Spatial and temporal change analysis method for vegetation coverage

Abstract

The invention relates to the field of ecological environment monitoring, and discloses a vegetation coverage spatio-temporal change analysis method, which comprises the following steps of: firstly, preprocessing remote sensing image data of a research area, including radiometric calibration, atmospheric correction and image splicing and cutting; then calculating a vegetation index value based on the normalized vegetation index, and extracting a vegetation coverage through a pixel bipartite model; and performing difference calculation on the vegetation coverage data in different periods, analyzing spatial and temporal change characteristics of vegetation coverage, and generating a vegetation coverage grading map and a change distribution map. According to the method, the spatial distribution and dynamic change trend of vegetation coverage can be accurately reflected, the defect that a traditional vegetation monitoring method is sensitive to soil background and atmosphere interference is overcome, the method has the advantages of being high in precision, wide in applicability and high in dynamic monitoring capacity, and a scientific basis is provided for formulation and effect evaluation of regional ecological protection measures.

Description

Technical Field

[0001] The present invention relates to the technical field of ecological environment monitoring, and particularly to a method for analyzing the spatio-temporal changes of vegetation coverage. Background Art

[0002] Vegetation coverage is an important indicator reflecting the growth status of surface vegetation and the quality of the ecosystem, and is of great significance for the dynamic monitoring of the regional ecological environment. Against the background of global climate change and the intensification of human activities, the dynamic changes of vegetation coverage in alpine and high-altitude regions are particularly hotspots in ecological research. However, there are some deficiencies in the existing technologies for vegetation coverage monitoring and change analysis.

[0003] Traditional vegetation monitoring methods mostly rely on ground measured data, but this method has limitations in terms of time and space scales and cannot meet the needs of large-scale and long-term dynamic monitoring. At the same time, although the extensive application of remote sensing technology has to a certain extent made up for the deficiencies of ground monitoring, some methods are affected by factors such as atmospheric interference and soil background effects during the data processing process, resulting in insufficient accuracy of vegetation coverage estimation. In addition, the existing methods lack systematicness in the analysis of vegetation changes in different periods, and have weak analytical capabilities for change trends and spatial distribution characteristics, making it difficult to comprehensively and accurately evaluate the dynamic changes of regional vegetation coverage.

[0004] Therefore, there is an urgent need for a method that can combine the high spatio-temporal resolution characteristics of remote sensing images, accurately estimate the vegetation coverage, and accurately analyze its dynamic change trends and spatial distribution characteristics, so as to provide a reliable basis for the protection and management of the ecological environment. Summary of the Invention

[0005] Aiming at the deficiencies of the existing technology, the present invention provides a method for analyzing the spatio-temporal changes of vegetation coverage, which solves the problems of insufficient accuracy, susceptibility to soil background and atmospheric interference, and weak analytical capabilities of dynamic changes in existing vegetation coverage monitoring.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: A method for analyzing the spatio-temporal changes of vegetation coverage, comprising the following steps:

[0007] Obtain the remote sensing image data of the target area, and preprocess the remote sensing image to eliminate the atmospheric and radiation effects, and extract the image data within the target area;

[0008] Calculate the vegetation index value of the pixels in the target area based on the remote sensing image data to reflect the vegetation distribution;

[0009] Estimate the vegetation coverage of the target area by using the pixel dichotomy model through the vegetation index value;

[0010] Create a vegetation coverage classification map based on vegetation coverage data and classify vegetation coverage;

[0011] By calculating the difference in vegetation coverage at different periods, the spatiotemporal variation characteristics of vegetation coverage in the target area are analyzed.

[0012] Preferably, the step of acquiring remote sensing image data of the target area includes:

[0013] Use the remote sensing image acquisition platform to obtain multispectral remote sensing images of the target area;

[0014] Filter image data with cloud cover less than the set value;

[0015] The acquired images are georeferenced and cropped to extract image data of the target area.

[0016] Preferably, the step of preprocessing the remote sensing image includes:

[0017] Perform radiometric correction on image data to eliminate errors caused by sensors and environmental factors during image acquisition;

[0018] Perform atmospheric correction and use the atmospheric correction model to remove the interference of atmospheric scattering and absorption on the image spectral information;

[0019] Multiple remote sensing images of the target area are stitched together to form remote sensing data that completely covers the target area.

[0020] Preferably, the step of calculating the vegetation index value of the pixels in the target area based on the remote sensing image data includes:

[0021] Extract the remote sensing data of the near infrared band and red light band of the target area, where the near infrared band is the 5th band of the Landsat8 image, and the red light band is the 4th band of the Landsat8 image;

[0022] The NDVI value of each pixel in the target area is calculated using the NDVI formula;

[0023] The calculated NDVI values ​​were processed for outliers, and the minimum and maximum values ​​of NDVI were determined by statistical cumulative probability distribution, and the NDVI values ​​with cumulative probabilities of 5% and 95% were taken as the upper and lower limits of the outliers, respectively;

[0024] The NDVI value range is limited to [-1, 1], and outliers exceeding this range are removed to reduce the impact of noise points on subsequent vegetation coverage calculations.

[0025] Preferably, the step of estimating the vegetation coverage of the target area by using the vegetation index value using a pixel binary model comprises:

[0026] Assume that the surface in the target area consists of a vegetated part and a bare soil part;

[0027] Use the pixel dichotomy model formula to estimate the vegetation coverage, and the formula is as follows:

[0028]

[0029] Among them, FVC represents the vegetation coverage of the pixel, NDVI soil is the NDVI value of the bare soil covered part, and NDVI veg is the NDVI value of the vegetated part.

[0030] Preferably, in the process of estimating the vegetation coverage, NDVI soil and NDVI veg are determined by the following methods:

[0031] Statistically analyze the cumulative probability distribution of NDVI values in the target area, and take the NDVI value with a cumulative probability of 5% as the lower limit of NDVI soil and take the NDVI value with a cumulative probability of 95% as the upper limit of NDVI veg ;

[0032] Adopt the large-sample pixel statistical method to eliminate abnormal NDVI values.

[0033] Preferably, the steps of making a vegetation coverage classification map based on the vegetation coverage data include:

[0034] Divide the value range of the vegetation coverage into multiple levels, and the classification criteria include extremely low vegetation coverage, low vegetation coverage, medium vegetation coverage, medium-high vegetation coverage, and high vegetation coverage;

[0035] Use the following classification ranges to classify the vegetation coverage:

[0036] Extremely low vegetation coverage: 0% to 10%;

[0037] Low vegetation coverage: 10% to 30%;

[0038] Medium vegetation coverage: 30% to 50%;

[0039] Medium-high vegetation coverage: 50% to 70%;

[0040] High vegetation coverage: 70% to 100%;

[0041] Use GIS software to visually process the classification results and generate a classification distribution map.

[0042] Preferably, the vegetation coverage classification further includes the following steps:

[0043] Statistically analyze the classification results, and calculate the area of each vegetation coverage level and its proportion in the total area of the target area;

[0044] Use map algebra method to generate the vegetation coverage level distribution map, and label the vegetation coverage level of the target area in units of pixels;

[0045] Use GIS tools to output the spatial distribution map of the vegetation coverage level, and conduct overlay analysis on the distribution results to identify the spatial distribution characteristics of the change in vegetation coverage level.

[0046] Preferably, the steps of analyzing the spatio-temporal change characteristics of the vegetation coverage degree of the target area by calculating the difference of the vegetation coverage degree data include:

[0047] Use the difference method to calculate the change value of the vegetation coverage degree in different periods. The specific formula is:

[0048] ΔFVC = FVC 2 - FVC 1

[0049] where ΔFVC is the change value of the vegetation coverage degree, and FVC 2 and FVC 1 are the vegetation coverage degrees in the later period and the earlier period respectively;

[0050] According to the magnitude of the change value of the vegetation coverage degree, divide the change area into the following levels: severe degradation, moderate degradation, slight degradation, no change, slight improvement, moderate improvement, and extreme improvement.

[0051] The present invention also provides a spatio-temporal change analysis device for vegetation coverage degree, including:

[0052] A data acquisition module, configured to acquire remote sensing image data of the target area and preprocess the image data;

[0053] A vegetation index calculation module, configured to calculate the vegetation index value of the target area based on the remote sensing image;

[0054] A vegetation coverage degree estimation module, configured to estimate the vegetation coverage degree of the target area based on the vegetation index value;

[0055] A grading module, configured to generate a vegetation coverage degree grading map based on the vegetation coverage degree data;

[0056] A change analysis module, configured to analyze the spatio-temporal change characteristics of the vegetation coverage degree of the target area by calculating the difference of the vegetation coverage degrees in different periods.

[0057] The present invention provides a spatio-temporal change analysis method for vegetation coverage degree. It has the following beneficial effects:

[0058] 1. The present invention uses the Normalized Difference Vegetation Index (NDVI) combined with the pixel dichotomy model to quantitatively calculate the vegetation coverage, which can minimize the influence of soil background and atmospheric interference to the greatest extent and ensure the accuracy of vegetation coverage estimation. At the same time, by screening the maximum and minimum values of NDVI through the cumulative probability distribution method, the interference of outliers to the calculation results is effectively avoided.

[0059] 2. By analyzing the difference in vegetation coverage in different periods, the present invention can intuitively reflect the change trend and spatial distribution characteristics of vegetation coverage in the study area, providing an efficient tool for vegetation dynamic monitoring. This method is particularly suitable for the long-term change assessment of ecologically sensitive areas and can provide a scientific basis for the formulation of ecological environmental protection policies.

[0060] 3. The present invention is applicable to remote sensing image data with various resolutions and study areas with different climatic conditions and ecological environments, and can effectively monitor the vegetation coverage from arid and semi-arid regions to alpine and high-altitude regions. This method has good versatility and scalability and strong adaptability.

[0061] 4. By analyzing the graded distribution of changes in vegetation coverage, the present invention can accurately evaluate the implementation effect of regional ecological projects. For example, the research in the source area of the Yellow River shows that the Sanjiangyuan Ecological Protection Project has significantly improved the vegetation coverage, further verifying the reliability and scientific value of the present invention in the evaluation of the effectiveness of ecological projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 is a schematic flow chart of the method of the present invention;

[0063] Figure 2 is a schematic structural diagram of the device of the present invention;

[0064] Figure 3 is a schematic diagram of the general situation of the study area in the embodiment of the present invention;

[0065] Figure 4 is a schematic diagram of the NDVI spatial distribution in 2016, 2019, and 2022 in the embodiment of the present invention;

[0066] Figure 5 is a schematic diagram of the vegetation coverage grades in 2016, 2019, and 2022 in the embodiment of the present invention;

[0067] Figure 6 is a schematic diagram of the graded distribution of changes in vegetation coverage from 2016 to 2022 in the embodiment of the present invention.

[0068] Among them, 10. Data acquisition module; 20. Vegetation index calculation module; 30. Vegetation coverage estimation module; 40. Grading module; 50. Change analysis module. DETAILED DESCRIPTION OF THE INVENTION

[0069] Next, in combination with the accompanying drawings of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0070] Please refer to the attached Figure 1 , the present invention provides a method for analyzing the spatio-temporal changes of vegetation coverage. Based on remote sensing image data and spatial analysis technology, it systematically analyzes the dynamic change characteristics of vegetation coverage in the target area, and clarifies the spatial distribution and trend of vegetation degradation or restoration.

[0071] The method for analyzing the spatio-temporal changes of vegetation coverage may include the following steps:

[0072] S1. Obtain the remote sensing image data of the target area and perform preprocessing;

[0073] S2. Calculate the vegetation index values of the pixels in the target area based on the remote sensing image data;

[0074] S3. Estimate the vegetation coverage of the target area by using the pixel dichotomy model through the vegetation index values;

[0075] S4. Make a vegetation coverage classification map based on the vegetation coverage data;

[0076] S5. Analyze the spatio-temporal change characteristics of the vegetation coverage in the target area by calculating the difference in vegetation coverage at different times.

[0077] The following is a detailed description of the specific implementation manners of each step of the method of the present invention.

[0078] For step S1, in this embodiment, step S1 obtains the remote sensing image data of the target area and preprocesses the remote sensing image to eliminate the influence of the atmosphere and radiation, and extracts the image data within the scope of the target area, providing reliable basic data for subsequent vegetation coverage analysis.

[0079] In the present invention, the acquisition of remote sensing image data preferably uses Landsat8 remote sensing images. This image has multi-spectral bands and a 30-meter spatial resolution, and can provide sufficiently detailed surface information to support large-scale vegetation coverage monitoring. At the same time, in order to reduce the influence of atmospheric interference such as cloud cover on the image data, the present invention ensures the clarity and accuracy of the image by screening image data with a cloud cover of less than 10%.

[0080] After the acquisition of image data, it is necessary to preprocess the data to eliminate the interference introduced by factors such as sensors, environmental conditions, and terrain, thereby improving the quality of the image data. In the present invention, the preprocessing process includes the following:

[0081] In this embodiment, first, radiometric correction is performed on the image. Specifically, by calibrating the digital number (DN value) of the remote sensing image and converting it into a reflectance or radiance value in physical terms, the systematic error that may exist during the data acquisition process by the sensor is eliminated, ensuring that the remote sensing image truly reflects the spectral characteristics of the ground surface.

[0082] Next, atmospheric correction is performed. The FLAASH (Fast Line-of-sight Atmospheric Analysis of Hypercubes) model is used for atmospheric correction. By removing the interference of atmospheric absorption and scattering on the spectral information in the remote sensing image, image data closer to the true reflectance of the ground surface is obtained. This step is particularly important for areas with complex atmospheric conditions such as plateaus and arid regions.

[0083] In this embodiment, to ensure the spatial positioning accuracy of the target area, geometric correction is performed on the image data. By using terrain control points and a digital elevation model (DEM), the geometric distortion problem of the image is corrected, thereby ensuring that the remote sensing image can accurately cover the target area.

[0084] The mosaicking of image data is also one of the important contents of this embodiment. In the case where the area of the target area is large, a single image may not cover the entire area. Therefore, in the present invention, seamless mosaicking of multiple images is performed to generate a remote sensing data set that completely covers the target area, laying a foundation for subsequent vegetation index calculation.

[0085] After the image mosaicking is completed, in combination with the vector boundary data of the target area, the image is cropped to extract the remote sensing data within the target area. The cropping process realizes accurate data extraction by matching the boundary vector data with the spatial coordinates of the remote sensing image, ensuring that subsequent analysis is only carried out for the target area.

[0086] After the image cropping is completed, the present invention further conducts a quality assessment of the image data, including detecting the spatial consistency, spectral consistency, and noise distribution of the image, to ensure that the processed image can meet the accuracy requirements for subsequent vegetation coverage analysis.

[0087] Through the above steps of image data acquisition and preprocessing, the present invention effectively improves the clarity and authenticity of the remote sensing image, providing reliable data support for subsequent vegetation coverage analysis. In this embodiment, all preprocessing steps can be implemented by common remote sensing software (such as ENVI 5.6), and have high versatility, capable of adapting to the needs of different data sources and target areas.

[0088] For step S2, in this embodiment, step S2 calculates the vegetation index values of the pixels in the target area based on remote sensing image data to reflect the spatial distribution and growth status of vegetation, providing basic data for subsequent vegetation coverage analysis.

[0089] In the present invention, it is preferably to use the Normalized Difference Vegetation Index (NDVI) as the vegetation index, which has the advantages of simple calculation and being able to significantly distinguish vegetation and non-vegetation areas. The calculation formula of NDVI is as follows:

[0090]

[0091] Wherein, NIR is the reflection value of the near-infrared band, and R is the reflection value of the red light band. For Landsat8 remote sensing image data, NIR corresponds to the 5th band of the image, and R corresponds to the 4th band of the image. This formula is based on the absorption and reflection characteristics of vegetation for different spectra: the strong absorption of red light and strong reflection of near-infrared light by healthy vegetation result in a higher NDVI value, while non-vegetation features (such as water bodies, bare soil, etc.) show a lower NDVI value.

[0092] In actual calculation, to ensure the accuracy and consistency of NDVI values, the present invention adopts the following steps for processing:

[0093] In this embodiment, first, the near-infrared band and red light band image data of the target area are extracted. During the extraction process, the required bands are directly selected through the cropped remote sensing data of the target area, avoiding unnecessary expansion of the calculation range. At the same time, the band data is standardized to adapt to subsequent mathematical operations.

[0094] Based on the above-extracted band data, the NDVI value of each pixel in the target area is calculated one by one. Specifically, a remote sensing software (such as ENVI or other software supporting raster calculation) is used to batch-process the formula to ensure the efficient completion of NDVI calculation for all pixels.

[0095] Due to the possible interference of noise, shadows, or other environmental factors in the actual image, the calculated NDVI values may be abnormal. In this embodiment, to eliminate the influence of these abnormal values on subsequent analysis, the cumulative probability distribution method is used to limit the range of NDVI values. Specifically, the NDVI values of all pixels in the target area are statistically sorted, and the NDVI value with a cumulative probability of 5% is taken as the lower limit, and the NDVI value with a cumulative probability of 95% is taken as the upper limit. The NDVI values below the lower limit or above the upper limit are excluded or replaced.

[0096] In this embodiment, after the above outlier processing, the range of NDVI values is effectively limited to [-1, 1], where:

[0097] The pixel area with a positive NDVI value usually represents the vegetation coverage area. The closer the NDVI value is to 1, the higher the vegetation density and the better the growth condition;

[0098] The pixel area with an NDVI value close to 0 represents bare land or sparse vegetation coverage;

[0099] The pixel area with a negative NDVI value represents water bodies or snow and ice coverage.

[0100] After the calculation and processing of NDVI values are completed, a complete NDVI distribution map of the target area is generated to visually display the spatial variation of the vegetation index. At the same time, through the statistical analysis of the NDVI distribution map, necessary basic data can be provided for the subsequent estimation of vegetation coverage.

[0101] To sum up, in this embodiment, through the reasonable extraction of remote sensing image band data, NDVI calculation, and combined with outlier processing, the accuracy and availability of the vegetation index in the target area are ensured. This step provides complete and reliable vegetation index data support for the subsequent vegetation coverage analysis and has good applicability in the application of remote sensing images with different resolutions and different types.

[0102] For step S3, in this embodiment, step S3 estimates the vegetation coverage of the target area using the pixel dichotomy model through the vegetation index value, aiming to quantify the vegetation coverage ratio of pixels and provide necessary data support for the subsequent vegetation coverage classification and change analysis.

[0103] In the present invention, the pixel dichotomy model is adopted as the core method for estimating vegetation coverage. This model assumes that the surface of the target area is composed of a vegetation-covered part and a bare soil-covered part, and the spectral information obtained by the sensor is a linear superposition of these two components in proportion. Specifically, the actual vegetation coverage (FVC, Fractional Vegetation Cover) of a certain pixel can be linearly decomposed through its normalized vegetation index (NDVI) and the NDVI values of pure vegetation pixels and pure bare soil pixels. The calculation formula is as follows:

[0104]

[0105] Where:

[0106] FVC: represents the vegetation coverage of the pixel, with a range of [0, 1];

[0107] NDVI: is the normalized vegetation index value of the current pixel;

[0108] NDVIsoil : The NDVI value of the pixel covered by pure bare soil;

[0109] NDVI veg : The NDVI value of the pixel covered by pure vegetation.

[0110] To ensure the accuracy of the calculation results, in the present invention, statistical methods are used to determine the NDVI soil and NDVI veg values. Specifically:

[0111] By performing cumulative probability distribution statistics on the NDVI values of all pixels in the target area, the NDVI value with a cumulative probability of 5% is taken as NDVI soil , to reflect the minimum vegetation index value of the bare soil covered part in the target area;

[0112] The NDVI value with a cumulative probability of 95% is taken as NDVI veg , to reflect the maximum vegetation index value of the vegetation covered part in the target area.

[0113] During the calculation process, in order to reduce the influence of data noise on the estimation of vegetation coverage, in the present invention, outlier values are also removed through the statistical analysis method of large sample pixels. These outlier values may come from sensor errors, image shadows or other environmental interferences. After removing the outlier values, the calculation of vegetation coverage can better reflect the actual surface conditions.

[0114] In this embodiment, based on the above formula and parameter determination method, the vegetation coverage of the target area is calculated pixel by pixel using remote sensing software (such as ENVI). During the calculation process, the NDVI values of each pixel are processed in batches and matched with the preset NDVI soil and NDVI veg to generate a vegetation coverage dataset.

[0115] After completing the estimation of vegetation coverage, the results are stored as a raster format file, and the value of each pixel represents the vegetation coverage ratio of the corresponding surface. Specifically:

[0116] Pixels with FVC values close to 0 represent areas with extremely low vegetation coverage or even no vegetation;

[0117] Pixels with FVC values close to 1 represent areas with complete vegetation coverage;

[0118] Pixels between the two represent areas with partial vegetation coverage, and the higher the value, the higher the vegetation coverage degree.

[0119] Through this step, the vegetation coverage information in the target area has been effectively quantified. The generated vegetation coverage data can not only intuitively reflect the surface vegetation coverage status, but also provide a solid data foundation for subsequent classification and change analysis.

[0120] For step S4, in this embodiment, step S4 produces a vegetation coverage classification map based on the vegetation coverage data and classifies the vegetation coverage. This step aims to clarify the spatial distribution characteristics of different vegetation coverage degrees in the target area through classification processing and visual display, laying a foundation for the comprehensive evaluation of vegetation coverage and change trend analysis.

[0121] In the present invention, the vegetation coverage data is derived from the calculation result of step S3, that is, the vegetation coverage ratio (FVC) of each pixel, and its value range is [0,1], and the higher the value, the higher the vegetation coverage. Based on these data, the vegetation coverage of the target area is classified by a classification method.

[0122] In order to more intuitively describe and analyze the vegetation coverage, the present invention sets a classification standard for vegetation coverage, and divides the vegetation coverage into five levels. The range of each level is as follows:

[0123] Very low vegetation cover: 0% to 10%;

[0124] Low vegetation cover: 10% to 30%;

[0125] Medium vegetation coverage: 30% to 50%;

[0126] Medium to high vegetation coverage: 50% to 70%;

[0127] High vegetation cover: 70% to 100%.

[0128] The classification process is implemented through the map algebra function in GIS tools (such as ArcGIS). Specifically, the FVC value of each pixel is compared with the above classification standards to determine its vegetation cover level, and unique identifiers (such as 1 to 5) are assigned to different levels. In the generated raster data, the value of each pixel is its corresponding vegetation cover level.

[0129] In order to further quantify the distribution of each vegetation coverage level in the target area, the classification results are statistically processed in this embodiment. The specific method is to calculate the number of pixels corresponding to each level and its proportion to the total number of pixels in the target area through GIS tools, so as to obtain the area distribution and percentage of each level. For example, the area of ​​the high vegetation coverage area is the sum of the areas of all pixels with FVC values ​​greater than 0.7, and its proportion is the ratio of this area to the total area of ​​the target area.

[0130] After generating the hierarchical data, the hierarchical results are visualized through GIS tools to create a vegetation coverage classification map of the target area. The classification map uses different colors or grayscale values to identify different vegetation coverage levels, visually presenting the spatial distribution characteristics of vegetation coverage. Specifically, the darker the color in the map, the higher the vegetation coverage, and the lighter the color, the lower the vegetation coverage.

[0131] In this embodiment, the creation of the vegetation coverage classification map realizes the spatial expression of the vegetation coverage information in the target area, enabling users to quickly identify the high and low distribution areas of vegetation coverage. At the same time, through hierarchical statistical data, the specific proportion of each level of vegetation coverage area can be determined, providing an accurate reference for subsequent analysis of vegetation coverage changes.

[0132] In summary, step S4 realizes the classified display and statistical quantification of vegetation coverage in the target area through the setting of classification criteria, the generation of classified data, and the production of classification maps, providing an important technical means for analyzing the spatio-temporal changes of vegetation coverage.

[0133] For step S5, in this embodiment, step S5 analyzes the spatio-temporal change characteristics of the vegetation coverage in the target area by calculating the difference in vegetation coverage at different times. This step quantitatively analyzes the vegetation coverage change through the difference method, and combines spatial distribution analysis to clarify the specific location and change trend of the change area, providing a reliable basis for evaluating the dynamic changes of the regional vegetation condition.

[0134] In the present invention, the calculation of the change in vegetation coverage is based on the vegetation coverage data at different times. Specifically, by calculating the difference between the later vegetation coverage and the previous vegetation coverage, the change value of the vegetation coverage in the target area is obtained. The mathematical expression for the difference calculation is as follows:

[0135] ΔFVC = FVC 2 - FVC 1

[0136] Where:

[0137] ΔFVC: represents the change value of vegetation coverage of a certain pixel between different time periods;

[0138] FVC 2 : represents the vegetation coverage value of a certain pixel in the target area in the later period;

[0139] FVC 1 : represents the vegetation coverage value of a certain pixel in the target area in the previous period.

[0140] In this embodiment, the change value of vegetation coverage is calculated for each pixel through the above formula, and the calculation results are stored as a raster format file. Each pixel value in the file represents the degree of change in vegetation coverage at that location between different time periods.

[0141] After calculating the change values, in order to more intuitively express the change trend of vegetation coverage, this embodiment classifies the change values of vegetation coverage. According to the magnitude and sign of the change values, the changed areas are divided into the following seven levels:

[0142] Severe degradation (significant negative value range);

[0143] Moderate degradation;

[0144] Slight degradation;

[0145] No change (range close to zero);

[0146] Slight improvement;

[0147] Moderate improvement;

[0148] Extreme improvement (significant positive value range).

[0149] By visualizing the spatial distribution of the classification results, a distribution map of the change levels of vegetation coverage is generated. Different colors or gray-scale values are used in the figure to identify each change level, enabling users to intuitively understand the distribution characteristics and trends of vegetation coverage changes in the target area.

[0150] In order to quantify the distribution of each change level, this embodiment conducts statistical processing on the classification results. Specifically, the number of pixels corresponding to each change level and its proportion of the total number of pixels in the target area are calculated, so as to obtain the area distribution and percentage of each change level. For example, the area of the slightly improved area is the sum of the areas of all pixels whose change values are within the slightly improved range, and its proportion is the ratio of this area to the total area of the target area.

[0151] On the basis of the change analysis, this embodiment further combines the geographical information and ecological background data of the target area to explore the driving factors of vegetation changes. Specifically, it includes:

[0152] Analyze whether the vegetation degradation area is consistent with the river valley areas or residential areas with frequent human activities;

[0153] Evaluate whether the improved area conforms to the implementation scope of the national ecological protection policy, such as the role of the Sanjiangyuan Ecological Protection Project;

[0154] Combined with natural factors such as terrain, climate, and soil, further explain the possible causes of the changes.

[0155] Through this step, the present invention realizes the quantitative analysis of the spatio-temporal changes in vegetation coverage and the display of the spatial distribution. The positive or negative value and the magnitude of the change directly reflect the improvement or degradation of vegetation coverage. At the same time, combined with the change level distribution map, the location of the change area can be quickly located, providing scientific support for the dynamic assessment of the regional ecological status and the formulation of governance strategies.

[0156] In summary, step S5 comprehensively analyzes the dynamic change characteristics of vegetation coverage in the target area through the mathematical operation of the difference method, the division of change levels, and the visualization display of the spatial distribution.

[0157] Generally speaking, the present invention processes the remote sensing image data of the target area, calculates the normalized difference vegetation index and vegetation coverage in sequence, generates a vegetation coverage classification map by using the pixel dichotomy model and the classification method, and analyzes the spatio-temporal change characteristics of vegetation coverage in the area through the calculation of the difference in vegetation coverage in different periods. This method can accurately quantify the vegetation coverage status, intuitively display the vegetation distribution and change trend, provide a scientific basis for ecological environment monitoring, land use planning, and the formulation of vegetation protection measures, and has strong applicability and popularization value.

[0158] The spatio-temporal change analysis device of vegetation coverage described below can be correspondingly referred to the spatio-temporal change analysis method of vegetation coverage described above.

[0159] Please refer to the attached Figure 2 , the present invention also provides a spatio-temporal change analysis device of vegetation coverage, including:

[0160] A data acquisition module 10, configured to acquire remote sensing image data of the target area and preprocess the image data;

[0161] A vegetation index calculation module 20, configured to calculate the vegetation index value of the target area based on the remote sensing image;

[0162] A vegetation coverage estimation module 30, configured to estimate the vegetation coverage of the target area based on the vegetation index value;

[0163] A classification module 40, configured to generate a vegetation coverage classification map based on the vegetation coverage data;

[0164] A change analysis module 50, configured to analyze the spatio-temporal change characteristics of the vegetation coverage in the target area by calculating the difference in vegetation coverage in different periods.

[0165] The device of this embodiment can be used to execute the above method embodiment, and its principle and technical effect are similar, which will not be elaborated here.

[0166] To better understand the present invention, the above method will be described in detail below with specific embodiments.

[0167] Example:

[0168] 1. General situation of the study area

[0169] In this example, the study area is the source region of the Yellow River, which is located in the northeastern part of the Qinghai-Tibet Plateau, spanning three provinces of Sichuan, Gansu, and Qinghai. The geographical coordinate range is from 31.5°N to 36.5°N and from 95.5°E to 103.5°E, with a basin area of approximately 134,000 square kilometers. The terrain of the source region of the Yellow River is high in the west and low in the east, with an average altitude of 4,473 meters. It belongs to the plateau continental climate. The vegetation in the region is mainly alpine vegetation, characterized by sensitivity to climate change and fragility of the ecological environment. As Figure 3 shown, it is a general situation map of the study area in the source region of the Yellow River, showing the geographical location and spatial distribution characteristics of the study area.

[0170] 2. Data sources and preprocessing

[0171] The remote sensing data used in this example is from the Landsat 8 image data of the United States Geological Survey (USGS), with a resolution of 30 meters. Images with less cloud cover and capable of accurately reflecting the vegetation cover in the three years of 2016, 2019, and 2022 were selected. At the same time, combined with the vector boundary data of the study area to ensure that the analysis scope is consistent with the study area.

[0172] The selection of remote sensing images is based on the following two conditions:

[0173] The images should be basically free of cloud interference to avoid affecting the calculation of NDVI;

[0174] The images of the selected years should be obtained in periods with similar vegetation growth conditions to reduce the impact of vegetation seasonal differences on change analysis.

[0175] Due to the possible limitations of the remote sensing system in data recording in terms of time, space, and radiation resolution, errors may occur. In this example, ENVI 5.6 software was used to preprocess the Landsat 8 image data, including steps such as radiometric calibration, atmospheric correction, image mosaicking, and cropping. After preprocessing, it is ensured that the quality of the image data meets the requirements for subsequent vegetation coverage monitoring and analysis.

[0176] 3. Method

[0177] NDVI is an index reflecting the vegetation growth status and spatial distribution density. Calculate the NDVI values for 2016, 2019, and 2022 according to the formula ( Figure 4 is the NDVI spatial distribution map), and the calculation results are shown in the following table. The basic characteristics of NDVI in the study area are statistically obtained.

[0178] Year NDVI Min NDVI Max NDVI Mean 2016 -1 1 0.504263 2019 -1 1 0.496373 2022 -1 1 0.281822

[0179] As can be seen from the table, the NDVI values generally show a spatial distribution of low in the northwest and high in the central and southern parts, reflecting the topographic and climatic distribution characteristics of the Yellow River source area.

[0180] Vegetation coverage extraction

[0181] In this embodiment, the pixel dichotomy model is used to calculate the vegetation coverage. To avoid the noise interference of the maximum and minimum NDVI values, the cumulative probability distribution method is used in this study, and the 95% and 5% quantile values of the NDVI are taken as NDVI veg and NDVI soil . The calculation parameters are shown in the following table.

[0182] Year <![CDATA[NDVI Min(NDVI soil )]]> <![CDATA[NDVI Max(NDVI veg )]]> 2016 0.152941 0.772549 2019 0.035294 0.796078 2022 -0.160784 0.576471

[0183] Based on the above parameters, the ENVI 5.6 software is used to calculate the vegetation coverage values in 2016, 2019, and 2022, and a classification map is generated, as Figure 5 shown.

[0184] Analysis of vegetation coverage change

[0185] Through the formula ΔFVC = FVC 2 -FVC 1 , the change of vegetation coverage from 2016 to 2022 is calculated. The calculation results are divided into 7 grades according to the change range, including severe degradation, moderate degradation, slight degradation, no change, slight improvement, moderate improvement, and extreme improvement. Attached Figure 6 is the classification distribution map of vegetation coverage change from 2016 to 2022, and the change statistics are shown in the following table.

[0186] Change type Number of pixels Percentage (%) Severe degradation 579267 2.74 Moderate degradation 1064964 5.04 Slight degradation 5463244 25.87 No change 880067 4.17 Slight improvement 12216586 57.85 Moderate improvement 641018 3.04 Extreme improvement 273211 1.29

[0187] The statistical results show that from 2016 to 2022, the positive change (improvement area) of vegetation coverage in the study area accounted for 62.18% of the total area, which is 2 times that of the negative change (degradation area). The improvement areas are mainly concentrated in the northern part of the study area, while the degradation areas are concentrated in the central and southern parts.

[0188] Through the implementation of the Sanjiangyuan Ecological Protection Project, this embodiment verifies the overall improvement trend of vegetation coverage in the Yellow River source area from 2016 to 2022. At the same time, it is also found that there is still vegetation degradation in some areas. The research results show that this method can effectively monitor the dynamic changes of vegetation in the Yellow River source area and provide scientific support for regional ecological environment protection.

[0189] Although embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for analyzing the spatiotemporal variation of vegetation coverage, characterized in that: The following steps are involved: Acquire remote sensing image data of the target area, pre-process the remote sensing image to eliminate the influence of atmosphere and radiation, and extract image data within the target area; Calculate the vegetation index value of the pixels in the target area based on remote sensing image data to reflect the vegetation distribution; The vegetation coverage of the target area is estimated by using the pixel binary model through the vegetation index value; Create a vegetation coverage classification map based on vegetation coverage data and classify vegetation coverage; By calculating the difference in vegetation coverage at different periods, the spatiotemporal variation characteristics of vegetation coverage in the target area are analyzed.

2. The method for analyzing the spatiotemporal variation of vegetation coverage according to claim 1, characterized in that: The step of acquiring remote sensing image data of the target area comprises: Use the remote sensing image acquisition platform to obtain multispectral remote sensing images of the target area; Filter image data with cloud cover less than the set value; The acquired images are georeferenced and cropped to extract image data of the target area.

3. The method for analyzing spatiotemporal changes of vegetation coverage according to claim 1, characterized in that: The step of preprocessing the remote sensing image comprises: Perform radiometric correction on image data to eliminate errors caused by sensors and environmental factors during image acquisition; Perform atmospheric correction and use the atmospheric correction model to remove the interference of atmospheric scattering and absorption on the image spectral information; Multiple remote sensing images of the target area are stitched together to form remote sensing data that completely covers the target area.

4. The method for analyzing spatiotemporal changes of vegetation coverage according to claim 1, characterized in that: The step of calculating the vegetation index value of the pixels in the target area based on the remote sensing image data comprises: Extract the remote sensing data of the near infrared band and red light band of the target area, where the near infrared band is the 5th band of the Landsat8 image, and the red light band is the 4th band of the Landsat8 image; The NDVI value of each pixel in the target area is calculated using the NDVI formula; The calculated NDVI values ​​were processed for outliers, and the minimum and maximum values ​​of NDVI were determined by statistical cumulative probability distribution, and the NDVI values ​​with cumulative probabilities of 5% and 95% were taken as the upper and lower limits of the outliers, respectively; The NDVI value range is limited to [-1, 1], and outliers exceeding this range are removed to reduce the impact of noise points on subsequent vegetation coverage calculations.

5. The method for analyzing spatiotemporal changes of vegetation coverage according to claim 1, characterized in that: The step of estimating the vegetation coverage of the target area by using the pixel binary model through the vegetation index value comprises: It is assumed that the surface within the target area consists of a vegetation-covered portion and a bare soil-covered portion; The vegetation coverage is estimated using the pixel binary model formula, which is as follows: Among them, FVC represents the vegetation coverage of the pixel, NDVI soil is the NDVI value of the bare soil cover part, NDVI veg is the NDVI value of the vegetation coverage area.

6. The method for analyzing spatiotemporal changes of vegetation coverage according to claim 5, characterized in that: In the process of estimating vegetation coverage, NDVI soil and NDVI veg The value of is determined as follows: The cumulative probability distribution of NDVI values ​​in the target area is calculated, and the NDVI value with a cumulative probability of 5% is taken as the NDVI soil The lower limit of the cumulative probability is 95%, and the NDVI value is taken as the NDVI veg The upper limit of A large sample pixel statistical method was used to eliminate abnormal NDVI values.

7. The method for analyzing spatiotemporal changes of vegetation coverage according to claim 1, characterized in that: The step of making a vegetation coverage classification map based on vegetation coverage data comprises: The range of vegetation coverage is divided into multiple levels, and the classification standards include very low vegetation coverage, low vegetation coverage, medium vegetation coverage, medium-high vegetation coverage and high vegetation coverage; Vegetation cover is classified using the following classification scale: Very low vegetation cover: 0% to 10%; Low vegetation cover: 10% to 30%; Medium vegetation coverage: 30% to 50%; Medium to high vegetation coverage: 50% to 70%; High vegetation cover: 70% to 100%; GIS software was used to visualize the classification results and generate a classification distribution map.

8. The method for analyzing spatiotemporal changes of vegetation coverage according to claim 7, characterized in that: The vegetation coverage classification also includes the following steps: The classification results are statistically analyzed to calculate the area of ​​each vegetation coverage level and its proportion to the total area of ​​the target area; The map algebra method is used to generate a vegetation coverage level distribution map, and the vegetation coverage level of the target area is marked in pixels; GIS tools were used to output the spatial distribution map of vegetation coverage levels, and the distribution results were overlaid and analyzed to identify the spatial distribution characteristics of changes in vegetation coverage levels.

9. The method for analyzing spatiotemporal changes of vegetation coverage according to claim 1, characterized in that: The step of analyzing the spatiotemporal variation characteristics of vegetation coverage in the target area by calculating the difference of vegetation coverage data comprises: The difference method is used to calculate the change value of vegetation coverage in different periods. The specific formula is: ΔFVC=FVC2-FVC1 Among them, ΔFVC is the change value of vegetation coverage, FVC2 and FVC1 are the vegetation coverage in the later and earlier periods, respectively; According to the value of vegetation cover change, the changed areas are divided into the following levels: severe degradation, moderate degradation, slight degradation, no change, slight improvement, moderate improvement and extreme improvement.

10. A device for analyzing the spatiotemporal variation of vegetation coverage, used to execute the method for analyzing the spatiotemporal variation of vegetation coverage as claimed in any one of claims 1 to 9, characterized in that: include: A data acquisition module is used to acquire remote sensing image data of the target area and pre-process the image data; Vegetation index calculation module, used to calculate the vegetation index value of the target area based on remote sensing images; A vegetation coverage estimation module is used to estimate the vegetation coverage of a target area based on a vegetation index value; A classification module, used for generating a vegetation coverage classification map based on vegetation coverage data; The change analysis module is used to analyze the spatiotemporal change characteristics of vegetation coverage in the target area by calculating the difference in vegetation coverage in different periods.

Images