th sampling time point; For the first The feature parameter value of the b-th feature parameter of a scene unit at the c-th sampling time after the repair time point; For the first The average value of all feature parameters of the b-th feature parameter of a scene unit after the repair time point; The preset minimum positive number is used to prevent the denominator from being 0. In this embodiment, the minimum positive number is set to 0. The value is set to 0.0001. As a first difference, in other embodiments of this application, the first difference may also be: .

[0076] The first difference reflects the stable state of the characteristic parameter; the smaller the value, the more stable the change of the characteristic parameter. This indicates whether the feature parameters have reached the point where they should be in a stable state from the repair time point to the c-th sampling time thereafter. If the value is large, it indicates that the c-th sampling time is more likely to be a time in the early fast response stage. Reduce focus on data from that specific moment; If the value is small, it indicates that the c-th sampling time may be in a period after the recovery speed has slowed down or the recovery has been interrupted, thus... Increase attention to data from relatively later times after the repair point; and then through Evaluate the stability of this feature parameter after the rapid response is completed. The smaller the value, the more stable the value of the feature parameter is after the fast response is completed, and the better the repair effect.

[0077] In other embodiments of this application, the expression for the fast response stability index may also be: .

[0078] The characteristic parameters of an incremental response typically do not exhibit significant changes at a single time point, but rather show a continuous and slow improvement process over multiple data collection and modeling cycles as ecological restoration measures continue to take effect. Therefore, the cumulative sum of the response levels of the characteristic parameters at multiple sampling times is used as a quantitative measure of the sustained responsiveness of the characteristic parameters. Preferably, in this embodiment, the expression for the sustained responsiveness is:

[0079]

[0080] In the formula, For the first The progressive responsiveness of the b-th feature parameter of a scene unit; For the first The fast response rate of the b-th feature parameter of a scene unit; This refers to the number of sampling times after the repair time point; For the first The minimum recovery identifier value of the b-th feature parameter of a scene unit between the repair time point and the c-th sampling time thereafter; For the first The response level of the b-th feature parameter of a scene unit at the c-th sampling time after the repair time point; It is a normalization function for maximum and minimum values; The preset minimum positive number is used to prevent the denominator from being 0. In this embodiment, the minimum positive number is set to 0. The value is set to 0.0001. In other embodiments of this application, the implementer may set the value according to the actual situation. The value of .

[0081] The larger the value, the more likely the feature parameter still exhibits continuous recovery characteristics at the c-th sampling time after the repair time point. Therefore, the corresponding response levels are accumulated to determine the continuous responsiveness of the feature parameter.

[0082] The main difference between asymptotic response characteristic parameters and fast response characteristic parameters lies in the recovery rate of a single burst, i.e., the fast response degree of the characteristic parameters. Therefore, further constraints can be imposed by using the fast response degree of the characteristic parameters. The larger it is, the more likely it is to be the first The less likely the b-th feature parameter of a scene unit is to be a fast response feature parameter, the lower the probability that it is. The larger the value, the more likely the feature parameter has continuous recoverability and tends to be an asymptotic response feature parameter.

[0083] Step S4: Determine the overall evaluation index of each scenario unit based on the degree of degradation trend of all characteristic parameters, rapid response stability index and progressive responsiveness, in order to conduct acceptance evaluation of the ecological restoration area.

[0084] When a certain characteristic parameter exhibits a high rapid response in the early stages of restoration, and its subsequent changes maintain high stability, it indicates that the restoration measures have produced both timely effects and formed a sustainable restoration state. Such characteristic parameters have high credibility and reference value in management evaluation. On the other hand, when a certain characteristic parameter has a relatively low rapid response, but exhibits a high gradual response over multiple collection cycles, it indicates that the restoration effect is continuously accumulating and gradually improving the ecological state, which also has important long-term management significance.

[0085] Therefore, based on the fast response stability index and the progressive responsiveness, combined with the fast response degree, a repair evaluation index for the characteristic parameters is constructed. Preferably, in this embodiment, the expression for the repair evaluation index is:

[0086]

[0087] In the formula, For the first The repair evaluation index of the b-th feature parameter of a scene unit; For the first The fast response rate of the b-th feature parameter of a scene unit; For the first The fast response stability index of the b-th feature parameter of a scene unit; For the first The progressive responsiveness of the b-th feature parameter of a scene unit.

[0088] Furthermore, the product of the degradation trend degree of each feature parameter of each scene unit and the repair evaluation index is calculated and denoted as the first product; the mean of the first product of all feature parameters of each scene unit is used as the overall evaluation index of each scene unit.

[0089] The overall evaluation index of all scene units is normalized using the maximum normalization method. In other embodiments of this application, the implementer may also use other normalization methods to normalize the overall evaluation index. The normalized value of the overall evaluation index is used to accept each scene unit: an evaluation threshold is set. If the overall evaluation index of each scene unit is greater than the evaluation threshold, the scene unit is judged to have passed acceptance; if the overall evaluation index of each scene unit is less than or equal to the evaluation threshold, the scene unit is judged to have failed acceptance and is included in the scope of continuous monitoring and tracking management, where supplementary repair, adjustment measures, or key supervision can be implemented subsequently. In this embodiment, the evaluation threshold is set to 0.8. In other embodiments of this application, the implementer may set the threshold according to the actual situation.

[0090] A schematic diagram of the process for obtaining the response level is shown below. Figure 2 As shown.

[0091] In summary, this application embodiment conducts multi-dimensional data analysis on scenarios by using data fluctuations of different characteristic parameters of different scenario units. First, it judges the degree of scenario degradation trend reflected by each characteristic parameter by analyzing the rate of change of data before repair. Since the evaluation methods for different characteristic parameters differ during the evaluation process, this application assesses the response intensity of each characteristic parameter at each time point, calculates the rapid response stability index and progressive responsiveness for each characteristic parameter, and judges the rapid response characteristics and progressive response characteristics of the characteristic parameters separately. By judging the rapid response of each characteristic parameter through response intensity, different weights are applied to calculate the rapid response stability index and progressive responsiveness of the characteristic parameters. Combined with the degree of degradation trend, an overall evaluation index for the scenario unit is obtained. Different evaluation methods can be used for characteristic parameters with different response characteristics, enabling comprehensive judgment and evaluation, achieving accurate assessment of ecological restoration projects, and improving the accuracy of ecological restoration project assessment by applying different processing methods.

[0092] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this application. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0093] The various embodiments in this application are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0094] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them; modifications to the technical solutions described in the foregoing embodiments, or equivalent substitutions of some of the technical feature parameters, do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for acceptance and evaluation of ecological restoration projects based on UAV 3D model interpretation, characterized in that, The method includes the following steps: The system acquires 3D drone models of the ecological restoration area at each sampling time and identifies various feature parameter values ​​of each scene unit in the 3D drone model. The sampling time includes the sampling time before ecological restoration and the sampling time after restoration begins. Based on the data change characteristics of each feature parameter within a preset time period before repair, the degree of degradation trend of each feature parameter is evaluated. Based on the differences between the data change trends of each feature parameter before repair and the data change trends at each time point after repair begins, the response degree of each feature parameter at each time point after repair begins is constructed to analyze the rapid response of each feature parameter and obtain the rapid response stability index of each feature parameter; based on the cumulative characteristics of the differences of each feature parameter after repair begins and the response degree at the first time point after repair begins, the asymptotic responsiveness of each feature parameter is determined. The overall evaluation index of each scenario unit is determined based on the degree of degradation trend of all characteristic parameters, rapid response stability index and progressive responsiveness, in order to conduct acceptance evaluation of the ecological restoration area. The process for obtaining the degree of degradation trend is as follows: Calculate the range of values ​​of each feature parameter at all times within a preset time period before repair, obtain the fitted line of values ​​at all times through a linear fitting algorithm, and use the positive fusion result of the absolute value of the slope of the fitted line of each feature parameter and the range as the degree of degradation trend of each feature parameter of each scene unit. The process of obtaining the response level of each feature parameter at each time point after the repair begins is as follows: Calculate the numerical difference between each feature parameter at each time after the start of repair and the previous time, and record the ratio of the difference to the time interval between two adjacent time points as the instantaneous recovery slope of each feature parameter at each time after the start of repair. The instantaneous recovery slope of the repair is compared with the slope of the fitted line to determine the recovery indicator value of each feature parameter at each time point after the repair begins; The response level of each feature parameter at each time point after the repair begins is determined based on the instantaneous recovery slope and the recovery identifier value.

2. The method for acceptance and evaluation of ecological restoration projects based on UAV 3D model interpretation as described in claim 1, characterized in that, The process for obtaining the recovery identifier value is as follows: If the instantaneous recovery slope has the same sign as the slope of the fitted line, then the recovery flag value is set to 0; otherwise, it is set to 1.

3. The method for acceptance and evaluation of ecological restoration projects based on UAV 3D model interpretation as described in claim 1, characterized in that, The process of obtaining the response level is as follows: The difference between the instantaneous recovery slope and the slope of the fitted line is mapped to a positive number; the product of the positive number and the recovery identifier value is used as the response degree of each feature parameter at each time point after the start of repair.

4. The method for acceptance and evaluation of ecological restoration projects based on UAV 3D model interpretation as described in claim 1, characterized in that, The analysis of the rapid response of each characteristic parameter yields the rapid response stability index of each characteristic parameter, specifically: Calculate the difference between the data value of each feature parameter at each time point after the repair begins and the mean of the data values ​​at all times, and record it as the first difference; obtain the minimum response degree of each feature parameter in the historical local time period after the repair begins. The fast response stability index of each characteristic parameter is negatively correlated with the first difference and the minimum response degree, respectively.

5. The method for acceptance and evaluation of ecological restoration projects based on UAV 3D model interpretation as described in claim 1, characterized in that, The process for obtaining the asymptotic responsiveness of each feature parameter value is as follows: The minimum recovery identifier value of each feature parameter in the historical local time period after the repair begins is used as the weight of the response degree of each feature parameter at each time point; The rapid response degree of each feature parameter is determined based on the aforementioned response degree; Obtain the weighted sum of the response levels at all times after the repair begins, and combine it with the fast response level to determine the asymptotic responsiveness of each feature parameter; the asymptotic responsiveness is directly proportional to the sum and inversely proportional to the fast response level.

6. The method for acceptance and evaluation of ecological restoration projects based on UAV 3D model interpretation as described in claim 5, characterized in that, The fast response rate is defined as the response level of each feature parameter at the first moment after the repair process begins.

7. The method for acceptance and evaluation of ecological restoration projects based on UAV 3D model interpretation as described in claim 5, characterized in that, The process of obtaining the overall evaluation index is as follows: The forward and reverse mapping values ​​of the fast response degree are used as the weights of the fast response stability index and the progressive responsiveness, respectively. The weighted sum of the fast response stability index and the progressive responsiveness is used as the repair evaluation index of each feature parameter. The fusion value of the product of the degradation trend of all characteristic parameters of each scene unit and the repair evaluation index is used as the overall evaluation index of each scene unit.

8. The method for acceptance and evaluation of ecological restoration projects based on UAV 3D model interpretation as described in claim 1, characterized in that, The acceptance assessment of the ecological restoration area specifically includes: If the normalized value of the overall evaluation index of each scenario unit is greater than the preset evaluation threshold, then each scenario unit is judged to have passed the acceptance test; otherwise, each scenario unit is judged to have failed the acceptance test.

Images