Intelligent acquisition method for vegetation coverage of mine ecological restoration

Abstract

The invention discloses a mine ecological restoration vegetation coverage intelligent acquisition method, which comprises the following steps: S1, data acquisition: acquiring image data of a mining area by using a high-resolution satellite remote sensing image and an aerial photography technology; s2, preprocessing: preprocessing the collected remote sensing image; s3, feature extraction: extracting a vegetation index from the preprocessed image; s4, applying a pixel bipartite model, and dividing each pixel into a vegetation part and a non-vegetation part by adopting the pixel bipartite model; s5, a machine learning algorithm; s6, model training and verification; s7, calculating the vegetation coverage; s8, analyzing and optimizing a result; and S9, performing dynamic monitoring. According to the method, the vegetation coverage of mine ecological restoration can be intelligently, rapidly and accurately obtained, the evaluation of the ecological restoration effect is facilitated, a scientific and reasonable ecological restoration scheme is formulated, and the method has important significance in promoting green development of mines and ecological civilization construction.

Description

Technical Field

[0001] The present invention relates to the technical field of mining area ecological restoration, and in particular to an intelligent acquisition method for vegetation coverage of mining ecological restoration. Background Art

[0002] Excessive exploitation of mineral resources in mines often leads to environmental deterioration of mine ecology. Since the deteriorated mine ecological environment often causes many hazards, such as regional water pollution and soil pollution degradation, it is very important to repair the deteriorated mine ecological environment. Vegetation restoration and reconstruction are often an indispensable part of mine ecological restoration work, so the vegetation coverage rate is often used as one of the important indicators to measure the effect of mine ecological environment restoration, which is of great significance to the effectiveness of mine restoration.

[0003] However, there is currently a lack of methods that can accurately and quickly obtain vegetation coverage. Due to the complex structure of the mining area and the large number of vegetation within it, the acquisition is easily affected by natural factors, such as atmospheric factors, wind speed, light intensity, etc., resulting in unclear images and inaccurate indexes. At the same time, after acquisition, it is not possible to combine existing technologies to accurately formulate a reasonable ecological restoration plan. As a result, when the plan is formulated later, due to natural factors and vegetation growth, the formulated plan has delays, and it is no longer possible to accurately solve the problems found in the previously acquired data. Therefore, in order to solve this series of problems, we propose an intelligent acquisition method for vegetation coverage for mine ecological restoration to solve the problem. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent method for obtaining vegetation coverage for mine ecological restoration to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: a method for intelligently obtaining vegetation coverage for mine ecological restoration, comprising the following steps:

[0006] S1. Data collection: using high-resolution satellite remote sensing images and aerial photography technology to obtain image data of the mining area;

[0007] S2, preprocessing, preprocessing the collected remote sensing images;

[0008] S3, feature extraction, extracting vegetation index from the preprocessed image;

[0009] S4, pixel binary model application, using the pixel binary model, each pixel is divided into vegetation and non-vegetation parts;

[0010] S5, machine learning algorithm, introduces machine learning algorithms, such as random forest, support vector machine (SVM) or deep learning network, to predict vegetation cover;

[0011] S6. Model training and verification: Use ground-based measured data to train and verify the model to ensure its accuracy and applicability;

[0012] S7, vegetation coverage calculation, using the above algorithm and model, calculate the vegetation coverage of each pixel and generate a vegetation coverage map;

[0013] S8. Result analysis and optimization: analyze the calculation results, compare them with historical data, and evaluate the ecological restoration effect;

[0014] S9. Dynamic monitoring: Establish a dynamic monitoring system, regularly update remote sensing data, and monitor changes in vegetation coverage in mining areas in real time to provide a scientific basis for mine ecological restoration.

[0015] Preferably, in step S2, the preprocessing includes image correction, denoising and atmospheric correction.

[0016] Preferably, in step S3, the feature extraction vegetation index includes the Normalized Difference Vegetation Index (NDVI) and the Enhanced Vegetation Index (EVI).

[0017] Preferably, the pixel binary model in step S4 assumes that the spectral reflectance of each pixel is a linear combination of vegetation and non-vegetation components, and calculates the NDVI values ​​of pure vegetation pixels and pure bare soil pixels to obtain vegetation coverage.

[0018] Preferably, the image correction includes geometric correction, radiation correction and color correction.

[0019] Preferably, the atmospheric correction first converts the remote sensing data into radiation brightness values, and then performs radiation calibration, uses dark pixels in the image (i.e., pixels with reflectivity close to 0) to estimate the atmospheric impact, and then performs correction to obtain the atmospheric quality index (AQI), and then uses a radiation transfer model to simulate the impact of the atmosphere on the remote sensing data, and then obtains the corrected remote sensing data through inversion calculation, estimates the atmospheric impact through ground measured data, and then removes these impacts from the remote sensing data.

[0020] Preferably, the calculation of the Enhanced Vegetation Index (EVI) is as follows:

[0021] EVI=G*(NIR-RED) / (NIR+C1*RED-C2*BLUE+L),

[0022] Among them, NIR represents the reflectance of the near-infrared band, RED represents the reflectance of the red light band, BLUE represents the reflectance of the blue light band, and G, C1, C2, and L are empirical coefficients used to adjust the sensitivity of EVI to different vegetation types and the degree of elimination of atmospheric influence.

[0023] Preferably, the calculation formula for calibration and apparent reflectivity in the radiation calibration is:

[0024] L=AG+N,P=πlh2 / Ecos(θ),

[0025] Among them, L is the apparent radiation brightness, A is the gain coefficient, G is the pixel radiation intensity, dimensionless; N is the deviation coefficient, P is the apparent reflectivity, dimensionless; π is a constant; h is the distance between the sun and the earth, in astronomical units; E is the solar irradiance, which is the solar zenith angle, in degrees.

[0026] Preferably, the atmospheric environment index (AQI) = max{IAQI1, IAQI2, IAQI3, ..., IAQIn}, and the atmospheric environment index (AQI) is the maximum value of the overall environmental sub-index IAQI of each pollutant item factor.

[0027] Compared with the existing technology, the beneficial effect of the present invention is that the method can intelligently, quickly and accurately obtain the vegetation coverage of mine ecological restoration, which is helpful to evaluate the effect of ecological restoration and formulate scientific and reasonable ecological restoration plans, which is of great significance to promoting the green development of mines and the construction of ecological civilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a flow chart of a method for intelligently acquiring vegetation coverage for mine ecological restoration according to the present invention. DETAILED DESCRIPTION

[0029] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] See also Figure 1 The present invention provides a technical solution: a technical solution for an intelligent method for obtaining vegetation coverage of mine ecological restoration, comprising the following steps:

[0031] S1. Data collection: using high-resolution satellite remote sensing images and aerial photography technology to obtain image data of the mining area;

[0032] S2. Preprocessing: The collected remote sensing images are preprocessed. The preprocessing includes image correction, denoising and atmospheric correction. Image correction includes geometric correction, radiation correction and color correction. Atmospheric correction first converts remote sensing data into radiation brightness values, and then performs radiation calibration. Dark pixels in the image (i.e. pixels with reflectivity close to 0) are used to estimate the atmospheric impact, so as to perform correction and obtain the atmospheric environment index (AQI). The radiation transfer model is then used to simulate the impact of the atmosphere on the remote sensing data. The corrected remote sensing data is then obtained through inversion calculation. The atmospheric impact is estimated through ground measured data, and then these impacts are removed from the remote sensing data. The calibration and apparent reflectance calculation formulas in radiation calibration are:

[0033] L=AG+N,P=πlh2 / Ecos(θ),

[0034] Among them, L is the apparent radiation brightness, A is the gain coefficient, G is the pixel radiation intensity, dimensionless; N is the deviation coefficient, P is the apparent reflectivity, dimensionless; π is a constant; h is the distance between the sun and the earth, in astronomical units; E is the solar irradiance, which is the solar zenith angle, in degrees,

[0035] Atmospheric environment index (AQI) = max{IAQI1, IAQI2, IAQI3, ..., IAQIn}, the atmospheric environment index (AQI) is the maximum value of the overall environmental sub-index IAQI of each pollutant project factor;

[0036] S3. Feature extraction: Extract vegetation index from the preprocessed image. The feature extraction vegetation index includes the Normalized Difference Vegetation Index (NDVI) and the Enhanced Vegetation Index (EVI). The model assumes that the spectral reflectance of each pixel is a linear combination of vegetation and non-vegetation components. The NDVI values ​​of pure vegetation pixels and pure bare soil pixels are calculated to obtain the vegetation coverage. The calculation of the Enhanced Vegetation Index (EVI) is shown as follows:

[0037] EVI=G*(NIR-RED) / (NIR+C1*RED-C2*BLUE+L),

[0038] Among them, NIR represents the reflectance of the near-infrared band, RED represents the reflectance of the red light band, BLUE represents the reflectance of the blue light band, and G, C1, C2, and L are empirical coefficients used to adjust the sensitivity of EVI to different vegetation types and the degree of elimination of atmospheric influences.

[0039] S4, pixel binary model application, using the pixel binary model, each pixel is divided into vegetation and non-vegetation parts;

[0040] S5, machine learning algorithm, introduces machine learning algorithms, such as random forest, support vector machine (SVM) or deep learning network, to predict vegetation cover;

[0041] S6. Model training and verification: Use ground-based measured data to train and verify the model to ensure its accuracy and applicability;

[0042] S7, vegetation coverage calculation, using the above algorithm and model, calculate the vegetation coverage of each pixel and generate a vegetation coverage map;

[0043] S8. Result analysis and optimization: analyze the calculation results, compare them with historical data, and evaluate the ecological restoration effect;

[0044] S9. Dynamic monitoring: Establish a dynamic monitoring system, regularly update remote sensing data, and monitor changes in vegetation coverage in mining areas in real time to provide a scientific basis for mine ecological restoration.

[0045] Working principle: The invention obtains image data of the mining area by using high-resolution satellite remote sensing images and aerial photography technology, performs image correction, denoising and atmospheric correction on the collected remote sensing images, and then obtains accurate image data of the mining area. The corrected data is then subjected to feature extraction to obtain the Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI). A pixel binary model is then used to divide each pixel into vegetation and non-vegetation parts. Machine learning algorithms such as random forests, support vector machines (SVM) or deep learning networks are then introduced to predict vegetation coverage. Ground measured data is used to obtain the vegetation coverage of the mining area. The model is trained and verified to ensure its accuracy and applicability. Through the above algorithms and models, the vegetation coverage of each pixel is calculated, and a vegetation coverage map is generated. The calculation results are analyzed and compared with historical data to evaluate the effect of ecological restoration. A dynamic monitoring system is established, remote sensing data is updated regularly, and changes in vegetation coverage in mining areas are monitored in real time to provide a scientific basis for mine ecological restoration. The vegetation coverage of mine ecological restoration can be obtained intelligently, quickly and accurately, which helps to evaluate the effect of ecological restoration and formulate scientific and reasonable ecological restoration plans, which is of great significance to promoting green development of mines and the construction of ecological civilization.

[0046] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0047] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent method for obtaining vegetation coverage for mine ecological restoration, characterized in that: The following steps are involved: S1. Data collection: using high-resolution satellite remote sensing images and aerial photography technology to obtain image data of the mining area; S2, preprocessing, preprocessing the collected remote sensing images; S3, feature extraction, extracting vegetation index from the preprocessed image; S4, pixel binary model application, using the pixel binary model, each pixel is divided into vegetation and non-vegetation parts; S5, machine learning algorithm, introduces machine learning algorithms, such as random forest, support vector machine (SVM) or deep learning network, to predict vegetation cover; S6. Model training and verification: Use ground-based measured data to train and verify the model to ensure its accuracy and applicability; S7, vegetation coverage calculation, using the above algorithm and model, calculate the vegetation coverage of each pixel and generate a vegetation coverage map; S8. Result analysis and optimization: analyze the calculation results, compare them with historical data, and evaluate the ecological restoration effect; S9. Dynamic monitoring: Establish a dynamic monitoring system, regularly update remote sensing data, and monitor changes in vegetation coverage in mining areas in real time to provide a scientific basis for mine ecological restoration.

2. According to claim 1, the intelligent acquisition method of vegetation coverage for mine ecological restoration is characterized by: In step S2, the preprocessing includes image correction, noise removal and atmospheric correction.

3. The intelligent acquisition method for vegetation coverage of mine ecological restoration according to claim 1 is characterized by: In step S3, the feature extraction vegetation index includes the Normalized Difference Vegetation Index (NDVI) and the Enhanced Vegetation Index (EVI).

4. The intelligent acquisition method for vegetation coverage of mine ecological restoration according to claim 1 is characterized by: The pixel binary model in step S4 assumes that the spectral reflectance of each pixel is a linear combination of vegetation and non-vegetation components, and calculates the NDVI values ​​of pure vegetation pixels and pure bare soil pixels to obtain the vegetation coverage.

5. The intelligent acquisition method for vegetation coverage of mine ecological restoration according to claim 2 is characterized by: The image correction includes geometric correction, radiation correction and color correction.

6. The intelligent acquisition method of vegetation coverage for mine ecological restoration according to claim 2 is characterized by: The atmospheric correction first converts the remote sensing data into radiation brightness values, and then performs radiation calibration. The dark pixels in the image (i.e., pixels with a reflectivity close to 0) are used to estimate the atmospheric impact, thereby performing correction and obtaining the atmospheric quality index (AQI). The radiation transfer model is then used to simulate the impact of the atmosphere on the remote sensing data, and then the corrected remote sensing data is obtained through inversion calculation. The atmospheric impact is estimated through ground measured data, and then these impacts are removed from the remote sensing data.

7. The intelligent acquisition method of vegetation coverage for mine ecological restoration according to claim 3 is characterized by: The calculation of the Enhanced Vegetation Index (EVI) is shown below: EVI=G*(NIR-RED) / (NIR+C1*RED-C2*BLUE+L), Among them, NIR represents the reflectance of the near-infrared band, RED represents the reflectance of the red light band, BLUE represents the reflectance of the blue light band, and G, C1, C2, and L are empirical coefficients used to adjust the sensitivity of EVI to different vegetation types and the degree of elimination of atmospheric influence.

8. The intelligent acquisition method for vegetation coverage of mine ecological restoration according to claim 6 is characterized by: The calculation formula of calibration and apparent reflectivity in the radiation calibration is: L=AG+N,P=πlh2 / Ecos(θ), Among them, L is the apparent radiation brightness, A is the gain coefficient, G is the pixel radiation intensity, dimensionless; N is the deviation coefficient, P is the apparent reflectivity, dimensionless; π is a constant; h is the distance between the sun and the earth, in astronomical units; E is the solar irradiance, which is the solar zenith angle, in degrees.

9. The intelligent acquisition method for vegetation coverage of mine ecological restoration according to claim 6 is characterized by: The atmospheric environment index (AQI) = max{IAQI1, IAQI2, IAQI3, ..., IAQIn}, and the atmospheric environment index (AQI) is the maximum value of the overall environmental sub-index IAQI of each pollutant item factor.

Images