Remote sensing monitoring classification system and sample library construction method for seawall ecological characteristics

Abstract

The invention relates to the technical field of data processing, and discloses a remote sensing monitoring classification system for seawall ecological features and a sample library construction method. The method comprises the following steps: fusing and processing a multi-source remote sensing image to obtain a standardized data set; extracting seawall section characteristic parameters, and generating preliminary type distribution data; fusing geographic and ecological information to form feature element data; calculating vegetation coverage and material characteristics to obtain quantitative indexes; establishing a three-level sample library structure; and updating the sample through similarity matching, and constructing a classification system. According to the invention, remote sensing monitoring standardized expression and intelligent classification of seawall ecological features are realized.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a remote sensing monitoring classification system for ecological characteristics of seawalls and a sample library construction method. Background Art

[0002] As an important coastal protection engineering facility, the ecological construction of seawalls is of great significance to maintaining the stability of coastal ecosystems. At present, the investigation and evaluation of the ecological characteristics of seawalls mainly rely on the combination of manual on-site surveys and remote sensing image interpretation. By collecting data such as basic information of seawalls, vegetation coverage, slope protection structure characteristics and surrounding ecological environment, a basic seawall database is established. In terms of data processing, GIS spatial analysis and remote sensing image processing are used to classify, count and evaluate the spatial distribution, structural type and ecological characteristics of seawalls, providing decision-making support for the ecological construction and management of seawalls.

[0003] However, the existing methods for analyzing the ecological characteristics of seawalls have the following shortcomings: first, the data acquisition method is single, which makes it difficult to fully reflect the multi-dimensional characteristics of seawalls; second, the data processing process lacks standardization and automation, and manual interpretation is labor-intensive and highly subjective; third, the quantitative evaluation of ecological characteristics lacks a systematic indicator system, making it difficult to make an objective comparison; finally, there is a lack of a complete sample library to provide an effective reference for the construction and renovation of seawalls. Summary of the invention

[0004] The present application provides a remote sensing monitoring classification system and sample library construction method for the ecological characteristics of seawalls, which are used to achieve standardized expression and intelligent classification of remote sensing monitoring of the ecological characteristics of seawalls.

[0005] In the first aspect, the present application provides a remote sensing monitoring classification system and sample library construction method for the ecological characteristics of seawalls, and the remote sensing monitoring classification system and sample library construction method for the ecological characteristics of seawalls include: multi-source data fusion processing of high-resolution satellite remote sensing images, aerial remote sensing images, drone orthophotos and field measurement data, obtaining digital orthophotos through aerial triangulation processing, and performing plane position and elevation correction on the digital orthophotos according to the control point coordinates to obtain a standardized multi-source data set; extracting the slope value, height value and width value of the seawall section according to the standardized multi-source data set, substituting the obtained characteristic parameter values ​​into preset expert rules for type identification, and generating preliminary seawall type distribution data; using the preliminary seawall The type distribution data performs feature correlation processing on basic geographic information, levee top ecological information, slope protection structure information and shore ecosystem information to form seawall ecological feature element data; based on the seawall ecological feature element data, the vegetation coverage of multispectral images is calculated, the image texture is used to determine the type of levee top greening, and a scoring operation is performed based on the slope protection material and ecological permeability to obtain quantitative feature indicators; based on the quantitative feature indicators, the sample data is standardizedly stored and encoded and multidimensional indexed according to a three-level structure to generate a hierarchical seawall sample library; similarity matching operations are performed on the spatial features, spectral features and structural features in the hierarchical seawall sample library, and the samples are updated through dynamic feedback processing to construct a seawall ecological feature classification system.

[0006] In the technical solution provided in this application, the accuracy and completeness of seawall feature information extraction are improved through the fusion processing of multi-source remote sensing data and aerial triangulation technology, and the information missing problem caused by a single data source is reduced; through standardized cross-sectional feature extraction and expert rule discrimination, the automatic identification of seawall types is realized, and the classification efficiency is improved; the feature association processing technology is used to integrate basic geographic information, ecological information, structural information and environmental information in multiple dimensions, which comprehensively reflects the ecological characteristics of the seawall; through quantitative vegetation coverage calculation, material feature analysis and ecological scoring, an objective evaluation index system is established to eliminate the uncertainty of subjective judgment; based on the three-level structure of the sample library design, the standardized storage and efficient retrieval of seawall feature information are realized, providing a reliable reference basis for subsequent seawall construction; through similarity matching and dynamic feedback mechanism, the continuous updating and optimization of the sample library is guaranteed, and the adaptability and accuracy of the classification system are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0008] Figure 1 It is a schematic diagram of an embodiment of the remote sensing monitoring classification system and sample library construction method of the ecological characteristics of the seawall in the embodiment of the present application. DETAILED DESCRIPTION

[0009] The embodiment of the present application provides a remote sensing monitoring classification system and sample library construction method for the ecological characteristics of seawalls. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0010] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 In the embodiment of the present application, an embodiment of the remote sensing monitoring classification system and sample library construction method of the ecological characteristics of the seawall includes: Step S101, multi-source data fusion processing is performed on high-resolution satellite remote sensing images, aerial remote sensing images, drone orthophotos and field measurement data, a digital orthophoto map is obtained through aerial triangulation processing, and the plane position and elevation deviation of the digital orthophoto map are corrected according to the coordinates of the control points to obtain a standardized multi-source data set; Step S102: extracting the slope value, height value and width value of the seawall section according to the standardized multi-source data set, substituting the obtained characteristic parameter values ​​into the preset expert rules for type identification, and generating preliminary seawall type distribution data; Step S103, using the preliminary seawall type distribution data, basic geographic information, levee crest ecological information, slope protection structure information and shore ecosystem information are subjected to feature association processing to form seawall ecological feature element data; Step S104: Calculate the vegetation coverage of the multispectral image based on the ecological characteristic element data of the seawall, determine the type of greening on the top of the seawall based on the image texture, and perform scoring calculations based on the slope protection material and ecological permeability to obtain quantitative characteristic indicators; Step S105: Perform standardized storage coding and multi-dimensional index construction on the sample data according to the three-level structure based on the quantitative characteristic indicators to generate a hierarchical seawall sample library; Step S106: perform similarity matching operations on the spatial features, spectral features and structural features in the hierarchical seawall sample library, update the samples through dynamic feedback processing, and build a seawall ecological feature classification system.

[0011] Specifically, it should be noted that in the present application, the multi-source data fusion processing of high-resolution satellite remote sensing images, aerial remote sensing images, UAV orthophotos and field measurement data is implemented. In the specific implementation process, the multi-source remote sensing monitoring data not only includes high-resolution satellite remote sensing images, aerial remote sensing images, UAV orthophotos and field measurement data, but also includes: panchromatic and multispectral fusion images, hyperspectral remote sensing images, lidar point cloud data, synthetic aperture radar images, thermal infrared remote sensing images, ground real-time monitoring site data and three-dimensional laser scanning data, etc.

[0012] Among them, the high-resolution satellite remote sensing images (spatial resolution better than 2 meters) are resampled, and the multispectral band data are fused through the Gram-schmidt mathematical transformation. The Gram-schmidt transformation orthogonalizes multiple bands according to their correlation and retains the independent information between the bands. For example, the near-infrared band and the red light band are orthogonally transformed to highlight the vegetation information. The fused data are aligned with the aerial remote sensing images (spatial resolution better than 1 meter) according to geographic coordinates. By selecting obvious ground points such as road intersections and building corners as the same-name points, the affine transformation and bilinear interpolation methods are used to achieve accurate alignment. The aligned images are fused with the drone orthophotos to enhance the expression of details in the local area. The drone images first need to be triangulated in the air, and the connection between the images is established by matching the connection points, and the external orientation elements are solved by the bundle method. In this process, ground control points are selected for constraints, and the coordinates of the control points are obtained by RTK measurement. For example, control points are arranged at the turning points and feature points of the seawall, and their three-dimensional coordinates are measured by RTK. The orthophoto image is corrected based on the coordinates of the control points to eliminate geometric deformation caused by terrain undulation and lens distortion.

[0013] After data fusion and geometric correction, a standardized multi-source data set is obtained, which contains spatial geometric information and spectral information. Based on this data set, the cross-sectional features of the seawall are extracted, and the edge detection and region segmentation of the seawall image are performed to extract the seawall contour. The seawall contour data is superimposed with the cross-sectional point cloud data measured on site, and the three-dimensional coordinates of each point on the cross section are calculated by spatial interpolation. A profile analysis is performed in the direction perpendicular to the coastline, and the ratio of the height difference between adjacent points to the horizontal distance is calculated to obtain the slope value, the height value is obtained by the elevation difference between the top and bottom of the seawall, and the width value is obtained by the horizontal distance between the points on both sides of the top road.

[0014] The extracted cross-section characteristic parameters are matched with expert rules. The expert rules specify the characteristic parameter thresholds of different types of seawalls. For example, the slope of a sloped seawall is usually between 15° and 45°, while a composite seawall contains a vertical section and a sloped section. By comparing the characteristic parameters with the rule thresholds, the seawalls are divided into two categories: sloped and composite. Further combined with the spatial topological relationship, a preliminary distribution layer of the seawall is generated.

[0015] Based on the preliminary classification results, a multi-dimensional feature correlation analysis was conducted on the seawall. The seawall trajectory was superimposed on the geographic base map to extract basic geographic information such as the latitude and longitude coordinates of the starting and ending points, the projection length, and the strike angle. Spectral analysis was performed on the top of the embankment to extract information on ecological elements such as vegetation and flower beds. Texture analysis was used to identify the surface material of the slope protection, such as fence panels, twisted king blocks, four-corner hollow squares and other structural types. Ecosystem types were identified in the area surrounding the seawall, including mangroves, bedrock beaches, sandy beaches, silty beaches, and salt marshes. Quantitative evaluation was performed based on feature correlation analysis. The normalized difference vegetation index (NDVI) was calculated for multispectral images, and the vegetation coverage was quantified by setting a classification threshold. Texture features were extracted from the top of the embankment, and the edge density and directional consistency were calculated to identify green belts, flower beds, and other greening types. The ecological permeability score of the slope protection material was calculated based on its characteristics, and the spatial distribution statistics of the surrounding ecosystem were conducted to form a quantitative feature indicator system.

[0016] When establishing a hierarchical sample library, a three-level organizational structure is adopted. The first level is the seawall type (slope type, composite type), the second level is the specific structural form (such as slope-platform-slope type, straight up and down slope type, etc.), and the third level is the combination of ecological characteristics. Each sample is given a unique identification code, and multimedia materials such as multi-angle images and point cloud data are associated with the identification code. A multi-dimensional retrieval mechanism is established to support queries based on geographical location, type characteristics, ecological indicators and other conditions. Finally, the samples are dynamically updated through similarity matching. Morphological analysis is performed on the spatial features in the sample library to calculate the contour curvature and continuity parameters. Band combination analysis is performed on the spectral features to calculate the vegetation index and material reflectance. Texture analysis is performed on the structural features to evaluate the matching degree between the slope protection type and the embankment top morphology. After the multi-dimensional feature similarity is normalized, a comprehensive score is given. Based on the scoring results, the samples are clustered and grouped, and the feature statistics of each group are extracted to form a classification system.

[0017] For example: for a certain coastal area, obtain Gaofen-2 satellite multispectral images, aerial remote sensing DOM images and drone oblique photography data. Multispectral bands are fused through Gram-schmidt transformation, and 20 ground control points are selected for geometric correction. The extraction of seawall section features shows that the average slope of this section of the seawall is 35°, the height is 4.5 meters, and the top width is 6 meters, which meets the characteristics of a sloped seawall. Further analysis found that there is a continuous green belt on the top of the embankment, and the slope protection adopts a concrete block plus ecological block structure, with mangroves and sandy mudflats distributed around it. After quantitative evaluation, the vegetation coverage is 0.6, the slope protection ecological score is 0.7, and the ecosystem distribution ratio is 0.4. The sample is coded and stored.

[0018] In the embodiments of the present application, the accuracy and completeness of the extraction of seawall feature information are improved through the fusion processing of multi-source remote sensing data and aerial triangulation technology, and the information missing problem caused by a single data source is reduced; through standardized cross-sectional feature extraction and expert rule discrimination, the automatic identification of seawall types is realized, and the classification efficiency is improved; the feature association processing technology is used to integrate basic geographic information, ecological information, structural information and environmental information in multiple dimensions, which comprehensively reflects the ecological characteristics of the seawall; through quantitative vegetation coverage calculation, material feature analysis and ecological scoring, an objective evaluation index system is established to eliminate the uncertainty of subjective judgment; based on the three-level structure of the sample library design, the standardized storage and efficient retrieval of seawall feature information are realized, providing a reliable reference basis for subsequent seawall construction; through similarity matching and dynamic feedback mechanism, the continuous updating and optimization of the sample library is guaranteed, and the adaptability and accuracy of the classification system are improved.

[0019] In a specific embodiment, the process of executing step S101 may specifically include the following steps: (1) Resample the bands of high-resolution satellite remote sensing images, input the resampled band data into the Gram-schmidt mathematical transformation, and obtain spectral fusion data; (2) Registering the spectral fusion data with the aerial remote sensing image according to the corresponding relationship of geographic coordinates to obtain the first registered image; (3) Perform affine transformation on the first registered image and the UAV orthophoto according to the same-name ground object points, and obtain the second registered image through bilinear interpolation; (4) Performing image stretching, equalization, and filtering enhancement processing on the second registered image using the on-site measurement data to generate enhanced image data; (5) Input the enhanced image data into the aerial triangulation program, and obtain the digital orthophoto map through tie point matching and bundle adjustment operations; (6) Substitute the coordinates of the control points into the correction equation, perform plane position correction and elevation correction calculations on the digital orthophoto, and output a standardized multi-source data set.

[0020] Specifically, the high-resolution satellite remote sensing images are subjected to band resampling. High-resolution remote sensing images contain multiple spectral bands, such as the blue light band (0.45-0.52μm), the green light band (0.52-0.60μm), the red light band (0.63-0.69μm) and the near-infrared band (0.76-0.90μm). The resampling process unifies the data of each band to the same spatial resolution, and uses the Gram-schmidt orthogonal transformation for spectral fusion. The key to the Gram-schmidt transformation is to gradually orthogonalize multiple bands according to their correlation and retain the independent information between the bands. For example, the near-infrared band and the red light band are orthogonally transformed to particularly highlight the vegetation information, which is helpful for the subsequent identification of the greening situation on the seawall.

[0021] The spectral fusion data is then spatially registered with the aerial remote sensing image. The registration process is based on the correspondence between geographic coordinates and selects ground control points as references. Ground control points are usually selected at locations with obvious features such as the corners of the seawall and the boundaries of the slope protection structure, and their precise three-dimensional coordinates are obtained by RTK measurement. During registration, an affine transformation model is used to transform the spectral fusion image into the same geographic coordinate system as the aerial remote sensing image. The registration accuracy is required to be controlled within one pixel to ensure that the two types of image data can be accurately superimposed. After obtaining the first registered image, it is necessary to further register it with the drone orthophoto. Drone images have higher spatial resolution, but the coverage is relatively small, and are mainly used to obtain local detailed features of the seawall. During registration, the same-name ground feature points are first selected in the two types of images, such as the feature points of the slope protection structure, the intersection of the road on the top of the embankment, etc. Based on these same-name points, an affine transformation relationship is established, and the image is resampled using the bilinear interpolation method to obtain the second registered image with precise spatial correspondence.

[0022] The second registered image also needs to be enhanced. Enhancement includes three key steps: image stretching, histogram equalization and filter enhancement. Image stretching adjusts the grayscale value range of the image to improve the image contrast; histogram equalization improves the overall brightness distribution of the image; filter enhancement highlights edge details through high-pass filtering, which is conducive to the subsequent extraction of seawall structure features. The parameter setting of the enhancement process needs to be verified and optimized in combination with field measurement data. The enhanced image data is input into the aerial triangulation program for processing. The core of aerial triangulation is to establish the connection between images through tie point matching, and then use the bundle method to solve the exterior orientation elements. Tie points are usually automatically extracted in the overlapping area of ​​the images, requiring the points to be evenly distributed and have obvious features. The bundle method adjustment comprehensively considers the constraints of the ground control points and the geometric relationship between the images, and obtains the optimal orientation parameters through iterative calculation. The digital orthophoto output of this process eliminates the geometric deformation caused by terrain undulation and camera tilt.

[0023] The digital orthophoto map is accurately corrected using the control point coordinates, that is, the correction equation is as follows: ; in: is the deviation correction displacement; , , is the correction coefficient in each direction; , , is the coordinate of the measuring point; , , is the reference point coordinate; is the weight factor. The coordinates of each pixel are corrected according to the actual geographical location, the systematic error and distortion are eliminated, and a standardized multi-source data set is output.

[0024] For example: first obtain high-resolution satellite multispectral images and panchromatic images, resample the bands to unify the spatial resolution to 2 meters, and then fuse the multispectral bands through Gram-schmidt transformation. Select 20 control points such as road intersections and building corners on the fused image and align them with the aerial remote sensing DOM image. The aligned image is then aligned with the drone oblique photography orthophoto, and 30 points of the same name such as the edge of the embankment road and the boundary of the slope protection structure are selected for precise correspondence. When enhancing the image, the grayscale value range is expanded to 0-255 through linear stretching, the equalization processing makes the histogram distribution more uniform, and the high-pass filtering highlights the contour edge. The aerial triangulation automatically extracts 200 connection points, and the bundle adjustment is performed in combination with 25 ground control points. Substitute into the correction equation to achieve centimeter-level plane position and elevation accuracy.

[0025] In a specific embodiment, the process of executing step S102 may specifically include the following steps: (1) Perform edge detection on the seawall images in the standardized multi-source dataset, segment the seawall area through region growing, and obtain the seawall contour data; (2) Superimpose the seawall contour data with the field measurement data, calculate the three-dimensional coordinates of each cross-section point through spatial interpolation, and obtain the seawall cross-section point cloud; (3) The seawall cross-section point cloud is cut into sections perpendicular to the coastline, and the height difference between adjacent cross-section points is divided by the horizontal distance to obtain the slope value; (4) Extract the elevation difference between the top and bottom points of the seawall from the seawall cross-section point cloud as the height value, and extract the horizontal distance between the points on both sides of the road at the top of the seawall as the width value; (5) Substituting the slope value, height value and width value into the stratified threshold discriminant formula, the seawall is divided into two types: slope type and composite type, and the seawall type discrimination result is obtained; (6) The seawall type identification results are associated with geographic coordinates through spatial topological relationships, and preliminary seawall type distribution data are output.

[0026] Specifically, edge detection is performed on the seawall image. The edge detection uses a multi-scale gradient operator to extract the boundary information between the seawall and the background at different resolutions. The gradient operator calculates the grayscale change rate of the image in the horizontal and vertical directions, and marks the pixels with gradient values ​​greater than the set threshold as edge points. After obtaining the preliminary edge point set, the seawall area is segmented by the region growing algorithm. Region growing starts from the pre-selected seed point and gradually merges adjacent similar pixels into the current area. The similarity judgment is based on multiple attributes such as pixel grayscale value and texture features. This process is iterated until no new pixels that meet the conditions can be found, and the seawall contour data is obtained.

[0027] The seawall contour data is spatially superimposed with the cross-section point data obtained by on-site RTK measurement. Each RTK measurement point contains accurate three-dimensional coordinate information. The discrete measurement points are expanded into a continuous cross-section point cloud by the Kriging spatial interpolation method. Kriging interpolation takes into account spatial autocorrelation and estimates the three-dimensional coordinates of unknown points based on the positional relationship and attribute values ​​of known points. In this way, densely and evenly distributed point cloud data can be obtained on each section. After obtaining the seawall cross-section point cloud, the section interception direction is determined. By analyzing the direction of the coastline, its normal direction is determined as the section interception reference. In the direction perpendicular to the coastline, the cross-section profile is extracted at fixed intervals. For adjacent point cloud data on each section, the ratio of its elevation difference to the horizontal distance is calculated, that is, the slope value of the section is obtained. Since the seawall section often presents a stepped feature, it is necessary to calculate the slope in sections and record the start and end positions of each section.

[0028] When extracting feature parameters from the seawall cross-section point cloud, the top and bottom feature points are first identified. The top point usually corresponds to the edge of the road on the top of the levee, with obvious elevation mutation characteristics; the bottom point is located at the junction of the seawall and the beach, and also shows a drastic change in elevation. The elevation difference between the two feature points is the height of the seawall. For the extraction of the levee top width, the edge points on both sides of the road are identified and their horizontal distance is calculated.

[0029] To identify the type of seawall, the following hierarchical threshold discriminant formula is used in the identification process: ; in: is the discriminant index; is the slope angle of the i-th section; is the height of the seawall; is the width of the embankment crest; is the structural characteristic index of the i-th segment; is the weight coefficient of each feature; is the segment correction coefficient; n is the number of cross-section segments.

[0030] When the discrimination index J is greater than the set threshold, it is judged as a composite seawall, otherwise it is judged as a slope seawall. The discrimination result needs to be associated with the geographic coordinate information to establish a spatial topological relationship. The type attribute of each seawall is bound to its geographic location through spatial indexing technology to form seawall type distribution data.

[0031] For example, when processing a certain coastal section, the seawall image is first extracted from the standardized multi-source data set. The general outline of the seawall is identified by edge detection, and then regional growth is performed based on pre-selected seed points. The seed points are selected at locations with obvious seawall features, such as the centerline of the road on the top of the embankment or the boundary of the slope protection structure. During the regional growth process, pixels with grayscale values ​​within 10 grayscale levels and texture feature similarity exceeding 0.8 are gradually included in the target area to obtain the seawall outline. Combined with the on-site RTK measurement data, a cross-section measurement point is arranged every 50 meters, and 5-7 feature points are arranged in each cross-section from the bottom to the top of the seawall. Kriging interpolation is used to generate continuous point clouds between cross-sections, and the cross-section spacing is set to 5 meters. Profile analysis is performed in the direction perpendicular to the coastline to calculate the slope values ​​of each section. For example, a cross-section is divided into three sections from the bottom to the top: the bottom section has a slope of 35 degrees, the middle section is a horizontal platform, and the top section has a slope of 40 degrees. At the same time, the total height of the seawall is measured to be 4.5 meters, and the width of the road on the top of the embankment is 6 meters. Substituting these parameters into the hierarchical threshold discriminant, considering that the section contains an obvious horizontal platform section and has a steep slope change, the discriminant index calculation result is greater than the threshold, so it is judged to be a composite seawall. This discrimination result is associated with the start and end coordinates of the section through the spatial topological relationship, and the attribute interpolation is performed between the front and rear sections to obtain the type distribution of this section of the seawall. This processing process fully utilizes the advantages of multi-source data, which not only ensures the accuracy of geometric feature extraction, but also realizes the automatic discrimination of seawall types.

[0032] In a specific embodiment, the process of executing step S103 may specifically include the following steps: (1) Superimpose the seawall trajectory lines in the preliminary seawall type distribution data with the geographic base map, extract the longitude and latitude coordinates of the starting and ending points of the seawall, the projection length and the strike angle, and obtain basic geographic information; (2) Band combination analysis was performed on the top area of ​​the seawall in the standardized multi-source dataset to separate the spectral characteristics of green plants, flower beds, and hardened ground to obtain ecological information on the top of the seawall; (3) The surface materials of the slope protection in the preliminary seawall type distribution data were distinguished through texture analysis, and classified into fence panels, twisted blocks, four-corner hollow blocks, concrete smooth structures, riprap slope protection, and mortar masonry to obtain the slope protection structure information; (4) Extract the area around the seawall from the standardized multi-source data set, identify the distribution range of mangroves, bedrock beaches, sandy beaches, muddy beaches, and salt marshes through spectral separation, and obtain information on the beach ecosystem; (5) Combine and associate basic geographic information, embankment ecological information, slope protection structure information, and shore ecosystem information according to spatial location correspondence; (6) Structural organization and unified coding of the associated multi-dimensional feature information are carried out to output the ecological feature element data of the seawall.

[0033] Specifically, the preliminary seawall type distribution data is processed with geographic information, and the seawall trajectory is superimposed on the high-precision geographic base map. The geographic base map contains standardized coordinate grids and terrain information, and spatial registration is used to ensure that the trajectory line corresponds accurately to the base map. The extraction process uses a vector data processing method to obtain the latitude and longitude coordinates of the start and end points from the endpoints of the trajectory line, calculate the length of the line segment on the projection plane, and calculate the strike angle through the endpoint coordinates. The basic geographic information obtained in this way describes the spatial distribution characteristics of the seawall. When performing band combination analysis on the top area of ​​the seawall in the standardized multi-source data set, a multispectral remote sensing image processing method is used. The near-infrared band is combined with the visible light band, and the spectral reflectance characteristics of different landforms are used for classification. Green plants have the characteristics of high reflectivity in the near-infrared band and low reflectivity in the red light band. The flower bed area shows regular green plant arrangement and flower spectral characteristics, while the hardened ground shows a higher reflectivity and a flat spectral curve. Through spectral feature comparison and spatial distribution feature analysis, the top area of ​​the levee is classified into types such as green belts, flower beds, scattered green plants and hardened ground, forming the ecological information of the levee top.

[0034] The extraction of slope protection structure information mainly relies on texture analysis. Different slope protection materials have characteristic texture patterns: fence panels present regular grid textures, twisted blocks present continuous corrugations, four-corner hollow squares form a regular grid arrangement, concrete smooth structure texture is relatively uniform, riprap slope protection presents irregular blocks, and mortar masonry presents a regular stone splicing texture. The texture feature parameters of each material are extracted through the grayscale co-occurrence matrix, including statistics such as energy, contrast, and entropy, and a material classification feature library is established. Then, the texture features of the area to be analyzed are matched with the feature library to obtain the material type of each section of the slope protection. The identification of the shore ecosystem around the seawall is assisted by spectral separation technology. Mangroves have obvious high reflectivity characteristics in the near-infrared band, and their spatial distribution is irregular and patchy; bedrock beaches have high reflectivity in the visible light band and rough surface texture; the spectral curve of sandy beaches is relatively flat in the visible light band, and the nearshore shows characteristic wave reflection; muddy beaches show low spectral reflectivity and fine surface texture; salt marshes have unique spectral characteristics in the near-infrared and short-wave infrared bands. Through spectral feature analysis and spatial morphological feature identification, the distribution range of various ecosystems can be divided.

[0035] When integrating the above-mentioned information in space, a unified spatial reference system is first established. Basic geographic information is used as a framework to provide a spatial positioning framework for the seawall. The ecological information of the levee crest, the slope protection structure information, and the shore ecosystem information are superimposed on the framework according to their spatial position correspondence. Spatial indexing technology is used to associate information at different levels to ensure that the characteristic information at each spatial position corresponds one to one.

[0036] The multi-dimensional feature information after association is structured and organized. A unified coding rule is designed, including position coding, type coding and feature coding. Position coding reflects the spatial location of the seawall section, type coding indicates the seawall structure type, and feature coding describes the ecological and structural characteristics. Standardized storage and rapid retrieval of information are achieved through coding annotation.

[0037] For example, a 2-kilometer-long seawall section was analyzed. The starting longitude was 120.5 degrees and latitude was 30.2 degrees, and the end longitude was 120.6 degrees and latitude was 30.3 degrees. The calculated projection length was 2150 meters, and the overall direction was 45 degrees northeast. Multispectral image analysis showed that there was a continuous green belt on the top of the embankment, and its near-infrared band reflectivity was significantly higher than that of the surrounding hardened road surface, with three flower beds dotted in the middle. Slope protection structure analysis found that this section of the seawall adopted a mixed structure: the upper section was a fence board slope protection, showing a regular grid texture; the middle section was a twisted king block, showing a characteristic corrugated shape; the lower section was a riprap slope protection, with an irregular block texture. The surrounding shore ecosystem includes the inner salt marsh belt, showing typical tidal gully texture and high-salt vegetation spectral characteristics, and the outer sandy beach, showing a uniform band distribution. After integrating this information according to spatial location, it is given a unified code and forms a feature element data set to describe the ecological characteristics of this section of the seawall.

[0038] In a specific embodiment, the process of executing step S104 may specifically include the following steps: (1) The multispectral images in the ecological characteristic element data of the seawall are used to perform normalized vegetation index calculations, and the vegetation coverage density is quantitatively graded by setting a classification threshold to obtain the percentage of vegetation coverage; (2) The ecological information of the dike top is segmented, and the spatial distribution characteristics of the dike top green belt, flower bed, and scattered vegetation are extracted by edge density and direction consistency calculation to obtain the distribution morphological data of the dike top greening; (3) The surface characteristics of different materials in the slope protection structure information are converted into permeability coefficients, and the ecological properties of the slope protection surface structure are quantitatively scored to obtain the ecological permeability score value; (4) Calculate the area and spatial distribution of various ecosystems in the beach ecosystem information to obtain the distribution ratio of ecosystems around the seawall; (5) Standardize the data of vegetation coverage percentage, embankment greening distribution morphology, ecological permeability score and ecosystem distribution ratio; (6) Perform weighted sum operation on each standardized index according to its weight, and output quantitative characteristic index.

[0039] Specifically, the vegetation index is calculated for the multispectral images in the ecological characteristic element data of the seawall. The normalized difference vegetation index (NDVI) is an important parameter to characterize the vegetation growth status. The multispectral image contains the near-infrared band (0.76-0.90μm) and the visible red band (0.63-0.69μm). Vegetation has high reflectivity in the near-infrared band and low reflectivity in the red band. By calculating the NDVI value and setting the classification threshold, the vegetation coverage density is divided into four levels: no vegetation coverage (NDVI < 0.1), low coverage (0.1 ≤ NDVI < 0.3), medium coverage (0.3 ≤ NDVI < 0.6) and high coverage (NDVI ≥ 0.6), so as to obtain the area proportion of each level.

[0040] When conducting an in-depth analysis of the ecological information on the top of the levee, regional segmentation is performed. The segmentation process is based on the image edge and regional growth algorithm, and the edge density parameters are calculated to identify the boundaries of different types of green areas. Among them, the green belt shows a continuous linear distribution feature, with uniform edge density and strong directionality; the flower bed area presents a regular geometric shape, with a high edge density and clear boundaries; the scattered vegetation shows a discrete distribution with uneven edge density. By calculating the spatial distribution density and main direction consistency of the edge points, the distribution characteristics of various greening forms are extracted to form the distribution form data of the greening on the top of the levee. The ecological evaluation of the slope protection structure mainly considers its permeability. There are significant differences in the permeability of different materials: fence panels and four-corner hollow squares have higher permeability, which is conducive to the habitat of small organisms; twisted blocks and riprap slope protection are second, which can provide space for plant growth; concrete smooth structure has the worst permeability, which is not conducive to ecological development. The permeability coefficient is assigned to each material through professional evaluation standards, and the comprehensive ecological permeability score is calculated in combination with the surface structure characteristics.

[0041] The spatial distribution analysis of the shore ecosystem needs to consider two aspects: area and spatial configuration. The following calculation formula is used: ; in: is a comprehensive ecosystem index; is the area of ​​ecosystem of type i; is the area weight coefficient; is the ecological importance coefficient; is the boundary length between the i-th and j-th ecosystems; is the ecological value coefficient of the boundary zone; is the spatial connection strength; is the overall coordination coefficient; n is the number of ecosystem types.

[0042] The above quantitative indicators are standardized to eliminate the differences in dimensions and numerical ranges. The maximum and minimum value method is used for standardization to map the values ​​of each indicator to the interval [0,1]. Finally, the standardized indicators are weighted and summed. The weight value is determined according to the degree of influence of each indicator on the ecological function of the seawall, and the quantitative characteristic indicators are obtained.

[0043] For example, to evaluate the ecological performance of a certain section of seawall, we first extract near-infrared and red band data from multispectral images to calculate NDVI. By statistically analyzing the distribution of NDVI values, we found that the NDVI values ​​of the top area of ​​the embankment are concentrated between 0.4 and 0.7, indicating that the vegetation is growing well. Regional segmentation analysis shows that there is a continuous green belt with a width of 3 meters on the top of the embankment, with uniform edge density and direction consistent with the direction of the seawall. A flower bed is set every 200 meters, showing a regular rectangular distribution. The slope protection adopts a composite structure, with ecological blocks (permeability coefficient 0.8) on the upper part and riprap structure (permeability coefficient 0.6) on the lower part. There are three ecosystems distributed around, namely mangroves, sandy beaches and silty beaches, with an area of ​​10 hectares, 15 hectares and 8 hectares respectively, and a good ecological transition zone is formed between mangroves and sandy beaches. Through standardization and weighted calculation, the comprehensive index of the ecological characteristics of this section of seawall is obtained to reflect its ecological function status.

[0044] In a specific embodiment, the process of executing step S105 may specifically include the following steps: (1) The quantitative characteristic indicators were hierarchically grouped, with seawall type as the first level, seawall structure as the second level, and ecological characteristics as the third level, resulting in a three-level hierarchical data structure; (2) Uniquely encode each seawall sample in the three-level hierarchical data structure, and obtain the sample identification code by splicing the type code, structure code, and ecological characteristic code; (3) Associating and binding the sample identification code with the multi-angle images, oblique photography point cloud data, and field photos in the ecological feature element data of the seawall to obtain a sample multimedia data set; (4) Extract multi-dimensional feature values ​​of the sample multimedia dataset according to geographic location, type characteristics, and ecological indicators to obtain the sample retrieval feature vector; (5) Convert the sample retrieval feature vector into an inverted index structure, establish a key-value pair mapping relationship according to the feature value, and obtain a sample multidimensional retrieval table; (6) Set update cycle identification and quality assessment scores for the sample multidimensional retrieval table and output a hierarchical seawall sample library.

[0045] Specifically, the quantitative characteristic indicators are hierarchically grouped to establish a three-level hierarchical data structure. The first level is the seawall type, including two major categories: slope type and composite type; the second level is the seawall structure, which is subdivided into continuous slope type and slope-platform-slope segmented type for slope type seawalls, and subdivided into upper straight and lower slope type, straight flat slope type, oblique straight oblique type, oblique flat straight type, upper oblique and lower straight type, and oblique straight flat type for composite seawalls; the third level is the ecological characteristics, which include four aspects: vegetation coverage, greening form on the top of the embankment, ecological properties of the slope protection structure, and surrounding ecosystems. For each seawall sample in the three-level hierarchical data structure, a unique coding rule is designed. The coding consists of three parts: type code (2 digits, such as 01 for slope type and 02 for composite type), structure code (2 digits, such as 01 for continuous slope type and 02 for segmented type in slope type seawalls), and ecological characteristic code (4 digits, representing vegetation coverage level, greening type on the top of the embankment, ecological property level of slope protection, and surrounding ecosystem type, respectively). By concatenating these three parts of the code, an 8-digit sample identification code is formed.

[0046] The generated sample identification code is associated with the multimedia data. The multimedia data includes multi-angle images (high-resolution images taken from different perspectives), oblique photography point cloud data (point clouds that record the three-dimensional structural information of the seawall) and field photos (on-site photos reflecting the actual conditions of the seawall). A data association table is established to establish a mapping relationship between each sample identification code and the corresponding multimedia file path. The association table also records metadata information such as the acquisition time and spatial resolution of the multimedia data. Feature values ​​are extracted from the sample multimedia data set to establish a multi-dimensional retrieval vector. Geographic location features include longitude and latitude coordinates, coastline direction angle, etc.; type features include cross-sectional morphological parameters, slope protection structure parameters, etc.; ecological indicators include vegetation index, ecological permeability score, etc. Each feature value is standardized to ensure a uniform numerical range. In this way, each sample is converted into a multi-dimensional feature vector for subsequent retrieval and matching.

[0047] Convert the extracted feature vectors into an inverted index structure. Inverted index is an efficient retrieval method that establishes a key-value pair mapping for each feature value. The key is the value range of the feature value, and the value is a list of sample identification codes with the feature. For example, for the vegetation coverage feature, multiple value intervals are established as keys, and the value corresponding to each interval is the number of all samples falling within the interval. In this way, a multidimensional retrieval table covering all feature dimensions is constructed. Finally, in order to ensure the timeliness and quality of the sample library, a dynamic maintenance mechanism is introduced to the retrieval table. Set an update cycle identifier to record the entry time and validity period of each sample. Calculate the quality assessment score based on the nature, accuracy and representativeness of the sample data. Clean up or update low-scoring samples regularly to ensure that the sample library always maintains a high-quality reference function.

[0048] For example, a new section of seawall is sampled. This seawall is a composite seawall (type code 02) with an oblique straight oblique structure (structure code 03). After ecological characteristic evaluation, the vegetation coverage is high (1), the top of the embankment is a continuous green belt (1), the slope protection ecology is medium (2), and there is a mangrove ecosystem around it (1), so the ecological characteristic code is 1121. After splicing, the sample identification code 02031121 is obtained. Then the multimedia data of the seawall is collected, including drone images taken at three angles of 0°, 45°, and 90°, point cloud data generated by oblique photogrammetry (point density reaches 50 points / square meter), and 20 field photos reflecting details such as the greening on the top of the embankment and the slope protection structure. When extracting the feature vector, the feature values ​​of multiple dimensions such as geographical location (120 degrees east longitude, 30 degrees north latitude, 45 degrees strike), type characteristics (slope 35 degrees, height 4.5 meters, width 6 meters) and ecological indicators (NDVI value 0.75, permeability score 0.6) were calculated. These feature values ​​are respectively used to establish inverted indexes. For example, the key-value pairs in the NDVI value range of 0.7-0.8 contain the sample number. Finally, a 6-month update cycle is set, and the quality score of 85 points is obtained based on the data quality and accuracy evaluation. Through this processing flow, the information of the seawall section is stored in the sample library in a standardized form.

[0049] In a specific embodiment, the process of executing step S106 may specifically include the following steps: (1) Perform morphological processing on the spatial features in the hierarchical seawall sample library, calculate the curvature value and continuity parameter of the seawall contour line, and obtain the spatial feature similarity matrix; (2) Perform band combination analysis on the spectral features in the hierarchical seawall sample library, calculate the difference values ​​of vegetation coverage and material reflectance, and obtain the spectral feature similarity matrix; (3) Perform texture analysis on the structural features in the hierarchical seawall sample library, calculate the matching degree between the slope protection type and the embankment top morphology, and obtain the structural feature similarity matrix; (4) Normalizing the spatial feature similarity matrix, spectral feature similarity matrix, and structural feature similarity matrix to obtain a comprehensive similarity score; (5) Cluster the sample data according to the comprehensive similarity score, perform feature statistics on the samples in each group, and obtain a type feature statistics table; (6) Structural reorganization of the characteristic values ​​in the type characteristic statistical table is carried out to generate a classification system of seawall ecological characteristics.

[0050] Specifically, in the final stage of the construction of the seawall ecological feature classification system, it is necessary to conduct multi-dimensional similarity analysis and classification integration on the samples in the hierarchical seawall sample library. The spatial features are morphologically processed, and the core of morphological processing is to analyze the geometric characteristics of the seawall contour line. The curvature value reflects the curvature of the seawall trajectory. By calculating the curvature of each point on the trajectory, the curvature sequence is obtained; the continuity parameter reflects the smoothness of the trajectory, which is calculated by analyzing the direction change between adjacent points. For the spatial features of any two samples, the Euclidean distance of their curvature sequence and continuity parameter is calculated to construct a spatial feature similarity matrix. The analysis of spectral features combines the bands of multispectral images. Typical band combinations include: near infrared, red and green band combinations, highlighting vegetation information; short-wave infrared, near infrared and red band combinations, highlighting moisture and material information. By calculating the reflectance characteristics of each band combination, the vegetation coverage index and material reflectance characteristics are obtained. The difference values ​​of these features are calculated for each pair of samples to form a spectral feature similarity matrix. The calculation of difference values ​​needs to consider the weight of each band, as different bands contribute differently to vegetation and material recognition.

[0051] The structural feature analysis focuses on the slope protection type and the top morphology of the embankment. The slope protection types include fence panels, twisted blocks, four-corner hollow squares and other forms, each of which has its own characteristic texture pattern. Texture feature parameters are extracted through the grayscale co-occurrence matrix, including statistics such as energy, contrast, and entropy. The top morphology focuses on the spatial distribution pattern of structures such as green belts and flower beds. These features are converted into feature vectors, and the cosine similarity between vectors is calculated to obtain the structural feature similarity matrix.

[0052] The three similarity matrices are normalized to eliminate the dimensional effects of different feature indicators. The normalization uses the maximum and minimum method to map all similarity values ​​to the interval [0,1]. Then the three normalized matrices are weighted and fused. The weight values ​​are determined according to the importance of each type of feature to the ecological function of the seawall to obtain a comprehensive similarity score. Based on the comprehensive similarity score, the hierarchical clustering algorithm is used to group the samples. First, each sample is treated as an independent category, and then the two categories with the highest similarity are gradually merged until the preset number of categories is reached. In each cluster group, the mean, standard deviation and other statistics of each feature are counted to form a type feature statistical table. The statistical table reflects the typical characteristics and variation range of each category.

[0053] Finally, the statistical table of type characteristics was reorganized to construct a hierarchical classification system. The first layer is the basic type of seawall (slope type, composite type), the second layer is the specific structural form, and the third layer is the combination of ecological characteristics. Each category is equipped with a detailed description of the characteristics and typical examples.

[0054] For example, the sample library of a coastal zone is analyzed, and the characteristics of a section of composite seawall are first extracted. Spatial feature analysis shows that the trajectory of this section of the seawall presents a weak curve shape, the mean of the calculated curvature sequence is 0.002, and the continuity parameter is 0.95, indicating that the direction of the seawall is relatively smooth. In the spectral analysis, the reflectivity of the near-infrared band is 0.45, the red light band is 0.12, and the vegetation index is 0.58, indicating that the greening condition is good; the reflectivity distribution of the short-wave infrared band shows that the slope protection material is mainly concrete. Structural analysis found that the slope protection adopts the combination of upper fence panels and lower riprap. The texture feature parameters show that the energy value of the fence panel area is 0.85, the contrast is 0.45, and the entropy is 0.65. These values ​​are highly consistent with the characteristics of the same type of slope protection in the sample library. The top of the embankment is a continuous green belt structure with uniform width and flower beds distributed at regular intervals. These feature values ​​are compared with other seawall samples in the sample library to calculate the similarity matrix. After cluster analysis, this seawall section was classified into the category of “composite-ecological slope protection-continuous greening”, showing high similarity with other samples in this category in terms of spatial morphology, vegetation coverage and structural characteristics.

[0055] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A remote sensing monitoring classification system for ecological characteristics of seawalls and a method for constructing a sample library, characterized in that: The remote sensing monitoring classification system and sample library construction method of the ecological characteristics of the seawall include: Perform multi-source data fusion processing on high-resolution satellite remote sensing images, aerial remote sensing images, drone orthophotos and field measurement data, obtain digital orthophotos through aerial triangulation, perform plane position and elevation correction on the digital orthophotos according to the control point coordinates, and obtain a standardized multi-source data set; Extracting the slope value, height value and width value of the seawall section according to the standardized multi-source data set, substituting the obtained characteristic parameter values ​​into the preset expert rules for type identification, and generating preliminary seawall type distribution data; Using the preliminary seawall type distribution data, basic geographic information, levee crest ecological information, slope protection structure information and shore ecosystem information are subjected to feature correlation processing to form seawall ecological feature element data; According to the ecological characteristic element data of the seawall, the vegetation coverage of the multispectral image is calculated, the type of greening on the top of the seawall is determined based on the image texture, and the scoring operation is performed based on the slope protection material and ecological permeability to obtain quantitative characteristic indicators; According to the quantitative characteristic indicators, the sample data is standardizedly stored and encoded and multi-dimensional indexes are constructed in a three-level structure to generate a hierarchical seawall sample library; Similarity matching operations are performed on the spatial features, spectral features and structural features in the hierarchical seawall sample library, and samples are updated through dynamic feedback processing to build a seawall ecological feature classification system.

2. The remote sensing monitoring classification system and sample library construction method of seawall ecological characteristics according to claim 1 is characterized in that: The high-resolution satellite remote sensing images, aerial remote sensing images, drone orthophotos and remote sensing images in field measurement data are subjected to data fusion processing, a digital orthophoto map is obtained through aerial triangulation processing, and the plane position and elevation deviation of the digital orthophoto map are corrected according to the control point coordinates to obtain a standardized multi-source data set, including: Resampling the high-resolution satellite remote sensing image bands, inputting the resampled band data into Gram-schmidt mathematical transformation to obtain spectral fusion data; Performing registration processing on the spectral fusion data and the aerial remote sensing image according to the corresponding relationship of geographic coordinates to obtain a first registered image; Perform affine transformation processing on the first registered image and the drone orthophoto according to the same-name ground object points, and obtain a second registered image through bilinear interpolation; Performing image stretching, equalization and filtering enhancement processing on the second registered image using the on-site measurement data to generate enhanced image data; Inputting the enhanced image data into an aerial triangulation program, and obtaining the digital orthophoto map through tie point matching and bundle adjustment operations; Substituting the control point coordinates into the correction equation, performing plane position correction and elevation correction calculation on the digital orthophoto map, and outputting the standardized multi-source data set.

3. The remote sensing monitoring classification system and sample library construction method of seawall ecological characteristics according to claim 1 is characterized in that: The method extracts the slope value, height value and width value of the seawall section according to the standardized multi-source data set, substitutes the obtained characteristic parameter value into the preset expert rules for type identification, and generates preliminary seawall type distribution data, including: Performing edge detection on the seawall image in the standardized multi-source data set, segmenting the seawall area by region growing, and obtaining seawall contour data; The seawall contour data is superimposed on the field measurement data, and the three-dimensional coordinates of each cross-section point are calculated by spatial interpolation to obtain a seawall cross-section point cloud; The seawall cross-section point cloud is cut into sections in a direction perpendicular to the coastline, and the height difference between adjacent cross-section points is divided by the horizontal distance to obtain the slope value; Extracting the elevation difference between the top and bottom points of the seawall from the seawall cross-section point cloud as the height value, and extracting the horizontal distance between the points on both sides of the road at the top of the seawall as the width value; Substituting the slope value, the height value and the width value into a layered threshold discriminant formula, the seawalls are divided into two types: slope type and composite type, and a seawall type discrimination result is obtained; The seawall type identification result is associated with the geographic coordinates through spatial topological relationship, and the preliminary seawall type distribution data is output.

4. The remote sensing monitoring classification system and sample library construction method of seawall ecological characteristics according to claim 1 is characterized in that: The preliminary seawall type distribution data is used to perform feature association processing on basic geographic information, levee top ecological information, slope protection structure information and shore ecosystem information to form seawall ecological feature element data, including: Overlaying the seawall trajectory line in the preliminary seawall type distribution data with the geographic base map, extracting the longitude and latitude coordinates of the starting and ending points of the seawall, the projection length and the strike angle, and obtaining the basic geographic information; Performing band combination analysis on the top area of ​​the seawall in the standardized multi-source data set, separating the spectral characteristics of green plants, flower beds, and hardened ground, and obtaining the ecological information of the top of the seawall; Distinguish the surface material of the slope protection in the preliminary seawall type distribution data by texture analysis, and classify them into fence panels, twisted blocks, four-corner hollow blocks, concrete smooth structures, riprap slope protection, and mortar masonry to obtain the slope protection structure information; Extracting the area around the seawall from the standardized multi-source data set, identifying the distribution range of mangroves, bedrock beaches, sandy beaches, muddy beaches and salt marshes through spectral separation, and obtaining the beach ecosystem information; Combining and associating the basic geographic information, the embankment top ecological information, the slope protection structure information and the shore ecosystem information according to the spatial position correspondence; The associated multi-dimensional feature information is structured and uniformly coded and labeled, and the ecological feature element data of the seawall is output.

5. The remote sensing monitoring classification system and sample library construction method of seawall ecological characteristics according to claim 1 is characterized in that: According to the ecological characteristic element data of the seawall, the vegetation coverage of the multispectral image is calculated, the type of greening on the top of the seawall is determined based on the image texture, and the scoring operation is performed according to the slope protection material and ecological permeability to obtain quantitative characteristic indicators, including: The multispectral image in the ecological characteristic element data of the seawall is subjected to normalized vegetation index calculation, and the vegetation coverage density is quantitatively graded by setting a classification threshold to obtain the vegetation coverage percentage; The ecological information of the dike top is segmented, and the spatial distribution characteristics of the green belt, flower bed and scattered vegetation on the dike top are extracted by edge density and direction consistency calculation to obtain the distribution morphological data of the greening on the dike top; The surface characteristics of different materials in the slope protection structure information are converted into permeability coefficients, and the ecological properties of the slope protection surface structure are quantitatively scored to obtain an ecological permeability score value; Perform area statistics and spatial distribution calculations on various ecosystems in the beach ecosystem information to obtain the distribution ratio of ecosystems around the seawall; Performing data standardization processing on the vegetation coverage percentage, the embankment greening distribution morphology data, the ecological permeability score value and the ecosystem distribution ratio; A weighted sum operation is performed on each standardized index according to its weight, and the quantitative characteristic index is output.

6. The remote sensing monitoring classification system and sample library construction method of seawall ecological characteristics according to claim 1 is characterized in that: The method of performing standardized storage coding and multi-dimensional index construction on the sample data according to the three-level structure based on the quantitative characteristic index to generate a hierarchical seawall sample library includes: The quantitative characteristic indicators are hierarchically grouped, with the seawall type as the first level, the seawall structure as the second level, and the ecological characteristics as the third level, to obtain a three-level hierarchical data structure; Uniquely encode each seawall sample in the three-level hierarchical data structure, and obtain a sample identification code by concatenating the type code, the structure code, and the ecological characteristic code; Associating and binding the sample identification code with the multi-angle images, oblique photography point cloud data, and field photos in the seawall ecological feature element data to obtain a sample multimedia data set; Extracting multi-dimensional feature values ​​of the sample multimedia data set according to geographical location, type characteristics, and ecological indicators to obtain a sample retrieval feature vector; Converting the sample retrieval feature vector into an inverted index structure, establishing a key-value pair mapping relationship according to the feature value, and obtaining a sample multidimensional retrieval table; An update cycle identifier and a quality assessment score are set for the sample multidimensional retrieval table, and the hierarchical seawall sample library is output.

7. The remote sensing monitoring classification system and sample library construction method of seawall ecological characteristics according to claim 1 is characterized in that: The similarity matching operation is performed on the spatial features, spectral features and structural features in the hierarchical seawall sample library, and the samples are updated through dynamic feedback processing to construct a seawall ecological feature classification system, including: Performing morphological processing on the spatial features in the hierarchical seawall sample library, calculating the curvature value and continuity parameter of the seawall contour line, and obtaining a spatial feature similarity matrix; Performing band combination analysis on the spectral features in the hierarchical seawall sample library, calculating the difference values ​​of vegetation coverage and material reflectance, and obtaining a spectral feature similarity matrix; Performing texture analysis on the structural features in the hierarchical seawall sample library, calculating the matching degree between the slope protection type and the embankment top morphology, and obtaining a structural feature similarity matrix; Normalizing the spatial feature similarity matrix, the spectral feature similarity matrix, and the structural feature similarity matrix to obtain a comprehensive similarity score; Clustering and grouping the sample data according to the comprehensive similarity score, performing feature statistics on the samples in each group, and obtaining a type feature statistics table; The characteristic values ​​in the type characteristic statistical table are structurally reorganized to generate the seawall ecological characteristic classification system.

Images