Grassland ecological degradation recovery effect dynamic evaluation method, device, equipment and medium

Abstract

The invention relates to a grassland ecological degradation recovery effect dynamic evaluation method and device, equipment and a medium, and the method comprises the steps: obtaining interannual remote sensing data for a target region, carrying out the vegetation feature extraction of the interannual remote sensing data based on the sliding operation of a predetermined window on the interannual remote sensing data, and obtaining a reference time sequence; calculating the relative change rate of vegetation in the target area based on the reference time sequence; and according to the relative change rate and a preset threshold value, evaluating the degradation or recovery condition of the vegetation in the target area. According to the method, the optimal spatial analysis scale, namely the window, is determined according to spatial heterogeneity characteristics (semi-variable function range), adaptive conversion from the pixel to the landscape scale can be realized, subjective interference can be eliminated, a reference time sequence reflecting vegetation characteristics can be accurately determined on the basis of the window, and the method has the advantages of being simple in structure and convenient to operate. And the degradation or recovery condition of the target area can be accurately evaluated based on the reference time sequence.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ecological environment remote sensing monitoring and dynamic evaluation, in particular, the present application relates to a grassland ecological degradation recovery effect dynamic evaluation method, device, equipment and medium. BACKGROUND

[0002] Constructing a scientific and reasonable evaluation benchmark is a key prerequisite for accurately evaluating the degradation recovery effect of an ecological system. The commonly used benchmark construction methods at present mainly include fixed benchmark method and spatial homogenization method, but both have significant limitations. The fixed benchmark method (such as the historical state method) assumes that the benchmark state is constant and does not change, which ignores the dynamic influence of climate change and vegetation natural succession. Studies have shown that the NDVI benchmark value of the typical grassland in Inner Mongolia has significantly increased at a rate of 0.012 / 10a in the past 20 years (P<0.01). If the fixed benchmark method is used, the degraded area will be overestimated by 18.3%. Therefore, this method will introduce evaluation errors because it cannot represent the natural succession process of the vegetation community. The spatial homogenization method usually uses a single-pixel historical benchmark. However, if the pixel itself is in a long-term degraded state, it cannot effectively evaluate the recovery effect. If a sliding window (such as 3x3km) is used, it has strong subjectivity because it lacks quantitative basis for spatial heterogeneity. At the same time, if the global mean value is directly used, the result is easily disturbed by outliers. Simulation data shows that when the proportion of outliers exceeds 5%, the mean deviation will exceed 15%. In addition, in the benchmark result verification link, the existing methods generally rely on point-shaped sample data to verify the planar evaluation results, which obviously lacks spatial representativeness. SUMMARY

[0003] The technical problem to be solved by the present application is to provide a grassland ecological degradation recovery effect dynamic evaluation method, device, equipment and medium, which aims to solve at least one of the above technical problems.

[0004] In a first aspect, the technical solution of the present application to solve the above technical problems is as follows: a grassland ecological degradation recovery effect dynamic evaluation method, the method comprising: obtaining annual remote sensing data for a target area, the target area being an area including vegetation, and the annual remote sensing data being selected from at least two years of remote sensing data corresponding to the target area; performing vegetation feature extraction on the annual remote sensing data based on a predetermined window sliding operation on the annual remote sensing data, wherein the window is quantitatively determined based on the range of the semi-variogram function, and the benchmark time series is data describing the change of the vegetation feature in a year, and each sliding operation corresponds to a benchmark value in the benchmark time series; calculating the relative change rate of the vegetation in the target area based on the benchmark time series; According to the relative change rate and the preset threshold, the degradation or recovery of the vegetation in the target region is evaluated.

[0005] The present application has the advantages that: the optimal spatial analysis scale (window) is determined according to the spatial heterogeneity characteristics (semi-variogram range), the adaptive conversion from a pixel to a landscape scale is realized, the subjective interference is eliminated, the sliding operation of the predetermined window on the inter-annual remote sensing data is adopted, the reference time series obtained can accurately reflect the natural succession process of the vegetation community, and thus the degradation or recovery of the target region can be accurately evaluated based on the reference time series.

[0006] Based on the above technical solution, the present application can be further improved as follows.

[0007] Further, the window is determined based on the following manner: According to the inter-annual remote sensing data and the semi-variogram, the estimated value of the semi-variance graph corresponding to the first sample point and the second sample point in the inter-annual remote sensing data is determined. Based on the estimated value, the window is determined.

[0008] Further, the sliding operation of the predetermined window on the inter-annual remote sensing data is used to extract the vegetation features of the inter-annual remote sensing data to obtain the reference time series, which includes: Based on the sliding operation of the window on the inter-annual remote sensing data, all the pixels in the window corresponding to each sliding operation are determined. For each sliding operation, the 85%-99% quantile is calculated based on all the pixels in the window corresponding to the sliding operation. For each sliding operation, the truncated data set is constructed based on the 85%-99% quantile and all the pixels corresponding to the sliding operation; and for each sliding operation, the reference value corresponding to the sliding operation is determined according to the truncated data set corresponding to the sliding operation. Based on the reference values corresponding to all the sliding operations, the reference time series corresponding to the inter-annual remote sensing data is determined.

[0009] Further, for each sliding operation, the reference value corresponding to the sliding operation is determined according to the truncated data set corresponding to the sliding operation, which includes: For each sliding operation, the candidate reference value corresponding to the sliding operation is determined according to the truncated data set corresponding to the sliding operation, the candidate reference value including the reference value determined based on the mean value of the truncated data set and the reference value determined based on the median value of the truncated data set; for each sliding operation, the normal distribution verification of the truncated data set is performed based on the Shapiro-Wilk normality test, and the reference value corresponding to the sliding operation is determined from the candidate reference value according to the verification result.

[0010] Further, the method further includes: acquire new inter-annual remote sensing data, and update the reference time series according to the new inter-annual remote sensing data.

[0011] Further, the method further comprises: acquiring verification data for the target region, the verification data being ground measured data or unmanned aerial vehicle photogrammetry data; performing consistency verification on the verification data and the inter-annual remote sensing data, and calibrating the evaluation result of the target region according to the verification result, the evaluation result being a result of evaluating the degradation or recovery of the vegetation in the target region.

[0012] Further, the method further comprises: For each sliding operation, it is determined whether the reference value corresponding to the sliding operation is a top community threshold value, based on the reference value corresponding to the sliding operation, a predetermined minimum contiguous area threshold value, a mean value and a standard deviation of all reference values corresponding to the target region.

[0013] In a second aspect, the present application provides a device for dynamically evaluating the effectiveness of grassland ecological degradation and recovery, which comprises: an acquisition module, configured to acquire inter-annual remote sensing data for a target region, the target region being a region comprising vegetation, and the inter-annual remote sensing data being remote sensing data selected from at least two years of remote sensing data corresponding to the target region; a reference time series determination module, configured to perform vegetation feature extraction on the inter-annual remote sensing data based on a sliding operation of a predetermined window on the inter-annual remote sensing data, to obtain a reference time series, wherein the window is quantitatively determined based on a semi-variogram range, and the reference time series is data describing the change of the vegetation feature within a year, and each sliding operation corresponds to a reference value in the reference time series; a relative change rate determination module, configured to calculate a relative change rate of the vegetation in the target region based on the reference time series; an evaluation module, configured to evaluate the degradation or recovery of the vegetation in the target region according to the relative change rate and a preset threshold value.

[0014] In a third aspect, the present application provides an electronic device, which comprises a memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor implements the method for dynamically evaluating the effectiveness of grassland ecological degradation and recovery when executing the computer program.

[0015] In a fourth aspect, the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executable on a processor to implement the method for dynamically evaluating the effectiveness of grassland ecological degradation and recovery.

[0016] Additional aspects and advantages of the application will be set forth in part in the description which follows, and in part will become apparent to those skilled in the art upon examination of the following and the attendant drawings or can be learned by practice of the application. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments of the present application will be briefly introduced.

[0018] Figure 1 A flowchart of a method for dynamically evaluating the recovery effect of grassland ecological degradation according to an embodiment of the present application is shown in FIG. 1. Figure 2 A flowchart of another method for dynamically evaluating the recovery effect of grassland ecological degradation according to an embodiment of the present application is shown in FIG. 2. Figure 3 The (a) to (d) in FIG. 3 are distribution diagrams of NDVI semi-variation function parameter results according to an embodiment of the present application. Figure 4 The (a) to (d) in FIG. 4 are distribution diagrams of NDVI different quantile and interval pixel values according to an embodiment of the present application. Figure 5 The (a) to (e) in FIG. 5 are distribution diagrams of NDVI different quantile and interval reference values according to an embodiment of the present application. Figure 6 The (a) to (d) in FIG. 6 are distribution diagrams of NDVI different quantile difference values according to an embodiment of the present application. Figure 7 The (a) to (c) in FIG. 7 are spatial distribution diagrams of NDVI original values and difference values according to an embodiment of the present application. Figure 8 The (a) to (c) in FIG. 8 are distribution feature diagrams of NDVI data according to an embodiment of the present application. Figure 9 A structural diagram of a device for dynamically evaluating the recovery effect of grassland ecological degradation according to an embodiment of the present application is shown in FIG. 9. Figure 10 A structural diagram of an electronic device according to an embodiment of the present application is shown in FIG. 10. DETAILED DESCRIPTION

[0019] The principles and features of the present application are described below, and the examples are only used to explain the present application and are not used to limit the scope of the present application.

[0020] The following describes in detail the technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems using specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The following embodiments of the present invention are described in conjunction with the accompanying drawings.

[0021] The solutions provided by the embodiments of the present invention can be applied to any application scenario requiring assessment of the degradation or restoration of vegetation-containing areas. The solutions provided by the embodiments of the present invention can be executed by any electronic device, for example, a user's terminal device, including at least one of the following: a smartphone, tablet computer, laptop computer, desktop computer, smart speaker, smart watch, smart TV, or smart in-vehicle device.

[0022] The embodiment of the present invention provides a possible implementation method, such as Figure 1 As shown, a flowchart of a method for dynamically evaluating the effectiveness of grassland ecological degradation restoration is provided. This solution can be executed by any electronic device, for example, a terminal device, or by a terminal device and a server. For ease of description, the method provided by the embodiment of the present invention will be described below using a terminal device as an example of the execution subject. Figure 1 As shown in the flowchart, the method may include the following steps: S10, obtaining inter-annual remote sensing data for a target area, where the target area is an area including vegetation, and the inter-annual remote sensing data is remote sensing data selected from remote sensing data of at least two years corresponding to the target area; S20, extracting vegetation features from the inter-annual remote sensing data based on a sliding operation of a predetermined window on the inter-annual remote sensing data to obtain a reference time series, wherein the window is quantitatively determined based on the semivariogram range, the reference time series is data describing changes in vegetation features within a year, and each sliding operation corresponds to a reference value in the reference time series; S30, calculating the relative change rate of vegetation in the target area based on the benchmark time series; S40: Evaluate the degradation or recovery of vegetation in the target area based on the relative change rate and a preset threshold.

[0023] Through the method of the present invention, the optimal spatial analysis scale, i.e., the window, is determined according to the spatial heterogeneity characteristics (semivariogram range), which can realize the adaptive conversion from pixel to landscape scale, eliminate subjective interference, and use the sliding operation of the predetermined window on the inter-annual remote sensing data. The obtained benchmark time series can accurately reflect the natural succession process of the vegetation community, so that the degradation or recovery of the target area can be accurately assessed based on the benchmark time series.

[0024] The core innovation of the present application is to propose and construct a spatio-temporal adaptive dynamic benchmark system to overcome the above limitations. The core technology includes: generation of benchmark time series corresponding to interannual remote sensing data, determination of adaptive window based on semi-variation function range, determination of 85%-99% quantile, extraction of benchmark value, and construction of multi-index coupled degradation recovery quantitative model. By iteratively generating benchmark time series, the vegetation succession dynamics can be effectively captured, and the evaluation error caused by ignoring temporal changes can be significantly reduced. According to the spatial heterogeneity characteristics (semi-variation function range), the optimal spatial analysis scale can be determined, which can realize adaptive conversion from pixel to landscape scale and eliminate subjective interference. Using the mean or median of 85%-99% quantile truncation to calculate the benchmark value can effectively suppress extreme value interference, thereby improving the robustness of the benchmark value. The establishment of a multi-source verification framework coupling ground sample, unmanned aerial survey and multi-temporal remote sensing inversion data can significantly improve the spatial representativeness of the verification results.

[0025] The present application aims to solve the three core contradictions in the existing grassland ecological degradation recovery evaluation benchmark construction technology; one is the contradiction between spatio-temporal staticity and ecological dynamics, i.e. fixed benchmark value cannot represent the natural succession process of vegetation community; two is the contradiction between spatial homogeneity assumption and heterogeneity reality, i.e. fixed window analysis ignores the spatial differentiation characteristics of landscape pattern; three is the deviation accumulation contradiction between evaluation results and ground true situation, i.e. lack of effective dynamic calibration mechanism leads to error propagation.

[0026] Based on this, the present application provides a grassland ecological degradation recovery effectiveness dynamic evaluation method, which can include the following steps, see Figure 2 to Figure 8 : S10, obtaining interannual remote sensing data for a target region, the target region being a region including vegetation, the interannual remote sensing data being remote sensing data selected from remote sensing data corresponding to at least two years of the target region; The interannual remote sensing data can be remote sensing data selected from remote sensing data corresponding to at least two years of the target region, for example, remote sensing data selected from remote sensing data corresponding to 2021 to 2022 of the target region. The interannual remote sensing data can be image data.

[0027] The target region can be a grassland, and the above interannual remote sensing data can be NDVI / EVI remote sensing data with a spatial resolution p∈{30m, 250m}. The above interannual remote sensing data can be processed to obtain an image, which is used for subsequent processing and can reflect the situation of the vegetation in the target region at different times in a year.

[0028] S20, performing vegetation feature extraction on the inter-annual remote sensing data based on a predetermined window sliding operation on the inter-annual remote sensing data, to obtain a baseline time series, wherein the window is quantitatively determined based on a range of a semi-variation function, and the baseline time series is data describing changes in vegetation features within a year, each sliding operation corresponding to a baseline value in the baseline time series; wherein the window is slid once on the inter-annual remote sensing data, corresponding to obtaining a baseline value, and the baseline value can reflect the vegetation features in the region corresponding to the window on the inter-annual remote sensing data. After sliding multiple times, the obtained multiple baseline values can form a baseline time series, better reflecting the time sequence of the vegetation features.

[0029] S30, calculating a relative change rate of the vegetation in the target region based on the baseline time series; The relative change rate can reflect the changes in the vegetation in the target region, i.e., quantitatively representing the changes in the vegetation in the target region by the relative change rate.

[0030] Optionally, one implementation of the above S30 is: S301, calculating an absolute change amount according to the baseline time series; Specifically, the calculation formula is as follows: ΔV(x, y, t) = V(x, y, t) - B t (x, y) (1) Wherein, B t (x, y) represents the baseline value at (x, y) in the t year, V(x, y, t) represents the actual observation value at (x, y) in the t year, and ΔV(x, y, t) represents the deviation between the actual observation value and the baseline value.

[0031] S302, calculating a relative change amount based on the absolute change amount.

[0032] Specifically, the calculation formula is as follows: Wherein, R(x, y, t) represents the relative change amount.

[0033] Before S301, the method further includes: S300, performing time consistency correction on the baseline time series to obtain a corrected baseline time series; One implementation of S300 is: Based on the inter-annual remote sensing data, a first average NDVI corresponding to the year corresponding to the inter-annual remote sensing data is calculated; the first average NDVI can be represented as μ GLOBAL (t), wherein t represents the year corresponding to the inter-annual remote sensing data; The reference time series is time-consistently corrected according to the first average NDVI and the predetermined second average NDVI, to obtain a corrected reference time series. The second average NDVI can be represented as μ GLOBAL (t0), where t0 is a reference starting year (reference year), and is usually selected as a year with high data quality or stable ecological state.

[0034] The above-mentioned time-consistent correction of the reference time series according to the first average NDVI and the predetermined second average NDVI can be specifically performed by correcting each reference value in the reference time series according to the first average NDVI and the predetermined second average NDVI, which can be represented as: where B′ t (x, y) represents the corrected reference value, B t (x, y) represents the reference value before correction, and (x, y) represents a position point corresponding to the reference value.

[0035] The above-mentioned formula for obtaining the absolute change amount and the relative change rate can be replaced by the corrected reference time series.

[0036] S40, according to the relative change rate and the preset threshold value, evaluating the degradation or recovery of the vegetation in the target region.

[0037] The preset threshold value includes a first threshold value and a second threshold value, the first threshold value is a threshold value corresponding to the degradation, and the second threshold value is a threshold value corresponding to the recovery. One implementation of the above-mentioned S40 can be: when the relative change rate is less than the first threshold value for n consecutive years, the target region is evaluated as the degradation; when the relative change rate is greater than the second threshold value for m consecutive years, the target region is evaluated as the recovery.

[0038] As an example, the first threshold value can be -15%, and the second threshold value can be 20%. n is 3, and m is 2.

[0039] The evaluation result of the target region can be represented as: Table 1 Status Decision criteria Ecological interpretation Degradation R <- 15% and for > 3 years Vegetation status is persistently significantly below baseline levels and long-term monitoring is required to confirm. Recovery R > +20% and persists > 2 years Vegetation status is significantly improved and stable, and short-term confirmation of recovery trend is possible.

[0040] Optionally, the above-mentioned window is determined based on the following method: S1, according to the inter-annual remote sensing data and the semi-variogram function, determining an estimated value of a semi-variance graph corresponding to the first sample point and the second sample point in the inter-annual remote sensing data; S2, determining the window based on the estimated value.

[0041] In S1, the first sample point and the second sample point are target regions with a distance of h k The two sampling points of the distance; specifically, the above semi-variogram function can be expressed as: Wherein, γ(h k ) is the estimated value of the semi-variogram with a distance of h k , h k is the distance between the first sample point and the second sample point, N(h k ) is the number of pixel pairs satisfying (x i -x j )∈(h k -V h , h k +V h ), a pixel pair includes two sample points, V h =0.5ρ, ρ represents the spatial resolution, z(x i ) is the average density of the first sample point x i , and z(x i +h k ) is the average density of the second sample point (x i +h k ).

[0042] In S2, based on the estimated value, one implementation of determining the window is: S21, based on the estimated value, a variogram function theoretical model is established, and the value of the range is calculated based on the variogram function theoretical model; as an example, taking a spherical model as an example, the variogram function theoretical model can be expressed as: Wherein, a is the range, c0 is the nugget value, and c is the structural variance.

[0043] S22, based on the value of the range and a predetermined parameter, the window is determined.

[0044] Specifically, the window can be determined based on the following formula: W=min(2a,W max ) (6) Wherein, W max is a predetermined parameter, which can be determined based on ecological zoning, for example, the typical grassland region W max =5km.

[0045] Optionally, in S20, based on the sliding operation of the predetermined window on the interannual remote sensing data, the vegetation feature extraction is performed on the interannual remote sensing data to obtain the baseline time series, including: S201, determining all pixels in the window corresponding to each sliding operation based on the sliding operation of the window in the inter-annual remote sensing data; S202, for each sliding operation, calculating the 85%-99% quantile based on all pixels in the window corresponding to the sliding operation; in the present scheme, the quantile corresponding to 85% can be represented as Q 85 , and the quantile corresponding to 99% can be represented as Q 99 .

[0046] S203, for each sliding operation, constructing a truncated data set based on the 85%-99% quantile corresponding to the sliding operation and all pixels; Specifically, the truncated data set Z ref can be represented as: Z ref = {z i | Q 85 ≤ z i ≤ Q 99} (7) Wherein, z i represents the value of all pixels in the window corresponding to each sliding operation.

[0047] S204, for each sliding operation, determining the reference value corresponding to the sliding operation according to the truncated data set corresponding to the sliding operation; S205, determining the reference time sequence corresponding to the inter-annual remote sensing data based on the reference values corresponding to all sliding operations.

[0048] Optionally, in the above S204, for each sliding operation, determining the reference value corresponding to the sliding operation according to the truncated data set corresponding to the sliding operation, comprising: S2041, for each sliding operation, determining the candidate reference value corresponding to the sliding operation according to the truncated data set corresponding to the sliding operation, the candidate reference value including the reference value determined based on the mean value of the truncated data set and the reference value determined based on the median value of the truncated data set; The above processing process can be specifically represented as: Wherein, and median(Z ref ) are candidate reference values.

[0049] S2042, for each sliding operation, performing normal distribution verification on the truncated data set based on Shapiro-Wilk normality test, and determining the reference value corresponding to the sliding operation from the candidate reference value according to the verification result.

[0050] Wherein, the process of normal distribution verification can be represented as: wherein a i denotes the coefficient of Shapiro-Wilk normality test, z i Generally, the sample value is obtained after sorting the data in the truncated data set (for example, in ascending order from small to large or from large to small). denotes the mean of the sample value obtained after sorting, and n denotes the total number of data in the truncated data set.

[0051] If W < W a (for example, a = 0.05), the null hypothesis is rejected, it is considered that the truncated data set does not conform to the normal distribution, and the benchmark value corresponding to the median mode is selected; if W ≥ W a , the null hypothesis is accepted, it is considered that the truncated data set conforms to the normal distribution, and the benchmark value corresponding to the mean mode is selected.

[0052] Optionally, the method further comprises: Obtaining new interannual remote sensing data, and updating the benchmark time series according to the new interannual remote sensing data.

[0053] The benchmark time series is dynamically iterated, so that the vegetation succession dynamics can be captured in real time, and the error caused by ignoring the time change can be eliminated.

[0054] Optionally, the method further comprises: For each sliding operation, whether the benchmark value corresponding to the sliding operation is a top community threshold value is determined based on the benchmark value corresponding to the sliding operation, a predetermined minimum contiguous area threshold value, a mean and a standard deviation of all benchmark values corresponding to the target region.

[0055] After each sliding operation, whether the benchmark value corresponding to the sliding operation is a top community threshold value is determined, if yes, the benchmark time series is determined based on the top community threshold value subsequently, if not, it is determined as a non-top community value and does not participate in the determination of the benchmark time series subsequently.

[0056] The specific implementation process of the above determination of whether the benchmark value corresponding to each sliding operation is a top community threshold value based on the benchmark value corresponding to the sliding operation, a predetermined minimum contiguous area threshold value, a mean and a standard deviation of all benchmark values corresponding to the target region is as follows: For each sliding operation, whether the benchmark value corresponding to the sliding operation satisfies a first condition and a second condition is determined, when the benchmark value corresponding to the sliding operation satisfies the first condition and the second condition, the benchmark value corresponding to the sliding operation is determined as a top community threshold value, otherwise, the benchmark value corresponding to the sliding operation is determined as a non-top community value.

[0057] wherein the first condition and the second condition are respectively: wherein, μ B represents the mean of all reference values in the target region, σ B represents the standard deviation of all reference values in the target region, S min represents the minimum contiguous area threshold, which can be determined by the vegetation type (for example, set to S min = 1 km 2 ).

[0058] Optionally, the method further comprises: obtaining verification data for the target region, the verification data being ground measured data or unmanned aerial vehicle surveying data; performing consistency verification on the verification data and the inter-annual remote sensing data, and calibrating the evaluation result of the target region according to the verification result, the evaluation result being the result of evaluating the degradation or recovery of the vegetation in the target region.

[0059] As an example, in order to verify the consistency of the remote sensing inversion value (i.e. the parameter value obtained based on the inter-annual remote sensing data) and the ground measured value (the parameter value expressing the same meaning as the inversion parameter value), the following pixel-quadrat matching model can be established: The verification condition is: |1-R remote |<0.15; wherein, V i field represents the measured value of the i-th ground quadrat in the ground measured data, V i remote represents the remote sensing pixel extraction value (remote sensing inversion value) corresponding to the i-th ground quadrat (the pixel in the inter-annual remote sensing data corresponding to the same position as the i-th ground quadrat), k1 is a normalization coefficient, usually k1 = 1 / k, k is the number of matched quadrats (covering different habitat types), R remote is a matching degree index of the inter-annual remote sensing data and the ground measured data. The number of matched quadrats refers to the logarithm of the ground quadrats and remote sensing pixels corresponding to the same value.

[0060] If R remote satisfies the verification condition, it can reflect the accuracy of the application scheme in evaluating the grassland ecological degradation and recovery effect based on the inter-annual remote sensing data.

[0061] Optionally, in the application scheme, in order to quantify the contribution of different driving factors such as climate factors and human factors to ecological degradation or recovery, a partial least squares regression (PLSR) method is used for driving force analysis, and the calculation formula is as follows: wherein R represents the target variable, i.e. the relative change rate R(x, y, t), X j represents driving factors, including temperature, precipitation, stocking rate and other factors, β0represents the intercept term, β j represents the regression coefficient of the jth driving factor, reflecting the direction and strength of its influence on the target variable, represents the residual term, i.e. the unexplained random error.

[0062] In order to better illustrate and understand the principles of the method provided by the present application, the scheme of the present application will be described below in combination with an optional specific embodiment. It should be noted that the specific implementation manner of each step in the specific embodiment should not be understood as a limitation on the scheme of the present application, and other implementation manners that can be thought of by those skilled in the art on the basis of the principles of the scheme provided by the present application should also be regarded as within the protection scope of the present application.

[0063] Reference is made to Figure 2 , and the grassland ecological degradation recovery effectiveness dynamic evaluation method provided by the present application is further described in combination with the following embodiment 1 and embodiment 2: Embodiment 1: In this embodiment, the alpine meadow in the Three-River Source Region is taken as the target region, and the grassland ecological degradation recovery effectiveness dynamic evaluation method is further described, which comprises the following steps: 1. Data input and preprocessing The interannual remote sensing data adopts MODIS NDVI data (spatial resolution 250m), from which 8-day remote sensing data are selected from the corresponding remote sensing data from 2001 to 2020 to synthesize, and after radiation correction, atmospheric correction and cloud removal processing, a maximum NDVI composite image containing the corresponding maximum NDVI of the growing season (May-September corresponding to the remote sensing data) is generated.

[0064] The ground data (ground measured data) comes from 126 ground sample plots, all of which uniformly cover the alpine meadow habitat type, and the vegetation coverage is investigated synchronously in August every year, with a data accuracy of ±5%.

[0065] 2. Spatial heterogeneity analysis and window optimization First, according to the maximum NDVI composite image and the semi-variogram function, the estimated value is calculated, the distance h k is in the range of 0-20km, and the step size p=500m. Then a spherical model is fitted to obtain the parameter results: nugget value c0=0.0, sill value c0+c=0.18, and range a=3.2km. Finally, the window size is determined, W max =8km, and the final window W=6.4km is calculated.

[0066] Robust benchmark calculation Firstly, based on the sliding operation of window in sample remote sensing data, all pixels in the window corresponding to each sliding operation are determined; for each sliding operation, based on all pixels in the window corresponding to the sliding operation, probability distribution modeling is performed, and quantile Q 85 = 0.78 and Q 99 = 0.92, outliers of NDVI < 0.2 or > 0.92 are removed. Then the Shapiro-Wilk normality test is performed, and the test result is W = 0.92, P = 0.008 < 0.05, the normality hypothesis is rejected, and the median mode is used to calculate the reference value B(x, y) corresponding to the sliding operation.

[0067] 4. Top community reference mapping Firstly, the threshold involved is set, the global reference mean μ B = 0.75, the standard deviation σ B = 0.11, and the top community threshold B(x, y) ≥ 0.86. Then the contiguity analysis is performed, and the 8-neighbor clustering algorithm is used to remove patches with an area < 1 km 2 , corresponding to 40 pixels.

[0068] 5. Dynamic reference generation According to the first average NDVI and the predetermined second average NDVI, the reference time series is corrected for temporal consistency, the reference year t0 = 2005 is selected, and after correction, the interannual fluctuation rate is reduced from the original 4.9% to 2.7%-4.1%.

[0069] 6. Degradation recovery quantification By calculating the absolute change and the relative change rate, the threshold determination conditions of degradation and recovery are obtained, the degradation determination condition is: R < -20% and lasts for ≥3 years, and the recovery determination condition is: R > +20% and lasts for ≥2 years. The output result is: the degradation area accounts for 12.3%, and the recovery area accounts for 6.8%.

[0070] 7. Construction of multi-source verification system Firstly, ground verification is performed, 30 independent ground sample plots are selected, the matching degree R remote = 0.98, and |1-R remote | = 0.02 < 0.15. Then the driving force analysis is performed, and the PLSR model result shows that the reduction of precipitation (VIP = 1.52, β = -0.67) and overgrazing (VIP = 1.33, β = -0.55) are the main causes of degradation, and the contribution rate is 81.5%.

[0071] 8. Calculation efficiency and verification accuracy The processing time of single scene image for spatial heterogeneity analysis was 4.2 minutes (Intel Xeon Gold 6248R, 128G memory). The validation accuracy R 2 = 0.91, and the Kappa coefficient was 0.84, indicating consistency with ground surveys.

[0072] In Example 2, the target area is the Hulun Buir meadow steppe, and the method for dynamically evaluating the recovery effect of grassland ecological degradation is further illustrated. The method includes: 1. Data input and preprocessing The remote sensing data used Landsat-8 OLI data (spatial resolution p = 30 m, 2013-2020, cloud cover <10%, after radiation calibration, terrain correction and NDVI calculation, 16-day composite images were generated).

[0073] The UAV observation data is 50 1km 2 orthophoto (spatial resolution 5cm), according to the image extraction vegetation coverage, data accuracy ±2%.

[0074] Spatial heterogeneity analysis and window optimization First, according to the composite image and the semi-variogram function, the estimated value is determined, and then the exponential model is fitted to obtain the parameter results: nugget value c0 = 0.01, sill value c0 + c = 0.15, and range a = 0.8km. Finally, the window size is determined, set W max = 1.2km, and the final window W = 1.2km is calculated.

[0075] 3. Robust benchmark calculation First, based on the sliding operation of the window in the sample remote sensing data, all pixels within the window corresponding to each sliding operation are determined; for each sliding operation, based on all pixels within the window corresponding to the sliding operation, the probability distribution modeling is performed, and the quantile Q 85 = 0.82 and Q 99 = 0.95 are calculated for the NDVI data within the window. The outliers with NDVI <0.2 or >0.95 are removed. Then the Shapiro-Wilk normality test is performed, and the test result is W = 0.96, P = 0.21 >0.05, the normality hypothesis is accepted, and the mean mode is used to calculate the benchmark value B(x, y).

[0076] Top community benchmark mapping First, the threshold involved is set, the global benchmark mean mu B = 0.68, the standard deviation sigma B = 0.09, and the top community threshold B(x, y) >= 0.77. Then perform a contiguous analysis, using a morphological erosion-dilation algorithm to remove areas <0.09km2 corresponding to 10 pixels.

[0077] 5. Dynamic benchmark generation According to the first average NDVI and the predetermined second average NDVI, the benchmark time sequence is time consistency corrected, the benchmark year t0=2015 is selected, and after the correction, the interannual fluctuation rate is stabilized at 1.8%-3.2%.

[0078] 6. Degradation recovery quantification By calculating the absolute change and the relative change rate, the threshold judgment conditions of degradation and recovery are obtained, the degradation judgment condition is: R<-25% and lasts for more than 3 years, and the recovery judgment condition is: R>=+25% and lasts for more than 2 years. The output result is: the minimum recovery patch is 0.09km 2 , and the NDVI of the recovery area is increased by 27.4%.

[0079] 7. Construction of multi-source verification system First, the unmanned aerial vehicle data is verified, and the matching degree R remote =1.03 is calculated, which satisfies |1-R remote |=0.03<0.15. Then, the driving force analysis is carried out, and the PLSR model result shows that the increase of precipitation (VIP=1.78, β=+0.62) and the decrease of livestock rate (VIP=1.45, β=+0.29) dominate the recovery, the contribution rate of precipitation is 62.3%, the contribution rate of livestock pressure relief is 28.7%, and the sum of the contribution rates is 91.0%.

[0080] 8. Calculation efficiency and spatial accuracy For spatial heterogeneity analysis, the single scene image processing time is 9.8 minutes (Intel Xeon Gold 6248R, 128G memory), and 0.09km 2 level degradation patches can be detected, which has the advantage of spatial resolution.

[0081] Through the scheme of the application, compared with the prior art, the following beneficial effects are obtained: 1. Spatial heterogeneity driven scale self-adaptation: the best spatial analysis scale is quantitatively determined based on the range of the semi-variogram function, the subjectivity of the fixed window is overcome, and the spatial differentiation of the landscape is accurately reflected; 2. Probability truncation filtering improves robustness: 85%-99% quantile constraints are used for data truncation, and the mean or median after truncation is calculated as the benchmark value, which effectively suppresses the interference of extreme values, thereby improving the robustness of the benchmark result; 3. Dynamic iteration of interannual benchmark sequence: an annual updated benchmark time sequence is established, the vegetation succession dynamics is captured in real time, and the error caused by ignoring time change is eliminated; 4. Multi-source data fusion verification framework: coupling ground plot measurement, unmanned aerial vehicle aerial survey and multi-temporal remote sensing inversion data, a verification system with strong spatial representation is constructed, which can effectively calibrate and evaluate the results.

[0082] This scheme significantly overcomes the spatial and temporal limitations of traditional fixed reference, and improves the evaluation accuracy to more than 85%, providing reliable technical support for the precise management of grassland ecosystems. This method can be widely applied to grassland ecological protection project effectiveness evaluation, degraded grassland restoration planning, grassland ecosystem carbon sink capacity dynamic monitoring, and can also provide key quantitative tools for monitoring and achieving "zero growth of land degradation".

[0083] Based on the same principle as the method shown in Figure 1 The embodiment of the application also provides a grassland ecological degradation recovery effectiveness dynamic evaluation device 20, as shown in Figure 9 The grassland ecological degradation recovery effectiveness dynamic evaluation device 20 can include an acquisition module 210, a reference time sequence determination module 220, a relative change rate determination module 230 and an evaluation module 240, wherein: the acquisition module 210 is configured to acquire annual remote sensing data for a target area, the target area being an area including vegetation, and the annual remote sensing data being selected from at least two years of remote sensing data corresponding to the target area; The reference time sequence determination module 220 is configured to perform vegetation feature extraction on the annual remote sensing data based on a sliding operation of a predetermined window on the annual remote sensing data, to obtain a reference time sequence, wherein the window is quantitatively determined based on a semi-variogram range, and the reference time sequence is data describing the change of the vegetation feature within a year, and each sliding operation corresponds to a reference value in the reference time sequence; The relative change rate determination module 230 is configured to calculate the relative change rate of the vegetation in the target area based on the reference time sequence; and the evaluation module 240 is configured to evaluate the degradation or recovery of the vegetation in the target area according to the relative change rate and a preset threshold.

[0084] Optionally, the window is determined based on the following manner: According to the annual remote sensing data and the semi-variogram function, an estimated value of a semi-variance graph corresponding to a first sample point and a second sample point in the annual remote sensing data is determined; Based on the estimated value, the window is determined.

[0085] Optionally, the reference time sequence determination module 220 is specifically configured to: Based on the sliding operation of the window in the annual remote sensing data, all pixels in the window corresponding to each sliding operation are determined; For each sliding operation, based on all pixels in the window corresponding to the sliding operation, the 85th-99th percentile is calculated; For each sliding operation, a truncated data set is constructed based on the 85%-99% quantile corresponding to the sliding operation and all pixels; for each sliding operation, a reference value corresponding to the sliding operation is determined according to the truncated data set corresponding to the sliding operation; Based on the reference values corresponding to all sliding operations, a reference time series corresponding to the interannual remote sensing data is determined.

[0086] Optionally, for each sliding operation, when determining the reference value corresponding to the sliding operation according to the truncated data set corresponding to the sliding operation, the reference time series determination module 220 is specifically configured to: For each sliding operation, a candidate reference value corresponding to the sliding operation is determined according to the truncated data set corresponding to the sliding operation, the candidate reference value including a reference value determined based on a mean value of the truncated data set and a reference value determined based on a median value of the truncated data set; for each sliding operation, a normal distribution verification is performed on the truncated data set based on a Shapiro-Wilk normality test, and the reference value corresponding to the sliding operation is determined from the candidate reference value according to a verification result.

[0087] Optionally, the device further comprises: An updating module configured to acquire new interannual remote sensing data, and update the reference time series according to the new interannual remote sensing data.

[0088] Optionally, the device further comprises: A calibration module configured to acquire verification data for the target region, the verification data being ground measured data or unmanned aerial vehicle aerial survey data; perform consistency verification on the verification data and the interannual remote sensing data, and calibrate an evaluation result of the target region according to a verification result, the evaluation result being a result of evaluating a degradation or recovery situation of vegetation in the target region.

[0089] Optionally, the device further comprises: A judgment module configured to, for each sliding operation, judge whether the reference value corresponding to the sliding operation is a top community threshold value based on the reference value corresponding to the sliding operation, a pre-determined minimum continuous area threshold value, a mean value and a standard deviation of all reference values corresponding to the target region.

[0090] The grassland ecological degradation and recovery effectiveness dynamic evaluation device can execute the grassland ecological degradation and recovery effectiveness dynamic evaluation method provided by the embodiments of the present application, and the implementation principles are similar. The actions performed by each module and unit in the grassland ecological degradation and recovery effectiveness dynamic evaluation device in the embodiments of the present application are corresponding to the steps in the grassland ecological degradation and recovery effectiveness dynamic evaluation method in the embodiments of the present application. The detailed function description of each module of the grassland ecological degradation and recovery effectiveness dynamic evaluation device can be referred to the description of the corresponding grassland ecological degradation and recovery effectiveness dynamic evaluation method in the foregoing, which will not be described here.

[0091] The grassland ecological degradation recovery effectiveness dynamic evaluation device can be a computer program (including program code) running in a computer device, for example, the grassland ecological degradation recovery effectiveness dynamic evaluation device is an application software; the device can be used to execute the corresponding steps in the method provided by the embodiments of the present application.

[0092] In some embodiments, the grassland ecological degradation recovery effectiveness dynamic evaluation device provided by the embodiments of the present application can be implemented in a combination of software and hardware, for example, the grassland ecological degradation recovery effectiveness dynamic evaluation device provided by the embodiments of the present application can be a processor in the form of a hardware decoding processor, which is programmed to execute the grassland ecological degradation recovery effectiveness dynamic evaluation method provided by the embodiments of the present application, for example, the processor in the form of a hardware decoding processor can use one or more application specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field programmable gate arrays (FPGAs) or other electronic components.

[0093] In some embodiments, the grassland ecological degradation recovery effectiveness dynamic evaluation device provided by the embodiments of the present application can be implemented in a combination of software and hardware, for example, the grassland ecological degradation recovery effectiveness dynamic evaluation device provided by the embodiments of the present application can be a processor in the form of a hardware decoding processor, which is programmed to execute the grassland ecological degradation recovery effectiveness dynamic evaluation method provided by the embodiments of the present application, for example, the processor in the form of a hardware decoding processor can use one or more application specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field programmable gate arrays (FPGAs) or other electronic components. Figure 9 The grassland ecological degradation recovery effectiveness dynamic evaluation device stored in the memory is shown, which can be software in the form of programs and plug-ins, and includes a series of modules, including the acquisition module 210, the reference time sequence determination module 220, the relative change rate determination module 230 and the evaluation module 240, for realizing the grassland ecological degradation recovery effectiveness dynamic evaluation method provided by the embodiments of the present application.

[0094] The modules described in the embodiments of the present application can be implemented by software or hardware. In some cases, the name of the module does not constitute a limitation on the module itself.

[0095] Based on the same principles as the method shown in the embodiments of the present application, the embodiments of the present application also provide an electronic device, which can include but is not limited to: a processor and a memory; the memory is used to store a computer program; the processor is used to execute the method shown in any embodiment of the present application by calling the computer program.

[0096] In one optional embodiment, an electronic device is provided, which can include but is not limited to: Figure 10As shown, Figure 10 The electronic device 4000 shown includes a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, for example, via a bus 4002. Optionally, the electronic device 4000 can also include a transceiver 4004, which can be used for data interaction between the electronic device and other electronic devices, such as data transmission and / or data reception, etc. It should be noted that in actual applications, the transceiver 4004 is not limited to one, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present application.

[0097] The processor 4001 can be a CPU (Central Processing Unit, central processor), a general-purpose processor, a DSP (Digital Signal Processor, data signal processor), an ASIC (Application Specific Integrated Circuit, application specific integrated circuit), an FPGA (Field Programmable Gate Array, field programmable gate array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It can implement or execute various exemplary logical blocks, modules and circuits described in combination with the present disclosure. The processor 4001 can also be a combination of computing functions, such as one or more microprocessor combinations, combinations of DSP and microprocessor, etc.

[0098] The bus 4002 can include a channel for transmitting information between the above-mentioned components. The bus 4002 can be a PCI (Peripheral Component Interconnect, peripheral component interconnect) bus or an EISA (Extended Industry Standard Architecture, extended industry standard architecture) bus, etc. The bus 4002 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 10 In the figure, only one thick line is used to represent the bus, but it does not mean that there is only one bus or only one type of bus.

[0099] The memory 4003 can be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, a magnetic disk storage or other magnetic storage devices, or any other medium capable of storing desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited thereto.

[0100] The memory 4003 is configured to store application code (computer program) for implementing the scheme of the present application, and the processor 4001 is configured to control the execution. The processor 4001 is configured to execute the application code stored in the memory 4003 to implement the content shown in the foregoing method embodiments.

[0101] The electronic device can also be a terminal device, Figure 10 The electronic device shown is only an example, and should not limit the functions and use range of the embodiments of the present application.

[0102] The embodiments of the present application provide a computer readable storage medium, and the computer readable storage medium stores a computer program. When the computer program runs on a computer, the computer can execute the corresponding content in the foregoing method embodiments.

[0103] According to another aspect of the present application, a computer program product or computer program is also provided, which includes computer instructions stored in a computer readable storage medium. A processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions to enable the computer device to perform the method provided in the various implementation manners of the above embodiments.

[0104] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0105] It should be understood that the flowchart and block diagrams in the drawings illustrate the architecture, functionality, and operation of possible implementations of various embodiments of the present application. In this regard, each block in the flowchart and block diagrams can represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks can sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and / or flowchart illustrations, and combinations thereof, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or combinations of hardware and software.

[0106] The computer readable storage medium of the present application embodiment can be, for example but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination thereof. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present application, the computer readable storage medium can be any tangible medium that contains or stores a program and can be used by an instruction execution system, apparatus, or device to retrieve and execute the program.

[0107] The computer readable storage medium described above bears one or more programs, when the one or more programs are executed by the electronic device, cause the electronic device to execute the method shown in the above embodiment.

[0108] The above description is merely the preferred embodiments of the present application and the explanation of the applied technical principles. It should be understood by those skilled in the art that the disclosed range of the present application is not limited to the technical solutions formed by the specific combinations of the above technical features, and should also cover other technical solutions formed by any combinations of the above technical features or their equivalent features without departing from the disclosed concept. For example, the technical solutions formed by the mutual replacement of the above features and the technical features disclosed in the present application (but not limited to) having similar functions.

Claims

1. A method for dynamically evaluating the effect of grassland ecological degradation recovery, characterized in that, The method comprises: obtaining inter-annual remote sensing data of a target region, the target region being a region comprising vegetation, the inter-annual remote sensing data being selected from at least two years of remote sensing data corresponding to the target region; extracting vegetation features from the inter-annual remote sensing data based on a sliding operation of a predetermined window on the inter-annual remote sensing data, to obtain a baseline time series, wherein the window is quantitatively determined based on a semi-variogram range, and the baseline time series is data describing the change of vegetation features within a year, each sliding operation corresponding to a baseline value in the baseline time series; calculating a relative change rate of vegetation in the target region based on the baseline time series; evaluating the degradation or recovery of vegetation in the target region according to the relative change rate and a preset threshold.

2. The method of claim 1, wherein, The window is determined based on the following method: determining an estimated value of a semi-variance graph corresponding to a first sample point and a second sample point in the inter-annual remote sensing data according to the inter-annual remote sensing data and a semi-variogram; determining the window based on the estimated value.

3. The method of claim 2, wherein, The extraction of vegetation features from the inter-annual remote sensing data based on the sliding operation of the predetermined window on the inter-annual remote sensing data to obtain the baseline time series comprises: determining all pixels within the window corresponding to each sliding operation based on the sliding operation of the window on the inter-annual remote sensing data; for each sliding operation, calculating the 85th-99th percentile based on all pixels within the window corresponding to the sliding operation; for each sliding operation, constructing a truncated data set based on the 85th-99th percentile and all pixels corresponding to the sliding operation; for each sliding operation, determining a baseline value corresponding to the sliding operation according to the truncated data set corresponding to the sliding operation; determining a baseline time series corresponding to the inter-annual remote sensing data based on the baseline values corresponding to all sliding operations.

4. The method of claim 3, wherein, For each sliding operation, determining a baseline value corresponding to the sliding operation according to the truncated data set corresponding to the sliding operation comprises: for each sliding operation, determining a candidate baseline value corresponding to the sliding operation according to the truncated data set, the candidate baseline value comprising a baseline value determined based on the mean of the truncated data set and a baseline value determined based on the median of the truncated data set; for each sliding operation, performing normal distribution verification on the truncated data set based on Shapiro-Wilk normality test, and determining the baseline value corresponding to the sliding operation from the candidate baseline value according to the verification result.

5. The method according to any one of claims 1 to 4, characterized in that, The method further comprises: obtaining new inter-annual remote sensing data, and updating the baseline time series according to the new inter-annual remote sensing data.

6. The method according to any one of claims 2 to 4, characterized in that, The method further comprises: obtaining verification data for the target region, the verification data being ground measured data or unmanned aerial vehicle survey data; performing consistency verification on the verification data and the inter-annual remote sensing data, and calibrating the evaluation result of the target region according to the verification result, the evaluation result being the result of evaluating the degradation or recovery of vegetation in the target region.

7. The method according to claim 3 or 4, characterized in that, The method further comprises: For each sliding operation, based on the reference value corresponding to the sliding operation, the predetermined minimum patch area threshold, the mean and standard deviation of all reference values corresponding to the target region, it is judged whether the reference value corresponding to the sliding operation is a top community threshold.

8. The device for evaluating the effect of the restoration of the ecological degradation of grassland, characterized in that, Comprise: An acquisition module is configured to acquire inter-annual remote sensing data for a target region, the target region being a region including vegetation, and the inter-annual remote sensing data being remote sensing data selected from at least two years of remote sensing data corresponding to the target region; A reference time series determination module is configured to perform vegetation feature extraction on the inter-annual remote sensing data based on a predetermined window sliding operation on the inter-annual remote sensing data, to obtain a reference time series, wherein the window is quantitatively determined based on a semi-variogram range, and the reference time series is data describing the change of vegetation features within a year, and each sliding operation corresponds to a reference value in the reference time series; A relative change rate determination module is configured to calculate a relative change rate of vegetation in the target region based on the reference time series; and an evaluation module is configured to evaluate the degradation or recovery of vegetation in the target region according to the relative change rate and a preset threshold.

9. An electronic device, comprising: The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the method in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the method in any one of claims 1-7.

Images