Vegetation recovery monitoring method based on kernel density

Abstract

The invention discloses a vegetation recovery monitoring method based on kernel density. The vegetation recovery monitoring method comprises the following steps: step 1, monitoring the vegetation recovery; data preparation: kNDVI data are collected and preprocessed, and the data are in a time sequence form; 2, unary linear trend analysis: carrying out linear trend analysis on the kNDVI data to obtain a trend coefficient; step 3, Mann-Kendall significance test is carried out, and Mann-Kendall trend test is used to evaluate whether the kNDVI time sequence has a significant trend or not; step 4, calculating SensSlope (slope estimation), and calculating SensSlope as the size estimation of the trend; step 5, spatial autocorrelation analysis: using Moran's I or other spatial autocorrelation methods to evaluate the spatial distribution condition of the trend; and step 6, result visualization: visualizing a trend result, including a spatial distribution diagram of the trend, a significance distribution diagram and a SensSlope distribution diagram. Compared with the prior art, the kNDVI nonlinear conversion method has the advantages that the nonlinear conversion process of the NDVI is linearized through a kernel method, the saturation phenomenon is reduced, and the adaptability is higher.

Description

Technical Field

[0001] The invention relates to the field of monitoring methods, and in particular to a vegetation restoration monitoring method based on kernel density. Background Art

[0002] The rapid development of a country cannot be separated from the strong guarantee of energy supply. As one of the main energy sources, the coal industry plays a vital role in it. While the large-scale and high-intensity coal mining in the Huainan mining area provides energy for regional development, it also inevitably causes a series of ecological and environmental problems such as surface subsidence, reduced farmland production and pollution. The improvement of climate conditions such as temperature and precipitation, and land reclamation and ecological restoration measures including soil improvement and reconstruction have also improved the ecological environment [5]. Due to the limitations of manpower, material resources and spatial distribution, traditional ground monitoring can usually only conduct point monitoring in local areas, which is difficult to fully and timely reflect the overall changes in the ecosystem. Satellite remote sensing technology relies on spectral data, which can cover a wide range of space and continuously obtain surface information in time, thereby realizing comprehensive and dynamic monitoring of the ecological environment in a large area.

[0003] Vegetation is an important component of terrestrial ecosystems. As a bridge between the atmosphere, hydrosphere, biosphere and lithosphere, its changes are both temporally complex and spatially diverse. Therefore, changes in vegetation are often used to measure changes in regional ecological environments. Various vegetation indices, including the Normalized Difference Vegetation Index (NDVI), are widely used to monitor the time series of vegetation growth.

[0004] NDVI (Normalized Difference Vegetation Index) is a widely used remote sensing indicator for assessing vegetation cover and biomass. However, NDVI has some limitations, mainly including:

[0005] (1) Affected by cloud cover and atmospheric conditions: Changes in cloud cover and atmospheric conditions may affect the accuracy of NDVI.

[0006] (2) Sensitive to lighting conditions: Changes in the sun’s altitude and other lighting conditions may affect NDVI readings.

[0007] (3) Not suitable for areas with low vegetation coverage: NDVI does not work well in areas with bare land or low vegetation coverage.

[0008] EVI (Enhanced Vegetation Index) was proposed to address some of the deficiencies of NDVI by using information from other bands to compensate for the saturation problem of NDVI. However, EVI still has saturation problems and may not be sensitive enough in areas with high vegetation cover. Summary of the invention

[0009] The technical problem to be solved by the present invention is to overcome the above technical defects and provide a kNDVI (kernel normalized difference vegetation index) to replace NDVI to evaluate the vegetation restoration effect in mining areas, which can more effectively and stably monitor the vegetation restoration effect in mining areas, and can effectively illustrate the monitoring method of the vegetation restoration rate in mining areas by using trend analysis and spatial analysis.

[0010] In order to solve the above problems, the technical solution of the present invention is: a vegetation restoration monitoring method based on kernel density, including the following monitoring methods:

[0011] Step 1: Data preparation: collect and preprocess kNDVI data in the form of time series with the same spatial resolution and time span;

[0012] Step 2: Univariate linear trend analysis: perform linear trend analysis on kNDVI data to obtain the trend coefficient;

[0013] Step 3: Mann-Kendall significance test: Use the Mann-Kendall trend test to assess whether there is a significant trend in the kNDVI time series.

[0014] Step 4: Calculate SensSlope (slope estimation). Calculate SensSlope as an estimate of the size of the trend.

[0015] Step 5: Spatial autocorrelation analysis, using Moran's I or other spatial autocorrelation methods to assess the spatial distribution of trends;

[0016] Step 6: Result visualization, visualize trend results, including spatial distribution map of trend, significance distribution map and SensSlope distribution map.

[0017] Furthermore, the kNDVI is a vegetation index calculated based on remote sensing image data. kNDVI uses the kernel density estimation method to smooth the NDVI, aiming to reduce noise interference and enhance spatial continuity, thereby improving the accuracy of the vegetation index. The calculation formula of kNDVI is as follows:

[0018]

[0019] Where NIR represents the near-infrared band, Red represents the infrared band, and σ represents the length scale parameter, which controls the sensitivity between dense or sparse vegetation pixels or areas and is usually set to the average of the reflectance of the near-infrared and red bands.

[0020] Furthermore, the kNDVI linearizes the nonlinear conversion process of NDVI through a kernel method.

[0021] The advantages of the present invention compared with the prior art are:

[0022] The present invention utilizes linearization processing: kNDVI linearizes the nonlinear conversion process of NDVI through a kernel method, thereby avoiding the problem of saturation of the vegetation index in areas with high vegetation coverage or high biomass due to the nonlinearity and saturation of the vegetation index, and reducing the saturation phenomenon: kNDVI alleviates the saturation problem of NDVI in densely vegetated areas, which makes the performance of kNDVI in high-coverage vegetation areas better than traditional NDVI, and has stronger adaptability: kNDVI can adapt to different vegetation types and environmental conditions by adjusting parameters, thereby improving the applicability and accuracy of the vegetation index. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 The present invention is a flow chart of a vegetation restoration monitoring method based on kernel density. DETAILED DESCRIPTION

[0024] The specific implementation of the present invention is further described below in conjunction with the accompanying drawings, wherein the same components are represented by the same reference numerals.

[0025] It should be noted that the words "front", "rear", "left", "right", "up" and "down" used in the following description refer to directions in the drawings, and the words "inside" and "outside" refer to directions toward or away from the geometric center of a specific component, respectively.

[0026] In order to make the contents of the present invention more clearly understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0027] like Figure 1 As shown, a vegetation restoration monitoring method based on kernel density includes the following monitoring methods:

[0028] Step 1: Data preparation: collect and preprocess kNDVI data in the form of time series with the same spatial resolution and time span;

[0029] Step 2: Univariate linear trend analysis: perform linear trend analysis on kNDVI data to obtain the trend coefficient;

[0030] Step 3: Mann-Kendall significance test: Use the Mann-Kendall trend test to assess whether there is a significant trend in the kNDVI time series.

[0031] Step 4: Calculate SensSlope (slope estimation). Calculate SensSlope as an estimate of the size of the trend.

[0032] Step 5: Spatial autocorrelation analysis, using Moran's I or other spatial autocorrelation methods to assess the spatial distribution of trends;

[0033] Step 6: Result visualization, visualize trend results, including spatial distribution map of trend, significance distribution map and SensSlope distribution map.

[0034] The kNDVI is a vegetation index calculated based on remote sensing image data. kNDVI uses the kernel density estimation method to smooth the NDVI, aiming to reduce noise interference and enhance spatial continuity, thereby improving the accuracy of the vegetation index. The calculation formula of kNDVI is as follows:

[0035]

[0036] In the formula, NIR represents the near-infrared band, Red represents the infrared band, represents the length scale parameter, which controls the sensitivity between dense or sparse vegetation pixels or areas and is usually set to the average of the reflectance of the near-infrared and red bands. The kNDVI linearizes the nonlinear conversion process of NDVI through the kernel method.

[0037] In specific use, to create a process using the Mann-Kendall (MK) trend test to analyze vegetation kNDVI data, you can follow the steps below.

[0038] 1. Data preparation - Collect and preprocess kNDVI data to ensure that the data is in time series form with the same spatial resolution and time span.

[0039] Input: time series kNDVI data;

[0040] Output: preprocessed kNDVI data;

[0041] 2. Univariate linear trend analysis - Perform linear trend analysis on the kNDVI data to obtain the trend coefficient. This step can be used as a preliminary understanding of the trend.

[0042] Input: preprocessed kNDVI data;

[0043] Output: trend coefficient and preliminary trend direction;

[0044] 3. Mann-Kendall significance test - Use the Mann-Kendall trend test to assess whether there is a significant trend in the kNDVI time series.

[0045] Input: time series kNDVI data;

[0046] Output: Z statistic, p value, significance level;

[0047] 4. Calculate SensSlope (slope estimate) - Calculate SensSlope as an estimate of the size of the trend. Compared with the slope of linear regression, SensSlope is less sensitive to outliers.

[0048] Input: kNDVI time series data;

[0049] Output: SensSlope value;

[0050] 5. Spatial Autocorrelation Analysis - Use Moran's I or other spatial autocorrelation methods to assess the spatial distribution of trends, especially clustering effects.

[0051] Input: MK trend test results and SensSlope results;

[0052] Output: spatial autocorrelation statistics and clustered area distribution;

[0053] 6. Result Visualization - Visualize trend results, such as spatial distribution map of trend, significance distribution map and SensSlope distribution map.

[0054] Input: MK test results, SensSlope results, spatial autocorrelation analysis results;

[0055] Output: Visual maps, trend reports;

[0056] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. A vegetation restoration monitoring method based on kernel density, characterized in that: The following monitoring methods are included: Step 1: Data preparation, collect and preprocess kNDVI data in the form of time series with the same spatial resolution and time span; Step 2: Univariate linear trend analysis: perform linear trend analysis on kNDVI data to obtain the trend coefficient; Step 3: Mann-Kendal l significance test, use the Mann-Kendal l trend test to evaluate whether there is a significant trend in the kNDVI time series; Step 4: Calculate SensSlope (slope estimation). Calculate SensSlope as an estimate of the size of the trend. Step 5: Spatial autocorrelation analysis, using Moran's I or other spatial autocorrelation methods to assess the spatial distribution of trends; Step 6: Result visualization, visualize trend results, including spatial distribution map of trend, significance distribution map and SensSlope distribution map.

2. The vegetation restoration monitoring method based on kernel density according to claim 1, characterized in that: The kNDVI is a vegetation index calculated based on remote sensing image data. kNDVI uses the kernel density estimation method to smooth the NDVI, aiming to reduce noise interference and enhance spatial continuity, thereby improving the accuracy of the vegetation index. The calculation formula of kNDVI is as follows: Where NIR represents the near-infrared band, Red represents the infrared band, and σ represents the length scale parameter, which controls the sensitivity between dense or sparse vegetation pixels or areas and is usually set to the average of the reflectance of the near-infrared and red bands.

3. The vegetation restoration monitoring method based on kernel density according to claim 1, characterized in that: The kNDVI linearizes the nonlinear conversion process of NDVI through a kernel method.

Images