A method for estimating ecological carrying capacity based on multi-source remote sensing data

Abstract

The present invention relates to the field of ecological technology, and in particular to an ecological carrying capacity estimation method based on multi-source remote sensing data. The method comprises the following steps: acquiring monitoring area sensor data and monitoring area remote sensing data and dividing them, obtaining natural area remote sensing images and development area remote sensing images; associating natural area remote sensing images and development area remote sensing images, obtaining natural area sensor data and development area sensor data; evaluating natural area remote sensing images, obtaining natural area ecological factor data; evaluating development area remote sensing images, obtaining cultivated land area ecological factor data; evaluating according to monitoring area sensor data, obtaining transition area ecological factor data, and merging them to obtain regional ecological factor data; estimating based on regional ecological factor data and high-resolution remote sensing images, obtaining regional ecological carrying capacity estimation data. The present invention can more accurately evaluate the carrying capacity of an ecosystem.

Description

Technical Field

[0001] The present invention relates to the field of ecological technology, and in particular to an ecological carrying capacity estimation method based on multi-source remote sensing data. Background Art

[0002] With the continuous development of human economic activities and the accelerated advancement of urbanization, the ecological environment has been increasingly threatened, and the stability and health of the ecosystem have become the focus of social attention. In this context, it is particularly important to scientifically evaluate and monitor the carrying capacity of the ecosystem, which is conducive to the formulation of reasonable ecological protection and resource utilization strategies. In the field of ecology, ecological carrying capacity refers to the ability of an ecosystem to accommodate and maintain a certain number and quality of biological populations or individuals without destroying its own structure and function. Traditional ecological carrying capacity assessment methods often rely on limited observation data, such as biological sample surveys and fixed-point observations. These data may not be sufficient to cover the entire study area, and it is difficult to fully and accurately reflect the characteristics and changes of the ecosystem. At the same time, traditional ecological carrying capacity assessment methods can usually only be used for local area assessments, and it is difficult to achieve comprehensive monitoring and assessment of the entire ecosystem. In particular, for cross-regional ecosystems, the applicability of traditional methods is limited. Summary of the invention

[0003] Based on this, it is necessary for the present invention to provide an ecological carrying capacity estimation method based on multi-source remote sensing data to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a method for estimating ecological carrying capacity based on multi-source remote sensing data includes the following steps:

[0005] Step S1: acquiring the monitoring area sensor data and the monitoring area remote sensing data, and performing multi-dimensional remote sensing data fusion based on the monitoring area remote sensing data, so as to obtain a high-resolution remote sensing image; performing surface resource division on the high-resolution remote sensing image, so as to obtain a natural area remote sensing image and a development area remote sensing image;

[0006] Step S2: spatially correlating the natural area remote sensing image and the development area remote sensing image according to the monitoring area sensor data, thereby obtaining the natural area sensor data and the development area sensor data;

[0007] Step S3: classifying the land cover types of the natural area remote sensing images, thereby obtaining regional land cover type classification spatial data; evaluating natural ecological factors based on the natural area sensing data and the regional land cover type classification spatial data, thereby obtaining natural area ecological factor data;

[0008] Step S4: performing remote sensing image classification of cultivated land types on the remote sensing image of the development area, thereby obtaining a remote sensing image of the cultivated land area; performing cultivated land ecological factor evaluation on the remote sensing image of the cultivated land area according to the sensor data of the development area, thereby obtaining ecological factor data of the cultivated land area;

[0009] Step S5: evaluating the transitional area ecological factors of the natural area remote sensing images and the cultivated area remote sensing images according to the monitoring area sensing data, thereby obtaining the transitional area ecological factor data; spatially merging the transitional area ecological factor data, the natural area ecological factor data and the cultivated area ecological factor data, thereby obtaining the regional ecological factor data;

[0010] Step S6: construct an ecological carrying capacity estimation model based on regional ecological factor data and high-resolution remote sensing images, and estimate the ecological carrying capacity of the regional ecological factor data and high-resolution remote sensing images according to the ecological carrying capacity estimation model, so as to obtain regional ecological carrying capacity estimation data.

[0011] The present invention obtains sensor data including environmental parameters such as soil moisture, temperature, and gas concentration. These data provide real-time environmental information and reflect the current state of the ecosystem. Remote sensing data can provide a wide range of surface information, including vegetation coverage, land use, water distribution, etc. By acquiring remote sensing data, the overall ecological situation of the monitored area can be fully understood. Fusion of remote sensing data from different sources and types can eliminate differences and inconsistencies between data and improve the accuracy and credibility of data. Through fusion processing, high-resolution remote sensing images are obtained, and the surface features of the monitored area can be observed more clearly, providing a more detailed data basis for subsequent analysis. Spatial association of sensor data with remote sensing images ensures spatial consistency and interrelationship of data. This helps to integrate information from different data sources into the same geographic space for analysis. By classifying the surface cover types of remote sensing images of natural areas, different ecological environment types, such as forests, grasslands, wetlands, etc., can be identified. Combined with sensor data and surface cover type classification results, ecological factors of natural areas are evaluated, including vegetation coverage, soil texture, water resources, etc., so as to understand the health status of natural ecosystems. By classifying the cultivated land types in remote sensing images of the development area, different types of cultivated land can be identified, such as rice fields, orchards, and farmlands. Combining the sensor data and the cultivated land classification results, the ecological factors of the cultivated land area are evaluated, including soil quality, pesticide use, crop health, etc., to assess the impact of human activities on the ecosystem. The ecological factors of the transition area between the natural area and the cultivated land area are evaluated, taking into account the special environmental characteristics of the area, such as edge effects and species migration. The ecological factor data of each region are spatially merged to obtain comprehensive and holistic regional ecological factor data, providing a data basis for comprehensive evaluation. Based on the regional ecological factor data and high-resolution remote sensing images, an ecological carrying capacity estimation model is constructed, taking into account the impact of different factors on the ecosystem and their interrelationships. Using the constructed ecological carrying capacity estimation model, the ecological carrying capacity of the regional ecological factor data and high-resolution remote sensing images is estimated, and the number and quality of biological populations or individuals that the ecosystem can withstand in the current state are obtained, so as to more accurately assess the carrying capacity of the ecosystem. At the same time, it provides a scientific basis for formulating reasonable protection and utilization strategies. By integrating sensor data, remote sensing data and high-resolution images, and applying ecological factor assessment and ecological carrying capacity estimation models, the present invention can more comprehensively and accurately assess the status and carrying capacity of ecosystems, thereby providing scientific support for ecological protection and resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0013] Figure 1 A schematic diagram of the steps of the method for estimating ecological carrying capacity based on multi-source remote sensing data of the present invention;

[0014] Figure 2 Detailed step flow diagram of step S1 in the present invention;

[0015] Figure 3 Detailed step flow diagram of step S15 in the present invention;

[0016] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0017] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are 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 technicians in this field without creative work are within the scope of protection of the present invention.

[0018] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0019] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0020] To achieve this, please refer to Figures 1 to 3 The present invention provides an ecological carrying capacity estimation method based on multi-source remote sensing data, the method comprising the following steps:

[0021] Step S1: acquiring the monitoring area sensor data and the monitoring area remote sensing data, and performing multi-dimensional remote sensing data fusion based on the monitoring area remote sensing data, so as to obtain a high-resolution remote sensing image; performing surface resource division on the high-resolution remote sensing image, so as to obtain a natural area remote sensing image and a development area remote sensing image;

[0022] In this embodiment, sensor data of the monitoring area, such as data from soil monitoring stations, meteorological stations, water quality monitoring stations, etc., and remote sensing data of the monitoring area, such as satellite images, are obtained through the monitoring organization platform. Multispectral and hyperspectral remote sensing data fusion technology is used to combine vegetation index, terrain and other information to achieve the generation of high-resolution remote sensing images. Pixel-level classification methods are used to divide surface resources of high-resolution remote sensing images, such as using machine learning algorithms such as support vector machines (SVM) to classify land use / cover, so as to obtain remote sensing images of natural areas and development areas.

[0023] Step S2: spatially correlating the natural area remote sensing image and the development area remote sensing image according to the monitoring area sensor data, thereby obtaining the natural area sensor data and the development area sensor data;

[0024] In this embodiment, geographic information system (GIS) technology is used to spatially associate the sensor data of the monitored area with the remote sensing images of the natural area and the developed area. For example, the sensor data of the natural area and the developed area, such as temperature and humidity, are obtained through pixel-level spatial matching and interpolation methods.

[0025] Step S3: classifying the land cover types of the natural area remote sensing images, thereby obtaining regional land cover type classification spatial data; evaluating natural ecological factors based on the natural area sensing data and the regional land cover type classification spatial data, thereby obtaining natural area ecological factor data;

[0026] In this embodiment, the land cover type classification is performed on the remote sensing images of the natural area, and the pixel-level classification method, such as the maximum likelihood classification algorithm, is used to classify the land cover into forest, grassland, water area and other categories, and obtain the regional land cover type classification spatial data. Then, based on the sensing data of the natural area and the land cover type classification spatial data, expert knowledge and model algorithms (such as decision tree algorithms) are used to evaluate natural ecological factors, such as biodiversity, vegetation coverage, etc.

[0027] Step S4: performing remote sensing image classification of cultivated land types on the remote sensing image of the development area, thereby obtaining a remote sensing image of the cultivated land area; performing cultivated land ecological factor evaluation on the remote sensing image of the cultivated land area according to the sensor data of the development area, thereby obtaining ecological factor data of the cultivated land area;

[0028] In this embodiment, the remote sensing image of the development area is divided into cultivated land types by using an object-oriented method and object-level classification technology, such as a rule-based classification algorithm, to divide the cultivated land into types such as rice fields and corn fields to obtain a remote sensing image of the cultivated land area. Then, based on the sensor data of the development area, such as soil moisture content, soil type, etc., the cultivated land ecological factors of the cultivated land area are evaluated, such as soil quality, crop yield, etc.

[0029] Step S5: evaluating the transitional area ecological factors of the natural area remote sensing images and the cultivated area remote sensing images according to the monitoring area sensing data, thereby obtaining the transitional area ecological factor data; spatially merging the transitional area ecological factor data, the natural area ecological factor data and the cultivated area ecological factor data, thereby obtaining the regional ecological factor data;

[0030] In this embodiment, spatial analysis technology is used to extract the transition area of ​​remote sensing images of natural areas and cultivated land areas, and pixel-level pixel mixing technology is used to obtain remote sensing images of transition areas. Then, based on the sensing data of the monitoring area, ecological factors of the transition area are evaluated, such as ecological stability, impact of human activities, etc. Finally, the ecological factor data of the transition area is spatially merged with the ecological factor data of the natural area and cultivated land area to generate ecological factor data of the entire area.

[0031] Step S6: construct an ecological carrying capacity estimation model based on regional ecological factor data and high-resolution remote sensing images, and estimate the ecological carrying capacity of the regional ecological factor data and high-resolution remote sensing images according to the ecological carrying capacity estimation model, so as to obtain regional ecological carrying capacity estimation data.

[0032] In this embodiment, an ecological model, such as an ecosystem service model or an ecological carrying capacity model, is used to combine regional ecological factor data and high-resolution remote sensing images to construct an ecological carrying capacity estimation model. Through model calculation and analysis, the ecological carrying capacity of the region is evaluated, including a comprehensive evaluation of natural resource utilization, ecosystem functions, etc., and ultimately regional ecological carrying capacity estimation data is obtained.

[0033] The present invention obtains sensor data including environmental parameters such as soil moisture, temperature, and gas concentration. These data provide real-time environmental information and reflect the current state of the ecosystem. Remote sensing data can provide a wide range of surface information, including vegetation coverage, land use, water distribution, etc. By acquiring remote sensing data, the overall ecological situation of the monitored area can be fully understood. Fusion of remote sensing data from different sources and types can eliminate differences and inconsistencies between data and improve the accuracy and credibility of data. Through fusion processing, high-resolution remote sensing images are obtained, and the surface features of the monitored area can be observed more clearly, providing a more detailed data basis for subsequent analysis. Spatial association of sensor data with remote sensing images ensures spatial consistency and interrelationship of data. This helps to integrate information from different data sources into the same geographic space for analysis. By classifying the surface cover types of remote sensing images of natural areas, different ecological environment types, such as forests, grasslands, wetlands, etc., can be identified. Combined with sensor data and surface cover type classification results, ecological factors of natural areas are evaluated, including vegetation coverage, soil texture, water resources, etc., so as to understand the health status of natural ecosystems. By classifying the cultivated land types in remote sensing images of the development area, different types of cultivated land can be identified, such as rice fields, orchards, and farmlands. Combining the sensor data and the cultivated land classification results, the ecological factors of the cultivated land area are evaluated, including soil quality, pesticide use, crop health, etc., to assess the impact of human activities on the ecosystem. The ecological factors of the transition area between the natural area and the cultivated land area are evaluated, taking into account the special environmental characteristics of the area, such as edge effects and species migration. The ecological factor data of each region are spatially merged to obtain comprehensive and holistic regional ecological factor data, providing a data basis for comprehensive evaluation. Based on the regional ecological factor data and high-resolution remote sensing images, an ecological carrying capacity estimation model is constructed, taking into account the impact of different factors on the ecosystem and their interrelationships. Using the constructed ecological carrying capacity estimation model, the ecological carrying capacity of the regional ecological factor data and high-resolution remote sensing images is estimated, and the number and quality of biological populations or individuals that the ecosystem can withstand in the current state are obtained, providing a scientific basis for formulating reasonable protection and utilization strategies.

[0034] Optionally, step S1 specifically includes:

[0035] Step S11: Acquire monitoring area sensor data and monitoring area remote sensing data;

[0036] In this embodiment, in order to obtain the sensor data and remote sensing data of the monitoring area, sensor equipment, such as weather stations and water quality monitoring stations, are first deployed on the ground to collect weather, water quality and other data; at the same time, satellites or drones are used to obtain high-resolution remote sensing images to cover the monitoring area and ensure the comprehensiveness and accuracy of data collection.

[0037] Step S12: performing atmospheric correction and radiation correction on the remote sensing data of the monitoring area, thereby obtaining pre-processed remote sensing data;

[0038] In this embodiment, atmospheric correction is performed on the acquired remote sensing data of the monitoring area to eliminate the influence of atmospheric factors on the image, and then radiation correction is performed to correct the difference in lighting conditions to improve the accuracy and consistency of the data, thereby obtaining pre-processed remote sensing data and ensuring the reliability of subsequent analysis.

[0039] Step S13: extracting terrain features from the pre-processed remote sensing data to obtain regional terrain feature data;

[0040] In this embodiment, pre-processed remote sensing data is used to extract terrain features, and digital elevation model (DEM) and other technologies are used to extract terrain information of the monitoring area, such as terrain undulation, slope, etc., to obtain regional terrain feature data and provide basic data support for subsequent analysis.

[0041] Step S14: constructing a three-dimensional image of the regional terrain according to the regional terrain feature data, and performing remote sensing data space mapping on the three-dimensional image of the regional terrain according to the pre-processed remote sensing data, thereby obtaining a high-resolution remote sensing image;

[0042] In this embodiment, a three-dimensional image of the regional terrain is constructed based on the regional terrain feature data, a terrain model is generated using DEM data, and spatial mapping of the terrain and image is performed in combination with pre-processed remote sensing data to ensure the consistency of the remote sensing image with the actual terrain, thereby obtaining high-resolution remote sensing images and providing accurate spatial information for the division of surface resources.

[0043] Step S15: Divide the high-resolution remote sensing image into surface resources, thereby obtaining remote sensing images of natural areas and remote sensing images of development areas.

[0044] In this embodiment, high-resolution remote sensing images are used to divide surface resources, and image classification and target extraction techniques are used to divide the monitored area into natural areas and development areas, such as forests, waters, cities, etc., to obtain remote sensing images of natural areas and remote sensing images of development areas, providing basic data for subsequent ecological factor assessments.

[0045] The sensor data obtained by the present invention provides real-time environmental parameters, such as soil moisture, temperature, etc., to help understand the current state of the ecosystem. Remote sensing data provides comprehensive surface information, such as vegetation coverage, water distribution, etc., providing a data basis for a comprehensive assessment of the ecosystem. These corrections help eliminate errors caused by atmospheric interference and surface reflection characteristics in remote sensing images, and improve the accuracy and credibility of the data. Through terrain feature extraction, the terrain information of the monitoring area, such as undulating terrain, high and low terrain, etc., can be obtained, providing a basis for subsequent terrain analysis and model construction. By constructing a three-dimensional terrain image, the terrain features of the monitoring area can be intuitively displayed, which helps to more deeply understand the impact of terrain on the ecosystem. Spatial mapping of remote sensing data with three-dimensional terrain images helps to associate remote sensing information with terrain features, providing more spatial information for subsequent remote sensing image analysis. By dividing the surface resources of high-resolution remote sensing images, the surface coverage types of natural areas and development areas can be identified, providing basic data for the analysis and evaluation of the ecological environment.

[0046] Optionally, step S15 is specifically:

[0047] Step S151: performing pixel statistics on the high-resolution remote sensing image to obtain pixel data of the remote sensing image;

[0048] In this embodiment, for high-resolution remote sensing images, pixel statistics are first performed, that is, the numerical information of each pixel in the image is counted, such as color value, brightness value, etc. These attributes are usually expressed in digital form, such as red, green, and blue component values ​​in RGB color mode. By counting each pixel attribute, the pixel data of the image can be obtained to prepare for subsequent analysis.

[0049] Step S152: performing adjacent pixel difference calculation on the remote sensing image pixel data, thereby obtaining adjacent pixel difference data;

[0050] In this embodiment, after obtaining the pixel data of the remote sensing image, the difference between adjacent pixels is calculated. When calculating the difference between adjacent pixels, it is necessary to consider the relationship between each pixel in the image and its surrounding adjacent pixels. This involves calculating the difference between pixels, for example, by calculating the difference between gray levels and color values. It usually includes traversing the neighborhood around each pixel and calculating the difference with the current pixel. Through this process, the difference data between adjacent pixels can be generated for subsequent processing.

[0051] Step S153: Connecting high-value pixel difference pixels of the high-resolution remote sensing image according to adjacent pixel difference data, thereby obtaining a high-value pixel difference pixel point connection image;

[0052] In this embodiment, when connecting high-value pixel difference pixels for high-resolution remote sensing images, it is necessary to connect the pixels with large adjacent pixel differences to form continuous edges or feature lines. This process usually includes threshold setting and pixel connection algorithm. First, it is determined which pixels have a difference that is considered "high" based on a pre-set threshold. Then, a pixel connection algorithm, such as a connectivity-based algorithm or an edge tracking algorithm, is used to connect these pixels to form feature lines or edges. Finally, a high-value pixel difference pixel connection image is obtained.

[0053] Step S154: dividing the high-value pixel difference pixel point connection image into natural area remote sensing images, thereby obtaining natural area remote sensing images; dividing the high-value pixel difference pixel point connection image into natural area remote sensing images, thereby obtaining development area remote sensing images.

[0054] In this embodiment, when dividing the remote sensing images of natural areas and developed areas for the high-value pixel difference pixel point connection image, an image segmentation or region segmentation method can be used. This includes a clustering algorithm based on pixel attributes or an edge-based segmentation algorithm. First, the image pixels are divided into different regions or categories according to specific image attributes, such as color, texture or shape. Then, according to the connectivity or edge information between adjacent pixels, these regions are further subdivided or merged to obtain remote sensing images of natural areas and developed areas. This provides image data with more semantic information for subsequent image analysis and application.

[0055] The present invention can obtain the numerical information of each pixel in the remote sensing image, including brightness, color, etc., through pixel statistics, providing basic data for subsequent analysis. By calculating the difference between adjacent pixels, subtle changes and edge features in the image can be revealed, which helps to extract the target object and boundary in the image. By connecting the difference data between adjacent pixels, a high-pixel difference pixel point connection image can be formed, which helps to identify local features and target objects in the image. By dividing the high-pixel difference pixel point connection image, the image can be divided into natural areas and development areas, which helps to classify and identify the surface cover type, and then analyze and evaluate the ecological environment.

[0056] Optionally, step S154 is specifically:

[0057] The high pixel difference pixel point connection image is divided into connected closed area images, so as to obtain a connected closed area image set;

[0058] In this embodiment, when dividing the connected closed region image for the high pixel difference pixel point connection image, a connectivity-based region growing algorithm is used. First, a seed pixel is selected as the starting point, and then the region is gradually expanded through the connectivity of the pixel neighborhood until it can no longer be connected, forming a closed region. This process is repeated until the entire image is traversed to obtain a connected closed region image set.

[0059] Performing closed area edge point detection on the connected closed area image set, thereby obtaining closed area edge point data;

[0060] In this embodiment, the closed region edge point detection is performed on the connected closed region image set, and the Canny edge detection algorithm is used. The algorithm uses the gradient information and non-maximum suppression in the image to detect edge points, and then connects the edges to form closed edges. In this way, the closed region edge point data can be obtained.

[0061] Calculate the angles of adjacent edge points on the edge point data of the closed area to obtain the angle data of adjacent edge points;

[0062] In this embodiment, the angles of adjacent edge points are calculated for the closed area edge point data, and the gradient direction of the edge points is calculated using the Sobel operator, and then the angles between adjacent edge points are calculated, so that the adjacent edge point angle data can be obtained.

[0063] Performing regular angle threshold evaluation on the angle data of adjacent edge points, thereby obtaining a regular angle threshold;

[0064] In this embodiment, regular angle thresholds are evaluated for the angle data of adjacent edge points, and a set of regular angle ranges is set, such as 0° to 15° for horizontal edges, 45° to 75° for vertical edges, etc. By comparing with these regular angle ranges, edge points that meet the rules are screened out.

[0065] The connected closed area image set is divided into closed areas according to the regular angle threshold, so as to obtain the natural area image set and the development area image set;

[0066] In this embodiment, the connected closed area image set is divided into closed areas according to the regular angle threshold, and the connected closed area images with angles that meet the rules are divided into the development area image set. On the contrary, the connected closed area images with angles that do not meet the rules are divided into the natural area image set. In this way, the natural area image set and the development area image set are obtained.

[0067] The natural area image set is spatially merged to obtain the natural area remote sensing image; the development area image set is spatially merged to obtain the development area remote sensing image.

[0068] In this embodiment, the natural area image set is spatially merged, and a region growing or segmentation merge algorithm can be used to merge adjacent natural areas with similar characteristics into a larger area, thereby obtaining a natural area remote sensing image. Similarly, the development area image set is spatially merged to obtain a development area remote sensing image.

[0069] The present invention can segment the connected areas in the image by connecting the closed area image division to form different regional image sets, which provides a basis for subsequent analysis and processing. Edge point detection can accurately find the boundary of the closed area, help determine the shape and size of the area, and provide boundary data for subsequent angle calculation and analysis. By calculating the angle between adjacent edge points, the curve characteristics and morphology of the closed area can be understood, and then used to judge the attributes and shape of the area. Evaluating the regularity and threshold of the angles of adjacent edge points can help determine the characteristics and types of the closed area, and further classify and identify the area. According to the regular angle threshold, the connected closed area image set is divided, and the image is divided into a natural area and a development area, which provides a basis for subsequent spatial merging. The natural area image set and the development area image set are spatially merged, and the adjacent areas can be integrated to obtain a larger range of natural area remote sensing images and development area remote sensing images, which helps to more comprehensively understand and analyze surface features and environment.

[0070] Optionally, the closed area division in step S154 is specifically as follows:

[0071] The angle data of adjacent edge points are classified and calculated according to the regular angle threshold, so as to obtain the regular edge point angle data and the irregular edge point angle data;

[0072] In this embodiment, the angle data of adjacent edge points are classified and calculated according to the regular angle threshold. For example, the horizontal edge angle range is set to 0° to 15°, and the vertical edge angle range is set to 75° to 90°. Angles outside these ranges are considered irregular. In this way, edge points can be divided into regular edge points and irregular edge points.

[0073] Performing statistical analysis on the closed area edge point data according to the regular edge point angle data, thereby obtaining high-frequency regular edge point area data and low-frequency regular edge point area data;

[0074] In this embodiment, the closed area edge point data is statistically analyzed based on the regular edge point angle data. For example, the frequency of occurrence of regular edge points in each closed area is counted. Regular edge points that appear frequently may represent obvious boundaries or structures, while regular edge points that appear infrequently may be noise or secondary features.

[0075] Performing statistical analysis on the closed area edge point data according to the irregular edge point angle data, thereby obtaining high-frequency irregular edge point area data and low-frequency irregular edge point area data;

[0076] In this embodiment, the closed area edge point data is statistically analyzed based on the irregular edge point angle data. For example, the frequency of occurrence of irregular edge points in each closed area is counted. Irregular edge points that appear frequently represent textured or cluttered areas, while irregular edge points that appear infrequently may be outliers or interference.

[0077] Perform intersection operation on high-frequency regular edge point area data and low-frequency irregular edge point area data to obtain development area data;

[0078] In this embodiment, the high-frequency regular edge point area data and the low-frequency irregular edge point area data are subjected to intersection operations to obtain the development area data. By performing intersection operations on the high-frequency regular edge point area and the low-frequency irregular edge point area, the development area data with obvious structure and boundaries can be obtained.

[0079] Perform intersection operation on low-frequency regular edge point area data and high-frequency irregular edge point area data to obtain natural area data;

[0080] In this embodiment, the intersection operation is performed on the low-frequency regular edge point region data and the high-frequency irregular edge point region data to obtain the natural region data. By performing the intersection operation on the low-frequency regular edge point region and the high-frequency irregular edge point region, the natural region data with messy texture or non-structural features can be obtained.

[0081] Dividing the connected closed area image set into regions according to the development area data, thereby obtaining the development area image set;

[0082] In this embodiment, the connected closed area image set is divided into regions according to the development area data, thereby obtaining the development area image set. The development area in the connected closed area image set is divided into the development area image set using the edge data included in the development area data.

[0083] The connected closed region image set is divided into regions according to the natural region data, so as to obtain the natural region image set.

[0084] In this embodiment, the connected closed region image set is divided into regions according to the natural region data, thereby obtaining the natural region image set. The natural regions in the connected closed region image set are divided into the natural region image set using the edge data contained in the natural region data.

[0085] The present invention classifies and calculates the adjacent edge point angle data according to the regular angle threshold, and can separate the regular edge point angle data and the irregular edge point angle data, which provides a basis for the subsequent regional division. The regular edge point angle data is statistically analyzed to obtain the regularity information of the edge point distribution, and then determine the high-frequency regular edge point area and the low-frequency regular edge point area, which is helpful to distinguish different types of objects or landform features. The irregular edge point angle data is statistically analyzed to obtain the distribution of the irregular edge point area, and then determine the high-frequency and low-frequency irregular edge point areas, which is helpful to identify the artificial area or natural area on the surface. By performing intersection operation on the high-frequency regular edge point area data and the low-frequency irregular edge point area data, the development area data and the natural area data can be obtained, and these data will be used for the final regional division. According to the development area data and the natural area data, the connected closed area image set is regionally divided, so as to obtain the development area image set and the natural area image set, which reflect the spatial distribution of the development and natural environment on the surface.

[0086] Optionally, step S3 specifically includes:

[0087] Step S31: extracting pixel spectral features from natural area remote sensing images to obtain natural surface spectral data;

[0088] In this embodiment, by extracting pixel spectral features from remote sensing images, spectral reflectance can be used to describe the spectral features of the surface of a natural area. This can include extracting the reflectance value of each pixel at different wavelengths from the remote sensing image to form a spectral curve that reflects the reflective properties of surface objects at different wavelengths.

[0089] Step S32: performing band reflectance statistics on the natural surface spectral data, thereby obtaining high-value blue band reflectance spectral data and high-value green band reflectance spectral data;

[0090] In this embodiment, the extracted spectral data is subjected to band reflectance statistics to obtain reflectance data of different bands (such as blue band and green band). Such statistics can help determine the spectral reflectance properties of natural surfaces in specific bands, and provide basic data for subsequent classification and analysis.

[0091] Step S33: dividing the natural area remote sensing image into blue-reflection areas according to the high-value blue band reflectance spectral data, thereby obtaining a high-value blue-reflection area image; dividing the natural area remote sensing image into green-reflection areas according to the high-value green band reflectance spectral data, thereby obtaining a high-value green-reflection area image;

[0092] In this embodiment, the reflectance data of the high-value blue band and the high-value green band are used to divide the remote sensing image into blue-reflected and green-reflected areas. Through a threshold or a classification algorithm based on spectral features, the areas reflecting blue and green features in the image can be extracted to form corresponding blue-reflected area images and green-reflected area images.

[0093] Step S34: performing vegetation index calculation and water index calculation according to the natural surface spectral data, thereby obtaining natural surface vegetation index data and natural surface water index data, and performing same index feature extraction on the natural surface spectral data according to the natural surface vegetation index data and the natural surface water index data, thereby obtaining same index natural spectral data;

[0094] In this embodiment, vegetation index (such as NDVI) and water index (such as NDWI) are calculated based on spectral data to obtain vegetation and water information on the natural surface. These indices reflect the spectral characteristics of vegetation and water, which are helpful for further classification and analysis. At the same time, spectral data with the same two indices can be extracted from the spectral data according to these indices to obtain natural spectral data with the same index.

[0095] Step S35: classifying the coverage type of the high-value green area image according to the natural surface vegetation index data, thereby obtaining a forest type image and a grassland type image; classifying the coverage type of the high-value blue area image according to the natural surface water index data, thereby obtaining a lake type image and a river type image;

[0096] In this embodiment, it involves using vegetation index data to classify high-value green areas, and using water index data to classify high-value blue areas. This can be achieved by setting a certain threshold and using a supervised or unsupervised classification algorithm, for example, a support vector machine (SVM), random forest, or K-means clustering method can be used. The classification results will reflect the spatial distribution of different land cover types, such as forests, grasslands, lakes, and rivers.

[0097] Step S36: dividing the natural area remote sensing image into wetland areas according to the natural spectrum data with the same index, thereby obtaining a wetland type image;

[0098] In this embodiment, the remote sensing image is divided into wetland areas according to the natural spectral data of the same index. This can be achieved by using the index data or classification algorithms such as support vector machine (SVM), random forest or K-means clustering based on the spectral characteristics of specific wetlands, so as to obtain wetland type images.

[0099] Step S37: spatially merging the wetland type image, the lake type image, the river type image, the forest type image, and the grassland type image to obtain regional land cover type classification spatial data;

[0100] In this embodiment, the images of different land cover types obtained by division are spatially merged to obtain complete regional land cover type classification spatial data. This can be achieved by image overlay, pixel-level spatial analysis or spatial operation of vector data to form a complete land cover type classification result.

[0101] Step S38: Evaluate natural ecological factors based on natural area sensor data and regional land cover type classification spatial data to obtain natural area ecological factor data.

[0102] In this embodiment, natural ecological factors are evaluated using natural area sensor data and surface cover type classification spatial data. This can be done by setting weights for the characteristics of different surface cover types based on expert experience, etc., to evaluate factors such as vegetation coverage, water resources, soil types, and biodiversity in natural ecosystems. This evaluation can provide a scientific basis for ecological environmental protection and management, and help formulate relevant policies and measures.

[0103] The present invention can obtain the spectral information of each point on the surface by extracting the pixel spectral features of the natural area remote sensing image, and provide basic data for the subsequent spectral data analysis. The band reflectance statistics of the natural surface spectral data can obtain the reflectance information of different bands, and extract the high blue band reflectance and high green band reflectance data from them, which is helpful to distinguish the surface cover type. Based on the high blue band reflectance and high green band reflectance data, the natural area remote sensing image is divided into blue reflection and green reflection areas, and the high blue reflection area image and the high green reflection area image are obtained, which are used to identify surface features such as water bodies and vegetation. Through the calculation of vegetation index and water body index, the vegetation and water body areas can be divided more accurately. The same index feature extraction is used to further strengthen the identification of surface cover types. Based on the vegetation index data and water body index data, the high green reflection and high blue reflection area images are classified into cover types, and different surface type images such as forests, grasslands, lakes and rivers are obtained, which provides classification data for ecological factor assessment. Through the natural spectral data of the same index, the natural area remote sensing images are divided into wetland areas to obtain wetland type images, which is helpful for the monitoring and protection of wetland resources. Different types of surface cover type images are spatially merged to obtain complete regional surface cover type classification spatial data, which provides data support for the overall evaluation of surface types. Combining natural area sensing data and surface cover type classification spatial data, the ecological factors of natural areas are evaluated to provide a scientific basis for the management and protection of the ecological environment.

[0104] Optionally, step S38 is specifically:

[0105] Step S381: extracting soil features and water features from the natural area sensor data, thereby obtaining natural area soil sensor data and natural area water sensor data;

[0106] In this embodiment, for soil feature extraction, the infrared band information in the hyperspectral remote sensing data is first used, combined with the soil spectral reflectance characteristics, and a linear spectral mixture model or a support vector machine algorithm is used to classify and identify the soil type, thereby obtaining natural area soil sensing data. For water body feature extraction, the blue light band reflectance information in the multispectral remote sensing data is combined with the spectral characteristics of the water body, and threshold segmentation or water body index calculation (such as water body index NDWI) is used to identify and extract water bodies, thereby obtaining natural area water body sensing data.

[0107] Step S382: integrating the regional land cover type classification spatial data according to the natural area soil sensor data and the natural area water body sensor data, thereby obtaining the natural area land cover sensor data;

[0108] In this embodiment, soil sensor data and water body sensor data are used in combination with regional surface cover type classification spatial data, and a pixel-level data fusion method, such as pixel-level weighted fusion or rule fusion, is adopted to integrate the soil and water body sensor data with the surface cover classification data to obtain natural area surface cover sensor data.

[0109] Step S383: Calculating the soil type distribution ratio of the natural area soil sensor data, thereby obtaining the natural area soil type distribution ratio data; calculating the water quality distribution ratio of the natural area water body sensor data, thereby obtaining the natural area water quality distribution ratio data;

[0110] In this embodiment, when calculating the soil type distribution ratio of soil sensor data, the soil type classification results at the pixel level can be used to count the number of pixels of each type of soil in the natural area and calculate their proportions. The water quality distribution ratio calculation can be based on indicators reflecting water quality conditions in water sensor data, such as water color and transparency, combined with water classification results, to count the number of water pixels of different water quality levels and their proportions.

[0111] Step S384: Calculate the vegetation coverage rate and the water area ratio of the regional land cover type classification spatial data, so as to obtain the natural area vegetation coverage rate data and the natural area water area ratio data;

[0112] In this embodiment, the vegetation coverage rate calculation can use the vegetation classification results in the surface cover classification data to count the number of vegetation pixels in the natural area and calculate its proportion to the total number of pixels, thereby obtaining the vegetation coverage rate data. The water body area ratio calculation is directly based on the water body classification results, and the proportion of the number of water body pixels to the total number of pixels is counted to obtain the water body area ratio data.

[0113] Step S385: Construct a natural ecological factor evaluation system based on the natural area soil type distribution ratio data, the natural area water quality distribution ratio data, the natural area vegetation coverage rate data and the natural area water area ratio data, and use the natural ecological factor evaluation system to perform natural ecological factor evaluation on the natural area surface cover sensor data, so as to obtain natural area ecological factor data.

[0114] In this embodiment, when constructing a natural ecological factor evaluation system, the weights and evaluation indicators of different ecological factors can be set according to data such as soil type distribution ratio, water quality distribution ratio, vegetation coverage rate and water area ratio, and according to relevant information, such as soil fertility, water health, biodiversity, etc. By comprehensively evaluating the impact of various factors on the natural regional ecosystem, the ecological factor data of the natural region is obtained, providing a scientific basis for ecological environment management and protection.

[0115] The present invention can obtain relevant data of soil and water body by extracting soil characteristics and water body characteristics from natural area sensor data, and provide basic data for subsequent soil type and water quality analysis. The soil sensor data and water body sensor data are integrated with the surface cover type classification spatial data to obtain the natural area surface cover sensor data, which further enriches the information of the surface cover type data. By analyzing the soil sensor data and water body sensor data, the distribution ratio of soil type and water quality in the natural area is calculated, which helps to understand the spatial distribution of soil type and water quality. The vegetation coverage rate and water area ratio of the surface cover type classification spatial data are calculated to obtain the vegetation coverage rate and water area ratio data of the natural area, which provides important parameters for the evaluation of the ecological environment. Based on the soil type distribution ratio, water quality distribution ratio, vegetation coverage rate and water area ratio data, a natural ecological factor evaluation system is constructed, and the surface cover sensor data is evaluated by the system to obtain the natural area ecological factor data, which provides a scientific basis for the comprehensive evaluation of the ecological environment.

[0116] Optionally, step S4 is specifically:

[0117] Step S41: performing a topological structure analysis on the remote sensing image of the development area, thereby obtaining the topological structure of the development area;

[0118] In this embodiment, the remote sensing image of the development area is subjected to topological structure analysis. First, an image segmentation algorithm (such as a method based on region growing) is used to segment the image into regional units with similar features. Then, an edge detection algorithm (such as Canny edge detection) is used to identify topological structural elements such as roads and buildings, and a linear feature extraction method (such as Hough transform) is used to extract features such as the center line of the road, and finally a topological structure map of the development area is established.

[0119] Step S42: evaluating the road density of the development area topology structure, thereby obtaining road density data;

[0120] In this embodiment, for the topological structure diagram, a road network extraction algorithm (such as a method based on graph theory) is used to calculate the length, density and distribution of the road network. Image analysis technology combined with mathematical morphological operations can be used to extract and analyze roads to quantify the density of roads.

[0121] Step S43: dividing the topological structure of the development area into low road density areas according to the road density data, thereby obtaining low road density area data;

[0122] In this embodiment, based on the road density data, a threshold segmentation or cluster analysis method is used to divide the development area into low road density areas. A threshold can be set to divide the area with a road density lower than the threshold into a low density area, thereby determining the low road density area.

[0123] Step S44: dividing the remote sensing image of the development area into regions according to the low road density area data, thereby obtaining a remote sensing image of the low road density area;

[0124] In this embodiment, the low road density area data is used to divide the development area remote sensing image using a pixel-level classification algorithm (such as a support vector machine classifier) ​​to obtain a low road density area remote sensing image. The pixel-level classification method can be used to classify each pixel in the remote sensing image as belonging to a low-density road area or other area.

[0125] Step S45: evaluating the terrain flatness of the remote sensing image of the low road density area to obtain regional terrain flatness data, and performing high flatness area distribution analysis based on the regional terrain flatness data to obtain high flatness area data;

[0126] In this embodiment, terrain flatness is evaluated for remote sensing images of low road density areas using terrain data (such as elevation data). Elevation data can be used to calculate indicators such as slope and aspect of the terrain, and the flatness of the terrain can be evaluated by analyzing these indicators, and statistical methods can be used to perform distribution analysis on high flatness areas.

[0127] Step S46: extracting pixel spectral features from the remote sensing image of the low road density area to obtain spectral data of the low road density area, and calculating vegetation index and water index on the spectral data of the low road density area to obtain vegetation index data of the low road density area and water index data of the low road density area;

[0128] In this embodiment, pixel-level spectral feature extraction is performed using remote sensing images, and a pixel-level spectral classification algorithm (such as pixel-based spectral feature extraction and classification) is used to extract spectral information of low road density areas. Then, vegetation index (such as NDVI) and water index (such as NDWI) are calculated based on the spectral information to obtain the distribution of vegetation and water bodies in low road density areas.

[0129] Step S47: performing a high vegetation index regional distribution analysis on the remote sensing image of the low road density area according to the low road density area vegetation index data, thereby obtaining high vegetation index area data; performing a high water index regional distribution analysis on the remote sensing image of the low road density area according to the low road density area water index data, thereby obtaining high water index area data;

[0130] In this embodiment, based on the vegetation index and water index data, a spatial analysis method is used to perform distribution analysis of high vegetation index areas and high water index areas on remote sensing images of low road density areas. Spatial statistical analysis methods, such as kernel density estimation, can be used to determine the distribution of high vegetation index areas and high water index areas, that is, high vegetation index area data and high water index area data.

[0131] Step S48: performing intersection operation on the high vegetation index area data, the high water body index area data and the high flatness area data, thereby obtaining the cultivated land area data, and performing cultivated land type remote sensing image classification on the development area remote sensing image according to the cultivated land area data, thereby obtaining the cultivated land area remote sensing image;

[0132] In this embodiment, the intersection operation is performed on the high vegetation index area, the high water index area and the high flatness area to identify the cultivated land area. Spatial analysis tools, such as GIS software, can be used to perform intersection operations on these areas, and the remote sensing image of the development area can be divided according to the results to obtain the remote sensing image of the cultivated land area.

[0133] Step S49: Evaluate the ecological factors of cultivated land on the remote sensing image of the cultivated land area according to the sensor data of the development area, so as to obtain the ecological factor data of the cultivated land area.

[0134] In this embodiment, the remote sensing image of the cultivated land area is used to evaluate the cultivated land ecological factors using the sensor data of the development area, such as soil moisture content, temperature, etc. Statistical analysis methods, such as linear regression analysis, can be used to evaluate the relationship between the cultivated land ecological factors and the sensor data, and generate ecological factor data of the cultivated land area.

[0135] The present invention can understand the spatial structure of the development area, including the connection relationship and layout between various objects, by performing topological structure analysis on remote sensing images, and provide a basis for subsequent road density assessment and regional division. Assessing road density can help assess the traffic conditions and urbanization level of the development area, so as to better understand the development level and urbanization level. Dividing the development area into low-density areas according to road density helps to identify relatively sparse areas in the city, and provides a basis for subsequent terrain and vegetation analysis. Dividing the images of low road density areas helps to more accurately analyze the terrain, vegetation and other features of these areas. Assessing terrain flatness and analyzing high flatness areas helps to understand the terrain characteristics and land use status of the development area. Through spectral feature extraction and index calculation, vegetation and water body information in low road density areas can be obtained, providing data support for subsequent vegetation and water body regional analysis. Analyzing high vegetation index and high water body index areas helps to understand vegetation coverage and water body distribution, and provides a basis for the identification of cultivated land areas. Obtaining cultivated land area data through intersection operation helps to identify land suitable for agricultural use, and provides a reference for the rational use of land resources. Ecological factor assessment of cultivated land areas can help us understand the ecological conditions of the agricultural production environment and provide guidance for sustainable agricultural development.

[0136] Optionally, step S49 is specifically:

[0137] Step S491: extracting soil features and water features from the development area sensor data, thereby obtaining the development area soil sensor data and the development area water sensor data;

[0138] In this embodiment, the principal component analysis (PCA) method is used to extract soil and water features using hyperspectral remote sensing data. For soil feature extraction, soil spectral features such as absorption peaks and reflectance are identified and classified in combination with the ground feature spectral library. Water feature extraction uses water spectral reflectance features and combines water indexes (such as NDWI) to extract and classify water, thereby obtaining soil sensor data and water sensor data of the development area.

[0139] Step S492: marking the cultivated land type on the cultivated land area remote sensing image, thereby obtaining a cultivated land type classification image;

[0140] In this embodiment, high-resolution remote sensing images are used to perform multi-category classification and labeling of cultivated land areas, such as wheat fields, corn fields, etc., using support vector machines (SVM) or deep learning methods. During the labeling process, the spectral characteristics, morphological characteristics and spatial relationships of the objects are considered to improve the classification accuracy, and finally an image of cultivated land type classification is obtained.

[0141] Step S493: integrating the cultivated land type classification image according to the soil sensor data of the development area and the water sensor data of the development area, thereby obtaining the cultivated land area sensor data;

[0142] In this embodiment, the soil sensor data and water sensor data of the development area are integrated with the cultivated land type classification image at the pixel level, and the cultivated land type to which each pixel belongs is determined according to the soil characteristics and water characteristics of each pixel to obtain the cultivated land area sensor data.

[0143] Step S494: calculating the soil distribution ratio of the cultivated land area sensor data, thereby obtaining the soil distribution ratio data of the cultivated land area; calculating the water distribution ratio of the cultivated land area sensor data, thereby obtaining the water distribution ratio data of the cultivated land area;

[0144] In this embodiment, the soil distribution ratio is calculated for the cultivated land area sensor data, and the soil coverage ratio of each type of cultivated land is calculated based on the number of soil characteristic pixels in each cultivated land type and the total number of pixels. The water body distribution ratio is calculated in a similar manner, and the water body coverage ratio of each type of cultivated land is calculated based on the number of water body characteristic pixels.

[0145] Step S495: performing water-soil connectivity analysis on the soil distribution ratio data of the cultivated land area and the water distribution ratio data of the cultivated land area, thereby obtaining water-soil connectivity data;

[0146] In this embodiment, based on the soil distribution ratio data and water body distribution ratio data of the cultivated land area, the connectivity between soil and water bodies is analyzed using spatial analysis methods (such as spatial weight matrix and spatial correlation function). By calculating the spatial correlation between the soil coverage area and the water coverage area, the water-soil connectivity data is obtained, which reflects the connection strength between water and soil in the cultivated land ecosystem.

[0147] Step S496: Calculating the plant coverage rate of the cultivated land type classification image, thereby obtaining plant coverage rate data;

[0148] In this embodiment, remote sensing images are used to calculate vegetation coverage, a pixel-level classification method is used to identify vegetation areas, and then plant coverage data is calculated based on the ratio of the number of vegetation pixel points to the total number of pixel points.

[0149] Step S497: Construct a cultivated land ecological factor evaluation system based on the cultivated land area soil distribution ratio data, plant coverage data, cultivated land area water body distribution ratio data and water-soil connectivity data, and use the cultivated land ecological factor evaluation system to evaluate the cultivated land area sensor data, thereby obtaining the cultivated land area ecological factor data.

[0150] In this embodiment, the cultivated land area soil distribution ratio data, plant coverage data, water body distribution ratio data and water-soil connectivity data are comprehensively considered to construct a cultivated land ecological factor evaluation system. The evaluation index weights are set according to relevant data, and the impact of different factors on the cultivated land ecosystem is comprehensively evaluated, and finally the cultivated land area ecological factor data is obtained to evaluate the health status and environmental quality of the cultivated land ecosystem.

[0151] The present invention can obtain the soil texture, humidity and water distribution of the development area by extracting soil characteristics and water characteristics, and provide a data basis for the subsequent calculation of soil distribution ratio and water distribution ratio. Marking the cultivated land type on the remote sensing image of the cultivated land area helps to accurately identify different types of cultivated land, and provides an accurate cultivated land range for the subsequent ecological factor evaluation. Integrating soil sensor data and water sensor data into the cultivated land type classification image can provide relevant information of soil and water for each cultivated land type, and provide a data basis for the ecological factor evaluation. Calculating the soil distribution ratio and water distribution ratio of the cultivated land area helps to understand the soil quality and water coverage of different types of cultivated land, and provides quantitative data support for the ecological factor evaluation. Through water-soil connectivity analysis, the connection between water and soil can be evaluated, the impact of water resources on land use and ecological environment can be understood, and a reference basis for ecosystem health assessment can be provided. Calculating the plant coverage rate helps to understand the vegetation coverage of different types of cultivated land, and provides vegetation-related data support for ecological factor evaluation. A cultivated land ecological factor assessment system is constructed based on data such as soil distribution ratio, water body distribution ratio, water-soil connectivity and plant coverage rate. By conducting ecological factor assessment on sensor data, the health status of the cultivated land ecological environment can be comprehensively evaluated, providing a scientific basis for land resource management and ecological protection.

[0152] Optionally, step S5 specifically includes:

[0153] Step S51: marking the region of the high-resolution remote sensing image according to the natural region remote sensing image and the cultivated region remote sensing image, thereby obtaining a monitoring region marked remote sensing image;

[0154] In this embodiment, a convolutional neural network (CNN) is used to supervise the learning of high-resolution remote sensing images, and the model is trained to identify different types of objects, such as water bodies, vegetation, and buildings. Remote sensing images of natural areas and cultivated areas are used for data enhancement and model verification to ensure the generalization ability of the model in different scenarios. The images of the monitored area are labeled at the pixel level, and semantic segmentation technology is used to ensure that each pixel is correctly classified.

[0155] Step S52: dividing the monitoring area marked remote sensing image into adjacent marked areas according to a preset area division ratio, thereby obtaining a transition area remote sensing image;

[0156] In this embodiment, the monitoring area marked remote sensing image is divided into adjacent areas of similar size according to a preset area division ratio, and a boundary detection algorithm is used to ensure the continuity and consistency between adjacent areas. This can use a pixel-based boundary detection method, such as the Canny edge detector, or a region-based method, such as connected region analysis.

[0157] Step S53: performing spatial correlation on the remote sensing image of the transition area and the sensor data of the monitoring area to obtain the sensor data of the transition area, and performing soil feature extraction and water feature extraction on the sensor data of the transition area to obtain soil sensor data of the transition area and water sensor data of the transition area;

[0158] In this embodiment, the sensing data of the monitoring area and the remote sensing image of the transition area are used to fill and predict the sensing data of the transition area using a method based on spatial association, such as spatial kriging interpolation or spatial autocorrelation analysis. Then, soil and water feature extraction algorithms, such as soil moisture index or water spectral feature extraction, are used to extract soil and water feature information from the remote sensing image of the transition area.

[0159] Step S54: Calculating the soil distribution ratio based on the soil sensor data in the transition area, thereby obtaining the soil distribution ratio data in the transition area; calculating the water distribution ratio based on the sensor data in the transition area, thereby obtaining the water distribution ratio data in the transition area;

[0160] In this embodiment, based on the extracted soil and water characteristics, combined with the ground object distribution information, a statistical analysis method such as least squares method or principal component analysis is used to calculate the soil and water distribution ratio in the transition area. This can associate the sensor data with the ground object type to infer the distribution of soil and water, and then determine their proportion in the entire transition area.

[0161] Step S55: Calculating the wetland percentage of the transition area remote sensing image, thereby obtaining wetland percentage data of the transition area;

[0162] In this embodiment, remote sensing images of the transition area are used to identify and classify wetlands at the pixel level using deep learning-based image classification methods, such as convolutional neural networks (CNN). Then, the number of pixels of wetlands in the entire transition area is calculated, and the proportion of wetlands in the transition area is calculated by pixel ratio.

[0163] Step S56: Calculating vegetation coverage and water area ratio of the remote sensing image of the transition area, thereby obtaining vegetation coverage data and water area ratio data of the transition area;

[0164] In this embodiment, the vegetation coverage is calculated by using a vegetation index calculation method, such as the Normalized Difference Vegetation Index (NDVI), using a remote sensing image of the transition area. At the same time, the area of ​​the water body is calculated using a water body boundary detection algorithm, such as a threshold-based method or an edge detection method. This can be achieved through pixel statistics or object detection to obtain the ratio data of the vegetation coverage and the water body area.

[0165] Step S57: integrating the remote sensing image of the transition area according to the soil sensing data of the transition area and the water sensing data of the transition area, thereby obtaining the sensing data of the transition area;

[0166] In this embodiment, data fusion technology, such as weighted average or multi-sensor fusion, is used to integrate the soil sensor data and water sensor data of the transition area into the remote sensing image of the transition area. This can be achieved by spatial alignment of remote sensing images and correction of sensor data to ensure consistency and accuracy between different data sources.

[0167] Step S58: constructing a transition area ecological factor evaluation system based on the transition area soil distribution ratio data, the transition area water body distribution ratio data, the transition area wetland ratio data, the transition area vegetation coverage rate data and the transition area water body area ratio data, and performing transition area ecological factor evaluation on the transition area sensor data through the transition area ecological factor evaluation system, thereby obtaining transition area ecological factor data;

[0168] In this embodiment, based on the soil and water distribution ratio data, wetland proportion data, vegetation coverage data and water area ratio data in the transition area, an ecological factor evaluation system for the transition area is constructed. This can use the analytic hierarchy process or expert evaluation method to determine the weight of each factor, thereby constructing an ecological factor evaluation system for the transition area. Then, the ecological factors of the transition area are comprehensively evaluated through the ecological factor evaluation system for the transition area combined with statistical analysis and spatial interpolation technology.

[0169] Step S59: spatially merge the transition area ecological factor data, the natural area ecological factor data and the cultivated area ecological factor data to obtain regional ecological factor data.

[0170] In this embodiment, the transition area ecological factor data, natural area ecological factor data and cultivated area ecological factor data are spatially merged using geographic information system (GIS) software. This may involve aligning the spatial ranges of different data sources and using spatial statistical methods, such as kriging interpolation or spatial analysis, to integrate and analyze the ecological factor data.

[0171] The present invention can mark the region of high-resolution remote sensing images by combining natural area remote sensing images and cultivated area remote sensing images, so as to accurately determine the scope and boundary of the monitoring area, and provide basic data for subsequent analysis. According to the preset regional division ratio, the marked remote sensing image of the monitoring area is divided into adjacent marked areas to obtain the remote sensing image of the transition area. This helps to reasonably divide the region in space and provides a basis for subsequent data integration and analysis. By spatially correlating the sensing data of the monitoring area and the remote sensing image of the transition area, the sensing data of the transition area is obtained. The soil and water body features of these data are further extracted to provide data support for the subsequent ecological factor evaluation. Based on the soil sensing data and sensing data of the transition area, the soil and water body distribution ratio is calculated, which helps to understand the land use and water resource distribution in the transition area. The wetland proportion calculation of the remote sensing image of the transition area can evaluate the distribution of wetland resources in the transition area, which provides important data for the evaluation of ecological factors. The vegetation coverage rate and water area ratio of the transition area are calculated to help understand the ecological environment of the transition area, and provide data support for the evaluation of ecological factors. Integrate soil sensor data and water sensor data in the transition area with remote sensing images to provide a variety of data sources for ecological factor assessment. Construct an ecological factor assessment system for the transition area based on data such as soil distribution ratio, water distribution ratio, wetland proportion, and vegetation coverage, and conduct a comprehensive assessment of the ecological environment in the transition area. Spatially merge the ecological factor data of the transition area with the ecological factor data of the natural area and the cultivated land area to form a complete regional ecological factor data, providing a comprehensive basis for regional ecological environment management and protection.

[0172] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0173] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for estimating ecological carrying capacity based on multi-source remote sensing data, characterized in that: The following steps are involved: Step S1: acquiring the monitoring area sensor data and the monitoring area remote sensing data, and performing multi-dimensional remote sensing data fusion based on the monitoring area remote sensing data, so as to obtain a high-resolution remote sensing image; performing surface resource division on the high-resolution remote sensing image, so as to obtain a natural area remote sensing image and a development area remote sensing image; Step S2: spatially correlating the natural area remote sensing image and the development area remote sensing image according to the monitoring area sensor data, thereby obtaining the natural area sensor data and the development area sensor data; Step S3: classifying the land cover types of the natural area remote sensing images, thereby obtaining regional land cover type classification spatial data; The natural ecological factors are evaluated based on the natural area sensor data and the regional land cover type classification spatial data, so as to obtain the natural area ecological factor data; Step S4: dividing the remote sensing image of the development area into cultivated land types, thereby obtaining a remote sensing image of the cultivated land area; According to the sensor data of the development area, the ecological factors of cultivated land are evaluated on the remote sensing images of the cultivated land area, so as to obtain the ecological factor data of the cultivated land area; Step S5: According to the monitoring area sensor data, the natural area remote sensing image and the cultivated area remote sensing image are evaluated for transition area ecological factors, so as to obtain transition area ecological factor data; the transition area ecological factor data, the natural area ecological factor data and the cultivated area ecological factor data are spatially merged to obtain regional ecological factor data. Step S5 is specifically as follows: Step S51: marking the region of the high-resolution remote sensing image according to the natural region remote sensing image and the cultivated region remote sensing image, thereby obtaining a monitoring region marked remote sensing image; Step S52: dividing the monitoring area marked remote sensing image into adjacent marked areas according to a preset area division ratio, thereby obtaining a transition area remote sensing image; Step S53: performing spatial correlation on the remote sensing image of the transition area and the sensor data of the monitoring area to obtain the sensor data of the transition area, and performing soil feature extraction and water feature extraction on the sensor data of the transition area to obtain soil sensor data of the transition area and water sensor data of the transition area; Step S54: Calculating the soil distribution ratio based on the soil sensor data in the transition area, thereby obtaining the soil distribution ratio data in the transition area; calculating the water distribution ratio based on the sensor data in the transition area, thereby obtaining the water distribution ratio data in the transition area; Step S55: Calculating the wetland percentage of the transition area remote sensing image, thereby obtaining wetland percentage data of the transition area; Step S56: Calculating the vegetation coverage rate and the water area ratio of the remote sensing image of the transition area, so as to obtain the vegetation coverage rate data and the water area ratio data of the transition area; Step S57: integrating the remote sensing image of the transition area according to the soil sensing data of the transition area and the water sensing data of the transition area, thereby obtaining the sensing data of the transition area; Step S58: constructing a transition area ecological factor evaluation system based on the transition area soil distribution ratio data, the transition area water body distribution ratio data, the transition area wetland ratio data, the transition area vegetation coverage rate data and the transition area water body area ratio data, and performing transition area ecological factor evaluation on the transition area sensor data through the transition area ecological factor evaluation system, thereby obtaining transition area ecological factor data; Step S59: spatially merging the transition area ecological factor data, the natural area ecological factor data, and the cultivated area ecological factor data to obtain regional ecological factor data; Step S6: construct an ecological carrying capacity estimation model based on regional ecological factor data and high-resolution remote sensing images, and estimate the ecological carrying capacity of the regional ecological factor data and high-resolution remote sensing images according to the ecological carrying capacity estimation model, so as to obtain regional ecological carrying capacity estimation data.

2. The method for estimating ecological carrying capacity based on multi-source remote sensing data according to claim 1, characterized in that: Step S1 is specifically as follows: Step S11: Acquire monitoring area sensor data and monitoring area remote sensing data; Step S12: Performing atmospheric correction and radiation correction on the remote sensing data of the monitoring area, thereby obtaining pre-processed remote sensing data; Step S13: extracting terrain features from the pre-processed remote sensing data to obtain regional terrain feature data; Step S14: constructing a three-dimensional image of the regional terrain according to the regional terrain feature data, and performing remote sensing data space mapping on the three-dimensional image of the regional terrain according to the pre-processed remote sensing data, thereby obtaining a high-resolution remote sensing image; Step S15: Divide the high-resolution remote sensing image into surface resources, thereby obtaining remote sensing images of natural areas and remote sensing images of development areas.

3. The method for estimating ecological carrying capacity based on multi-source remote sensing data according to claim 2, characterized in that: Step S15 is specifically as follows: Step S151: performing pixel statistics on the high-resolution remote sensing image to obtain pixel data of the remote sensing image; Step S152: performing adjacent pixel difference calculation on the remote sensing image pixel data, thereby obtaining adjacent pixel difference data; Step S153: performing high pixel difference pixel connection on the high-resolution remote sensing image according to the adjacent pixel difference data, connecting the pixels with large adjacent pixel differences, thereby obtaining a high pixel difference pixel connection image; Step S154: dividing the high-value pixel difference pixel point connection image into natural area remote sensing images, thereby obtaining natural area remote sensing images; dividing the high-value pixel difference pixel point connection image into natural area remote sensing images, thereby obtaining development area remote sensing images.

4. The method for estimating ecological carrying capacity based on multi-source remote sensing data according to claim 3, characterized in that: Step S154 is specifically as follows: The high pixel difference pixel point connection image is divided into connected closed area images, so as to obtain a connected closed area image set; Performing closed area edge point detection on the connected closed area image set, thereby obtaining closed area edge point data; Calculate the angles of adjacent edge points on the edge point data of the closed area, so as to obtain the angle data of adjacent edge points; The angle data of adjacent edge points are evaluated according to the regular angle threshold. A set of regular angle ranges is set, with 0° to 15° as horizontal edge and 45° to 75° as vertical edge. By comparing with these regular angle ranges, the edge points that meet the rules are screened out, thereby obtaining the regular angle threshold. The connected closed area image set is divided into closed areas according to the rule angle threshold, so as to obtain the natural area image set and the development area image set; Spatially merge the natural area image set to obtain the natural area remote sensing image; The image sets of the development area are spatially merged to obtain the remote sensing image of the development area.

5. The method for estimating ecological carrying capacity based on multi-source remote sensing data according to claim 4, characterized in that: The closed area division in step S154 is specifically as follows: The angle data of adjacent edge points are classified and calculated according to the regular angle threshold, so as to obtain the angle data of regular edge points and the angle data of irregular edge points; Performing statistical analysis on the closed area edge point data according to the regular edge point angle data, thereby obtaining high-frequency regular edge point area data and low-frequency regular edge point area data; Performing statistical analysis on the closed area edge point data according to the irregular edge point angle data, thereby obtaining high-frequency irregular edge point area data and low-frequency irregular edge point area data; Perform intersection operation on high-frequency regular edge point area data and low-frequency irregular edge point area data to obtain development area data; Perform intersection operation on low-frequency regular edge point area data and high-frequency irregular edge point area data to obtain natural area data; Dividing the connected closed area image set into regions according to the development area data, thereby obtaining the development area image set; The connected closed region image set is divided into regions according to the natural region data, so as to obtain the natural region image set.

6. The method for estimating ecological carrying capacity based on multi-source remote sensing data according to claim 1, characterized in that: Step S3 is specifically as follows: Step S31: extracting pixel spectral features from natural area remote sensing images to obtain natural surface spectral data; Step S32: performing band reflectance statistics on the natural surface spectral data, thereby obtaining high-value blue band reflectance spectral data and high-value green band reflectance spectral data; Step S33: dividing the natural area remote sensing image into blue-reflection areas according to the high-value blue band reflectance spectral data, thereby obtaining a high-value blue-reflection area image; dividing the natural area remote sensing image into green-reflection areas according to the high-value green band reflectance spectral data, thereby obtaining a high-value green-reflection area image; Step S34: performing vegetation index calculation and water index calculation according to the natural surface spectral data, thereby obtaining natural surface vegetation index data and natural surface water index data, and performing same index feature extraction on the natural surface spectral data according to the natural surface vegetation index data and the natural surface water index data, thereby obtaining same index natural spectral data; Step S35: classifying the coverage type of the high-value green area image according to the natural surface vegetation index data, thereby obtaining a forest type image and a grassland type image; According to the natural surface water index data, the coverage type of the high-amount blue reflection area image is classified to obtain lake type images and river type images; Step S36: dividing the natural area remote sensing image into wetland areas according to the natural spectrum data with the same index, thereby obtaining a wetland type image; Step S37: spatially merging the wetland type image, the lake type image, the river type image, the forest type image, and the grassland type image to obtain regional land cover type classification spatial data; Step S38: Evaluate natural ecological factors based on natural area sensor data and regional land cover type classification spatial data to obtain natural area ecological factor data.

7. The method for estimating ecological carrying capacity based on multi-source remote sensing data according to claim 6, characterized in that: Step S38 is specifically as follows: Step S381: extracting soil features and water features from the natural area sensor data, thereby obtaining natural area soil sensor data and natural area water sensor data; Step S382: integrating the regional land cover type classification spatial data according to the natural area soil sensor data and the natural area water body sensor data, thereby obtaining the natural area land cover sensor data; Step S383: Calculating the soil type distribution ratio of the natural area soil sensor data, thereby obtaining the natural area soil type distribution ratio data; Calculate the water quality distribution ratio of the natural area water body sensor data to obtain the natural area water quality distribution ratio data; Step S384: Calculate the vegetation coverage rate and the water area ratio of the regional land cover type classification spatial data, so as to obtain the natural area vegetation coverage rate data and the natural area water area ratio data; Step S385: Construct a natural ecological factor evaluation system based on the natural area soil type distribution ratio data, the natural area water quality distribution ratio data, the natural area vegetation coverage rate data and the natural area water area ratio data, and use the natural ecological factor evaluation system to perform natural ecological factor evaluation on the natural area surface cover sensor data, so as to obtain natural area ecological factor data.

8. The method for estimating ecological carrying capacity based on multi-source remote sensing data according to claim 1, characterized in that: Step S4 is specifically as follows: Step S41: performing a topological structure analysis on the remote sensing image of the development area, thereby obtaining the topological structure of the development area; Step S42: evaluating the road density of the development area topology structure, thereby obtaining road density data; Step S43: dividing the development area topology into low road density areas according to the road density data, thereby obtaining low road density area data; Step S44: dividing the remote sensing image of the development area into regions according to the low road density area data, thereby obtaining a remote sensing image of the low road density area; Step S45: evaluating the terrain flatness of the remote sensing image of the low road density area to obtain regional terrain flatness data, and performing high flatness area distribution analysis based on the regional terrain flatness data to obtain high flatness area data; Step S46: extracting pixel spectral features from the remote sensing image of the low road density area to obtain spectral data of the low road density area, and calculating vegetation index and water index on the spectral data of the low road density area to obtain vegetation index data of the low road density area and water index data of the low road density area; Step S47: performing a high vegetation index regional distribution analysis on the remote sensing image of the low road density area according to the low road density area vegetation index data, thereby obtaining high vegetation index area data; performing a high water index regional distribution analysis on the remote sensing image of the low road density area according to the low road density area water index data, thereby obtaining high water index area data; Step S48: performing intersection operation on the high vegetation index area data, the high water body index area data and the high flatness area data, thereby obtaining the cultivated land area data, and performing cultivated land type remote sensing image classification on the development area remote sensing image according to the cultivated land area data, thereby obtaining the cultivated land area remote sensing image; Step S49: Evaluate the ecological factors of cultivated land on the remote sensing image of the cultivated land area according to the sensor data of the development area, so as to obtain the ecological factor data of the cultivated land area.

9. The method for estimating ecological carrying capacity based on multi-source remote sensing data according to claim 8, characterized in that: Step S49 is specifically as follows: Step S491: extracting soil features and water features from the development area sensor data, thereby obtaining the development area soil sensor data and the development area water sensor data; Step S492: marking the cultivated land type on the cultivated land area remote sensing image, thereby obtaining a cultivated land type classification image; Step S493: integrating the cultivated land type classification image according to the soil sensor data of the development area and the water sensor data of the development area, thereby obtaining the cultivated land area sensor data; Step S494: calculating the soil distribution ratio of the cultivated land area sensor data, thereby obtaining the soil distribution ratio data of the cultivated land area; calculating the water distribution ratio of the cultivated land area sensor data, thereby obtaining the water distribution ratio data of the cultivated land area; Step S495: performing water-soil connectivity analysis on the soil distribution ratio data of the cultivated land area and the water distribution ratio data of the cultivated land area, thereby obtaining water-soil connectivity data; Step S496: Calculating the plant coverage rate of the cultivated land type classification image, thereby obtaining plant coverage rate data; Step S497: Construct a cultivated land ecological factor evaluation system based on the cultivated land area soil distribution ratio data, plant coverage data, cultivated land area water body distribution ratio data and water-soil connectivity data, and use the cultivated land ecological factor evaluation system to evaluate the cultivated land area sensor data, thereby obtaining the cultivated land area ecological factor data.

Images