Beach identification and area remote sensing estimation method
By combining the methods of normalized vegetation index, normalized water body index and microwave data, high and low water level line models are generated, and the area of the continental beach is identified and calculated, which solves the problem of inaccurate extraction of water level lines in large spaces, and achieves high-precision The identification and area estimation of the continental beach are suitable for different types of continental beaches. They are highly adaptable and versatile, and help in ecological and environmental protection and resource management.
Patent Information
- Application Number
- CN202411260792.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-10
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-09-10
AI Technical Summary
The prior art is difficult to accurately identify and extract high and low water levels in large spaces, resulting in errors in the identification and area estimation of the continental beaches, especially inaccurate extraction of permanent water bodies in the river bank line and the center of the continental beaches.
By combining normalized vegetation index, normalized water body index, microwave data, and random forest methods, high and low water level line models are generated, and different types of beaches are identified and their area is calculated.
It realizes high-precision beach identification and area estimation in large areas, reduces the cumbersomeness and error of manual measurement, and can obtain beach information in a timely manner. It is suitable for different types of beaches, and has strong adaptability and versatility. , assist in ecological environment protection and resource management.
Smart Images

Figure CN119025974A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of ecology and remote sensing technology, and in particular to a beach identification and area remote sensing estimation method. Background Art
[0002] Beach refers to a landform feature formed on the shoreline of a river or lake or in the middle of a river. It is usually formed by the accumulation of sediments such as sand and silt, and gradually forms with the movement of water. Beaches are very important to the local ecological environment. The beach vegetation growing on them provides food sources and habitats for some aquatic organisms, and the sediments deposited in some areas also provide fertile soil and sufficient arable land for agricultural production.
[0003] Beaches are mainly distributed on the convex banks of lakes and rivers, and the land they form has become an indispensable part of the local ecological environment and human production and life. Beach wetlands are mainly divided into river convex banks, lake shorelines, lake shoreline convexities, and lake center islands or lake center islands according to their location and shape. Different beach types have different development potentials and corresponding ecological protection measures. The microbial communities formed by the lake shoreline convexities and lake center island types have greatly improved the biodiversity of the ecological environment and the self-purification capacity of the water body. The beaches on the lake shorelines or river convex banks are used for agricultural development, and the sediments deposited also provide rich inorganic salts for crops. In addition, the soil will be further deposited and the soil fertility will be increased under the flooding of water bodies during the flood season every year. Therefore, in order to determine the development potential of beaches and implement different ecological protection measures for different types of beaches, we need to obtain specific distribution information of different types of beaches.
[0004] At present, the methods for obtaining the distribution information of beaches mainly include manual surveys and first using remote sensing methods and then manually screening areas. Among them, manual surveys are mainly conducted through manual field surveys. The distribution information of beaches is determined through manual field surveys and then the beach area is counted. The problem with this method is that it is difficult to effectively and real-time reflect the spatiotemporal change information when facing the research area with large temporal and spatial differences. The remote sensing method is currently a widely used method with objective and convenient characteristics, and can obtain information with a large temporal and spatial span. There are many existing remote sensing methods. The more common one is to use the time series method to extract the area where the surface changes between land and water bodies in a year and determine it as a beach. However, this method requires the use of remote sensing to obtain accurate high and low water levels. At present, there is no very suitable model for the research on accurately extracting high and low water levels in large spaces, and then there is no way to correctly and accurately extract the river shoreline, river convex bank and permanent water body in the center of the beach in beach identification.
[0005] In view of the above problems, considering that the high and low water level lines in large space cannot be accurately identified and the shoals in large areas cannot be accurately identified, this method comprehensively considers the high and low water level acquisition methods, the morphological characteristics of various shoals and their characteristics in remote sensing images, and proposes a method for extracting the distribution of shoals by acquiring high and low water level lines and river morphological characteristics through remote sensing, in order to achieve the purpose of synchronously extracting shoals with high precision in a large area. Summary of the invention
[0006] The purpose of the present invention is to provide a method for identifying a shoal and estimating its area using remote sensing, aiming to solve the above problems.
[0007] The present invention provides a method for identifying a shoal and estimating its area by remote sensing, comprising:
[0008] Determine the land surface index based on the normalized vegetation index and the normalized water body index, and obtain the pure water body area based on the land surface index;
[0009] Based on the area of the pure water body, determine the time corresponding to when the pure water body area is maximum and when the pure water body area is minimum, and determine the time to obtain the high water level and the time to obtain the low water level;
[0010] When obtaining the low water level, the random forest method, microwave data, normalized vegetation index and normalized water index are used to obtain the proportion of non-pure pixel water bodies, and the low water level is obtained by using the water level generation model in combination with the proportion of non-pure pixel water bodies and the pure water level line;
[0011] When obtaining the high water mark, a function of microwave data and water body proportion is established, microwave data is used to obtain the water body proportion of non-pure pixels, and the high water mark is obtained by using a water mark generation model in combination with the water body proportion of non-pure pixels and the pure water body water mark;
[0012] The high water level line and low water level line are combined with the beach identification model to obtain the beach category and determine the beach area.
[0013] Preferably, before determining the land surface index according to the normalized vegetation index and the normalized water index, the method further comprises:
[0014] The remote sensing data of the beach to be identified is obtained, and the improved normalized difference water index, normalized water index, normalized vegetation index and backscatter coefficient are determined according to the remote sensing data, specifically:
[0015] The improved normalized difference water index is determined according to the following formula:
[0016]
[0017] Among them, MNDWI represents the improved normalized difference water index, p(G) represents the green band reflectance, and p(MIR) represents the mid-infrared band reflectance;
[0018] The normalized water index is determined according to the following formula:
[0019]
[0020] Among them, NDWI represents the normalized water index, and p(NIR) represents the near-infrared band reflectance;
[0021] The normalized vegetation index is determined according to the following formula:
[0022]
[0023] Among them, NDVI represents the normalized vegetation index, p(R) represents the red band reflectance;
[0024] The backscatter coefficient is determined according to the following formula:
[0025] σ ij =VV ij ×VH ij ;
[0026] Among them, σ ij represents the backscatter coefficient, VV represents the microwave data of vertical transmission and vertical reception polarization, VH represents the microwave data of vertical transmission and horizontal reception polarization, i represents the time sequence number, and j represents the grid sequence number.
[0027] Preferably, determining a land surface index according to a normalized vegetation index and a normalized water body index, and obtaining a pure water body area according to the land surface index includes:
[0028] The surface index is determined according to the following formula:
[0029]
[0030] SI ij represents the surface index, δ represents the empirical threshold of pure water bodies;
[0031] The pure water area is determined according to the following formula:
[0032]
[0033] PWS i represents the pure water area, J represents the number of grids, and b represents the data resolution of microwave data.
[0034] Preferably, determining the time corresponding to when the pure water body area is maximum and when the pure water body area is minimum based on the pure water body area, and determining the time to obtain the high water level line and the time to obtain the low water level line, comprises:
[0035] Get the maximum and minimum values of the pure water area;
[0036] Set the time corresponding to the maximum value as the flood season, and set i=i max , determine i max Get the time for the high water mark;
[0037] Set the time corresponding to the minimum value as the dry season time, and set i = i min , determine i min Get the time for the low water mark.
[0038] Preferably, when obtaining the low water mark, the proportion of non-pure pixel water bodies is obtained by using the random forest method, microwave data, normalized vegetation index and normalized water body index, and the low water mark is obtained by using the water mark generation model in combination with the proportion of non-pure pixel water bodies and the pure water body water mark, including:
[0039] Establish a nonlinear functional relationship between input parameters and water body proportion:
[0040] PW=f RF (C)+ε;
[0041] Among them, PW represents the proportion of water body, C represents the input vector of input parameters, and f RF represents a nonlinear function between input parameters and water body proportion, wherein the input parameters include MNDWI, NDVI, NDWI, p(G), p(R) and microwave data;
[0042] A random forest model was established, and the proportion of non-pure pixel water bodies was determined based on the random forest model; the expression of the random forest model is:
[0043]
[0044] Among them, g(PW|C) represents the integrated decision tree, n represents the number of regression trees, and g i (PW|C) represents a sub-decision tree;
[0045] Get the time at the low water mark, i = i min When , the pure water body distribution is used to obtain the pure water body water level line l0;
[0046] With any point on l0 as the origin, the normal as the y-axis, and the direction pointing to the land as the positive direction, we can establish a coordinate system. Among them, y min Represents the coordinate value of the point on the low water level, y0 means i=i min The coordinate value of the point on the water level of pure water body at time , represents the average water percentage of the pixels adjacent to the point, a represents the data resolution of the random forest model, means i=i min The serial number of the point on the water level of the pure water body;
[0047] Get the low water level for ten consecutive years and determine the average water level of the low water level for ten years as the low water level.
[0048] Preferably, when obtaining the high water mark, a function of microwave data and water body proportion is established, microwave data is used to obtain the non-pure pixel water body proportion, and a water mark generation model is used to obtain the high water mark in combination with the non-pure pixel water body proportion and the pure water body water mark, including:
[0049] Establish the function of backscatter coefficient and water proportion: PW = p(σ j ), where PW represents the proportion of water bodies, j represents the grid number, σ j represents the backscattering coefficient;
[0050] Determine the water percentage of non-pure pixels based on the function of backscatter coefficient and water percentage;
[0051] Get the time at the high water mark, that is, i=i max When , the pure water body distribution is used to obtain the pure water body water level line l1;
[0052] Take any point on l1 as the origin, the normal as the y-axis, and the direction pointing to the land as the positive direction to establish a coordinate system, and we get Among them, y max Represents the coordinate value of the point on the high water mark, y1 represents i=i max The coordinate value of the point on the water level of pure water body at time , represents the average water percentage of the pixels adjacent to the point, b represents the data resolution of the microwave data, and τ represents i=i max The serial number of the point on the water level of the pure water body;
[0053] Get the high water mark of ten consecutive years and determine the average water mark of ten years as the high water mark.
[0054] Preferably, the high water mark and the low water mark are combined with the beach identification model to obtain the beach category and determine the beach area, including:
[0055] When the beach type is a lake center beach, the area of the lake center beach is determined according to the following formula:
[0056]
[0057] Among them, S1 represents the area of the island in the center of the lake, and n represents i=i min The number of the non-pure water pixel in the water body, N means i=i min The total number of non-pure water pixels in the water body, m represents i=i maxThe serial number of the non-pure water body pixel in the water body at time i, M represents i = i max The total number of non-pure water body pixels in the water body at time i, S0 represents the area of the permanent water body surrounded by the shoal of the central island.
[0058] Preferably, by using the high water level line and the low water level line in combination with the shoal recognition model, the shoal category is obtained, and the shoal area is determined, including:
[0059] Using the time series method, the area where the surface type changes between water and land and is not included in the water body in a natural year is extracted and determined as a strip-shaped shoal;
[0060] When the shoal category is a strip-shaped shoal, the width of the strip-shaped shoal is determined according to the low water level line and the high water level line of the strip-shaped shoal
[0061] The area of the strip-shaped shoal is determined according to the following formula:
[0062]
[0063] Among them, S2 represents the area of the strip-shaped shoal, L max represents the length of the high water level line, L min represents the length of the low water level line, represents the width of the strip-shaped shoal.
[0064] Preferably, by using the high water level line and the low water level line in combination with the shoal recognition model, the shoal category is obtained, and the shoal area is determined, including:
[0065] Obtain the radius of curvature of each point on the low water level line, and establish a function of the water level line and the radius of curvature, R = p(x), where R represents the radius of curvature and x represents the length of a certain point on the low water level line from the river source;
[0066] Set the point corresponding to R = 0 and p(x) ≤ r0 as x g ;
[0067] Establish a plane rectangular coordinate system with point x g as the origin, the tangent line of point x g as the x-axis, the counterclockwise direction as the positive direction, the normal line of point x g as the y-axis, and the direction pointing to the bank as the positive direction;
[0068] Fit the boundary of the shoal to a quadratic curve, determine the shoal category as a water convex bank shoal, and set p(x g ) = r0 and the point with the smallest |x dg | has coordinates (x d0 , y d0 ) on the plane rectangular coordinate system, then the quadratic curve equation is:
[0069]
[0070] The area of the convex bank and beach of the water body is determined according to the quadratic curve equation:
[0071]
[0072] Among them, S3 represents the area of the convex bank and beach of the water body, g represents the vertex number, and G represents the maximum value of g.
[0073] Preferably, the high water level line and the low water level line are combined with the beach identification model to obtain the beach category and determine the beach area, including: the beach area is the sum of the area of the center of the lake beach, the area of the strip beach and the area of the convex bank beach of the water body, that is, S=S1+S2+S3.
[0074] Compared with the prior art, the beneficial effect of the present invention is that the present invention utilizes the distribution and morphological characteristics of the beach, extracts the high and low water level lines, combines the remote sensing technology to extract the water body and land surface characteristics, accurately identifies different beach types through remote sensing, and accurately obtains the distribution information of the beach. The present invention uses remote sensing images to comprehensively consider multiple factors such as the water body vegetation index, the surface index of the pure water body, the flood season time, the dry season time, etc., so that the identification result is more comprehensive and reliable. The information of the beach to be identified can be obtained in time, and the real-time monitoring and dynamic analysis of the land area of the beach can be realized. It is suitable for different types of beaches and has strong adaptability and versatility. Through the remote sensing technology of the present invention, the recognition efficiency is greatly improved, the tediousness and errors of manual measurement are reduced; it is helpful to evaluate the ecological environment of the beach and provide a basis for environmental protection; accurately grasp the land area of the beach and is conducive to better resource management and allocation. BRIEF DESCRIPTION OF THE DRAWINGS
[0075] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.
[0076] Figure 1 The present invention is a schematic diagram of the process of a beach identification and area remote sensing estimation method. DETAILED DESCRIPTION
[0077] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0078] like Figure 1 As shown, the present invention provides a method for identifying a shoal and estimating its area by remote sensing, including: determining a surface index according to a normalized vegetation index and a normalized water body index, and obtaining a pure water body area according to the surface index.
[0079] Based on the pure water body area, the time corresponding to when the pure water body area is maximum and when the pure water body area is minimum is determined, and the high water level acquisition time and the low water level acquisition time are determined.
[0080] When obtaining the low water level line, the random forest method, microwave data, normalized vegetation index and normalized water body index are used to obtain the proportion of non-pure pixel water bodies. The low water level line is obtained using the water level generation model combined with the proportion of non-pure pixel water bodies and the pure water body water level line.
[0081] When obtaining the high water mark, a function of microwave data and water body proportion is established, microwave data is used to obtain the water body proportion of non-pure pixels, and the high water mark is obtained using a water mark generation model in combination with the water body proportion of non-pure pixels and the pure water body water mark.
[0082] The high water level line and low water level line are combined with the beach identification model to obtain the beach category and determine the beach area.
[0083] In some embodiments of the present application, before determining the surface index according to the normalized vegetation index and the normalized water index, it also includes: acquiring remote sensing data of the beach to be identified, and determining an improved normalized difference water index, a normalized water index, a normalized vegetation index and a backscatter coefficient according to the remote sensing data, specifically:
[0084] The improved normalized difference water index is determined according to the following formula:
[0085]
[0086] Among them, MNDWI represents the improved normalized difference water index, p(G) represents the green band reflectance, and p(MIR) represents the mid-infrared band reflectance;
[0087] The normalized water index is determined according to the following formula:
[0088]
[0089] Among them, NDWI represents the normalized water index, and p(NIR) represents the near-infrared band reflectance;
[0090] The normalized vegetation index is determined according to the following formula:
[0091]
[0092] Among them, NDVI represents the normalized difference vegetation index, p(R) represents the red band reflectance;
[0093] The backscatter coefficient is determined according to the following formula:
[0094] σ ij =VV ij ×VH ij ;
[0095] Among them, σ ij represents the backscatter coefficient, VV represents the microwave data of vertical transmission and vertical reception polarization, VH represents the microwave data of vertical transmission and horizontal reception polarization, i represents the time sequence number, and j represents the grid sequence number.
[0096] In this embodiment, the microwave data is screened using polarization mode and imaging orbit as ascending orbit. The images are then cropped according to the study area and RefinedLee filtering is used to eliminate certain SAR image coherent speckle noise.
[0097] In some embodiments of the present application, a land surface index is determined according to a normalized vegetation index and a normalized water body index, and a pure water body area is obtained according to the land surface index, including:
[0098] The surface index is determined according to the following formula:
[0099]
[0100] SI ij represents the surface index, δ represents the empirical threshold of pure water bodies;
[0101] The pure water area is determined according to the following formula:
[0102]
[0103] PWS i represents the pure water area, J represents the number of grids, and b represents the data resolution of microwave data.
[0104] In some embodiments of the present application, based on the pure water body area, determining the time corresponding to the maximum and minimum pure water body area, and determining the time to obtain the high water level and the time to obtain the low water level, including: obtaining the maximum and minimum values of the pure water body area; setting the time corresponding to the maximum value as the flood season time, and setting i=i max , determine i max Get the time for the high water mark; set the time corresponding to the minimum value as the dry season time, and set i = i min , determine i min Get the time for the low water mark.
[0105] In this embodiment, the time corresponding to the maximum value of the pure water body area in a year is the flood season, and the time corresponding to the minimum value of the pure water body area in a year is the dry season.
[0106] In some embodiments of the present application, when obtaining the low water level, the proportion of non-pure pixel water bodies is obtained by using the random forest method, microwave data, normalized vegetation index and normalized water body index, and the low water level is obtained by using the water level generation model in combination with the proportion of non-pure pixel water bodies and the pure water body water level, including:
[0107] Establish a nonlinear functional relationship between input parameters and water body proportion:
[0108] PW=f RF (C)+ε;
[0109] Among them, PW represents the proportion of water body, C represents the input vector of input parameters, and f RF represents a nonlinear function between input parameters and water body proportion, wherein the input parameters include MNDWI, NDVI, NDWI, p(G), p(R) and microwave data;
[0110] A random forest model was established, and the proportion of non-pure pixel water bodies was determined based on the random forest model; the expression of the random forest model is:
[0111]
[0112] Among them, g(PW|C) represents the integrated decision tree, n represents the number of regression trees, and g i (PW|C) represents a sub-decision tree;
[0113] Get the time at the low water mark, i = i min When , the pure water body distribution is used to obtain the pure water body water level line l0;
[0114] With any point on l0 as the origin, the normal as the y-axis, and the direction pointing to the land as the positive direction, we can establish a coordinate system. Among them, y minRepresents the coordinate value of the point on the low water level, y0 means i=i min The coordinate value of the point on the water level of pure water body at time , represents the average water percentage of the pixels adjacent to the point, a represents the data resolution of the random forest model, means i=i min The serial number of the point on the water level of the pure water body;
[0115] Get the low water level for ten consecutive years and determine the average water level of the low water level for ten years as the low water level.
[0116] In this embodiment, the existing MNDWI, NDVI, NDWI, p(G), p(R) and microwave data are used as influencing factors that are highly correlated with the water body proportion, and higher-resolution water and non-water body distribution remote sensing data are used to obtain the water body proportion in a grid at a remote sensing data resolution.
[0117] Random forest builds multiple decision trees in the training phase and then calculates the average prediction value of these decision trees as the output of the method. The input feature space is divided into a large number of regression trees, each of which is generated by a bootstrap sample. A bootstrap sample consists of two-thirds of the training sample and the remaining one-third of the data is used to validate each tree.
[0118] Random forest sorting parameters are used to sort the influence of water body proportion, and the top 4 are determined as the key vectors of water body proportion. The key vectors are input into the trained model to obtain the water body proportion map. min When the pure water body distribution is fitted with the pure water body water level line l0, a coordinate system is set with any point on l0 as the origin, the normal as the y-axis and the direction pointing to the land as the positive direction.
[0119] In some embodiments of the present application, when obtaining the high water mark, a function of microwave data and water body proportion is established, microwave data is used to obtain the non-pure pixel water body proportion, and the high water mark is obtained by using a water mark generation model in combination with the non-pure pixel water body proportion and the pure water body water mark, including:
[0120] Establish the function of backscatter coefficient and water proportion: PW = p(σ j ), where PW represents the proportion of water bodies, j represents the grid number, σ j represents the backscattering coefficient;
[0121] Determine the water percentage of non-pure pixels based on the function of backscatter coefficient and water percentage;
[0122] Get the time at the high water mark, that is, i=i max When , the pure water body distribution is used to obtain the pure water body water level line l1;
[0123] Take any point on l1 as the origin, the normal as the y-axis, and the direction pointing to the land as the positive direction to establish a coordinate system, and we get Among them, y max Represents the coordinate value of the point on the high water mark, y1 represents i=i max The coordinate value of the point on the water level of pure water body at time , represents the average water percentage of the pixels adjacent to the point, b represents the data resolution of the microwave data, and τ represents i=i max The serial number of the point on the water level of the pure water body;
[0124] Get the high water mark of ten consecutive years and determine the average water mark of ten years as the high water mark.
[0125] In this embodiment, when obtaining the high water mark, the area is in the flood season, with frequent cloudy and rainy weather, and the accuracy and availability of optical remote sensing data are greatly reduced. Therefore, microwave remote sensing data is used to obtain the water mark, but the resolution of microwave remote sensing data is low. In the field, there are mountains, rocks, vegetation and other materials on the shore. Some pixels on the shore are a combination of water and non-water bodies, and the high water mark cannot be accurately extracted. Therefore, it is necessary to establish a function of the backscatter coefficient and the proportion of water. According to i=i max When the pure water body distribution is fitted with the pure water body water level line l1, a coordinate system is set with any point on l1 as the origin, the normal as the y-axis and the direction pointing to the land as the positive direction.
[0126] In some embodiments of the present application, the high water mark and the low water mark are combined with the beach identification model to obtain the beach category and determine the beach area, including:
[0127] When the beach type is a lake center beach, the area of the lake center beach is determined according to the following formula:
[0128]
[0129] Among them, S1 represents the area of the island in the center of the lake, and n represents i=i min The number of the non-pure water pixel in the water body, N means i=i min The total number of non-pure water pixels in the water body, m represents i=i max The pixel number of the non-pure water body in the water body, M represents i=i max S0 is the total number of non-pure water pixels in the water body, and S0 is the permanent water body area surrounded by the lake center islands and shoals.
[0130] In this embodiment, since there may be a lake island or a lake island in the center of the lake, the surroundings of the lake island, the permanent water body in the center of the lake island, and the surroundings of the lake island are all islands. maxWhen the area surrounded by water bodies has non-pure water pixels, the beach area is in the area i=i min Non-pure water pixel removal i=i max The non-pure water pixel part and the additional pixel at i=i min The pure water part is contained by the non-pure water pixel part contained by the water body.
[0131] In some embodiments of the present application, the high water mark and the low water mark are combined with the beach identification model to obtain the beach category and determine the beach area, including:
[0132] Using the time series method, the area where the land surface type changes between water and land in a natural year and is not contained by water is extracted and determined as a strip of beach;
[0133] When the beach type is a strip beach, the width of the strip beach is determined according to the low water level and high water level of the strip beach.
[0134] The area of the strip beach is determined according to the following formula:
[0135]
[0136] Among them, S2 represents the area of the strip beach, L max Indicates the length of the high water mark, L min Indicates the length of the low water mark. Indicates the width of the strip of beach.
[0137] In this embodiment, the strip-shaped beach is the most common type of beach, which is usually distributed around the water body and formed due to the change of water level in the hydrological rhythm throughout the year. When extracting the strip-shaped beach, the length of the water level line and the width of the beach are used to obtain the area of the strip-shaped beach. The time series method is used to extract the area where the surface type changes between the water body and the land and is not contained by the water body in a natural year. This area is the strip-shaped beach. The distance between the two points where the normal line intersects the high water line at a point on the low water line is the width of the strip-shaped beach. The width of the strip-shaped beach is taken every K distance, and the average value is taken.
[0138] In some embodiments of the present application, the high water mark and the low water mark are combined with the beach identification model to obtain the beach category and determine the beach area, including:
[0139] Obtain the radius of curvature of each point on the low water line, and establish a function of the water line and the radius of curvature, R = p(x), where R represents the radius of curvature and x represents the distance from a point on the low water line to the source of the river;
[0140] When R=0 and p(x)≤r0, the corresponding point is x g ;
[0141] Establish a plane rectangular coordinate system with point x g as the origin, the tangent line of point x g as the x-axis, the counterclockwise direction as the positive direction, the normal line of point x g as the y-axis, and the direction pointing to the shore as the positive direction;
[0142] Fit the boundary of the shoal into a quadratic curve, determine that the shoal category is the convex bank shoal of the water body, and set p(x g ) = r0 and the coordinates of the point with the minimum |x dg | on the plane rectangular coordinate system are (x d0 , y d0 ), then the quadratic curve equation is:
[0143]
[0144] Determine the area of the convex bank shoal of the water body according to the quadratic curve equation as:
[0145]
[0146] where S3 represents the area of the convex bank shoal of the water body, g represents the vertex number, and G represents the maximum value of g.
[0147] In this embodiment, when setting the origin, take the point where R = 0 and p(x) ≤ r0 (r0 is a threshold) as x g , g is the vertex number, and set G as the maximum value of g. Establish a plane rectangular coordinate system with the point at x g as the origin, the tangent line of this point as the x-axis, the counterclockwise direction as the positive direction, the normal line of this point as the y-axis, and the direction pointing to the shore as the positive direction.
[0148] In some embodiments of the present application, using the high water level line and the low water level line in combination with the shoal identification model, obtain the shoal category and determine the shoal area, including: the shoal area is the sum of the area of the mid-lake shoal, the area of the strip-shaped shoal and the area of the convex bank shoal of the water body, that is, S = S1 + S2 + S3.
[0149] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions of the present invention or make equivalent replacements, and these modifications or equivalent replacements cannot make the modified technical solutions deviate from the spirit and scope of the technical solutions of the present invention.
[0150] The system provided in the above embodiment is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the modules or steps in the embodiments of the present invention can be decomposed or combined. For example, the modules in the above embodiment can be combined into one module, or further divided into multiple sub-modules to complete all or part of the functions described above. The names of the modules and steps involved in the embodiments of the present invention are only for distinguishing the modules or steps, and are not regarded as improper limitations of the present invention.
[0151] Those skilled in the art should be able to appreciate that the modules and method steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, computer software or a combination of the two, and the programs corresponding to the software modules and method steps can be placed in random access memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disks, removable disks, CD-ROMs or any other form of storage medium known in the technical field. In order to clearly illustrate the interchangeability of electronic hardware and software, the composition and steps of each example have been generally described in the above description according to the function. Whether these functions are performed in electronic hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.
Claims
1. A method for identifying a shoal and estimating its area by remote sensing, characterized in that: include: Determine the land surface index based on the normalized vegetation index and the normalized water body index, and obtain the pure water body area based on the land surface index; Based on the area of the pure water body, determine the time corresponding to when the pure water body area is maximum and when the pure water body area is minimum, and determine the time to obtain the high water level line and the time to obtain the low water level line; When obtaining the low water level, the random forest method, microwave data, normalized vegetation index and normalized water index are used to obtain the proportion of non-pure pixel water bodies, and the low water level is obtained by using the water level generation model in combination with the proportion of non-pure pixel water bodies and the pure water level line; When obtaining the high water mark, a function of microwave data and water body proportion is established, microwave data is used to obtain the water body proportion of non-pure pixels, and the high water mark is obtained by using a water mark generation model in combination with the water body proportion of non-pure pixels and the pure water body water mark; The high water level line and low water level line are combined with the beach identification model to obtain the beach category and determine the beach area.
2. The method for identifying and estimating the area of a shoal by remote sensing according to claim 1, characterized in that: Before determining the land surface index based on the normalized difference vegetation index and the normalized difference water index, it also includes: The remote sensing data of the beach to be identified is obtained, and the improved normalized difference water index, normalized water index, normalized vegetation index and backscatter coefficient are determined according to the remote sensing data, specifically: The improved normalized difference water index is determined according to the following formula: Among them, MNDWI represents the improved normalized difference water index, p(G) represents the green band reflectance, and p(MIR) represents the mid-infrared band reflectance; The normalized water index is determined according to the following formula: Among them, NDWI represents the normalized water index, and p(NIR) represents the near-infrared band reflectance; The normalized vegetation index is determined according to the following formula: Among them, NDVI represents the normalized difference vegetation index, p(R) represents the red band reflectance; The backscatter coefficient is determined according to the following formula: σ ij !!VV ij ×VH ij 4 Among them, σ ij represents the backscatter coefficient, VV represents the microwave data of vertical transmission and vertical reception polarization, VH represents the microwave data of vertical transmission and horizontal reception polarization, i represents the time sequence number, and j represents the grid sequence number.
3. The method for identifying and estimating the area of a shoal by remote sensing according to claim 2, characterized in that: The land surface index is determined based on the normalized vegetation index and the normalized water body index, and the pure water body area is obtained based on the land surface index, including: The surface index is determined according to the following formula: SI ij represents the surface index, δ represents the empirical threshold of pure water bodies; The pure water area is determined according to the following formula: PWS i represents the pure water area, J represents the number of grids, and b represents the data resolution of microwave data.
4. The method for identifying and estimating the area of a shoal by remote sensing according to claim 3, characterized in that: Determining the time corresponding to when the pure water body area is maximum and when the pure water body area is minimum based on the pure water body area, and determining the time to obtain the high water level line and the time to obtain the low water level line, including: Get the maximum and minimum values of the pure water area; Set the time corresponding to the maximum value as the flood season, and set i=i max , determine i max Get the time for the high water mark; Set the time corresponding to the minimum value as the dry season time, and set i = i min , determine i min Get the time for the low water mark.
5. The method for identifying and estimating the area of a shoal by remote sensing according to claim 4, characterized in that: When obtaining the low water level, the random forest method, microwave data, normalized vegetation index and normalized water index are used to obtain the proportion of non-pure pixel water bodies. The low water level is obtained by combining the proportion of non-pure pixel water bodies and the pure water level using the water level generation model, including: Establish a nonlinear functional relationship between input parameters and water body proportion: PW=f RF (C)+e; Among them, PW represents the proportion of water body, C represents the input vector of input parameters, and f RF represents a nonlinear function between input parameters and water body proportion, wherein the input parameters include MNDWI, NDVI, NDWI, p(G), p(R) and microwave data; A random forest model was established, and the proportion of non-pure pixel water bodies was determined based on the random forest model; the expression of the random forest model is: Among them, g(PW|C) represents the integrated decision tree, n represents the number of regression trees, and g i (PW|C) represents a sub-decision tree; Get the time at the low water mark, i = i min When , the pure water body distribution is used to obtain the pure water body water level line l0; With any point on l0 as the origin, the normal as the y-axis, and the direction pointing to the land as the positive direction, we can establish a coordinate system. Among them, y min Represents the coordinate value of the point on the low water level, y0 means i=i min The coordinate value of the point on the water level of pure water body at time , represents the average water percentage of the pixels adjacent to the point, a represents the data resolution of the random forest model, means i=i min The serial number of the point on the water level of the pure water body; Get the low water level for ten consecutive years and determine the average water level of the ten-year low water level as the low water level.
6. The method for identifying and estimating the area of a shoal by remote sensing according to claim 5, characterized in that: When obtaining the high water mark, a function of microwave data and water body proportion is established, microwave data is used to obtain the water body proportion of non-pure pixels, and the high water mark is obtained by using the water mark generation model in combination with the water body proportion of non-pure pixels and the pure water body water mark, including: Establish the function of backscatter coefficient and water proportion: PW = p(σ j ), where PW represents the proportion of water bodies, j represents the grid number, σ j represents the backscattering coefficient; Determine the water percentage of non-pure pixels based on the function of backscatter coefficient and water percentage; Get the time at the high water mark, that is, i=i max When , the pure water body distribution is used to obtain the pure water body water level line l1; Take any point on l1 as the origin, the normal as the y-axis, and the direction pointing to the land as the positive direction to establish a coordinate system, and we get Among them, y max Represents the coordinate value of the point on the high water mark, y1 represents i=i max The coordinate value of the point on the water level of pure water body at time , represents the average water percentage of the pixels adjacent to the point, b represents the data resolution of the microwave data, and τ represents i=i max The serial number of the point on the water level of the pure water body; Get the high water mark of ten consecutive years and determine the average water mark of ten years as the high water mark.
7. The method for identifying and estimating the area of a shoal by remote sensing according to claim 6, characterized in that: The high water mark and low water mark are combined with the beach identification model to obtain the beach category and determine the beach area, including: When the beach type is a lake center beach, the area of the lake center beach is determined according to the following formula: Among them, S1 represents the area of the island in the center of the lake, and n represents i=i min The number of the non-pure water pixel in the water body, N means i=i min The total number of non-pure water pixels in the water body, m represents i=i max The pixel number of the non-pure water body in the water body, M represents i=i max S0 is the total number of non-pure water pixels in the water body, and S0 is the permanent water body area surrounded by the lake center islands and shoals.
8. The method for identifying and estimating the area of a shoal by remote sensing according to claim 7, characterized in that: The high water mark and low water mark are combined with the beach identification model to obtain the beach category and determine the beach area, including: Using the time series method, the area where the land surface type changes between water and land in a natural year and is not contained by water is extracted and determined as a strip of beach; When the beach type is a strip beach, the width of the strip beach is determined according to the low water level and high water level of the strip beach. The area of the strip beach is determined according to the following formula: Among them, S2 represents the area of the strip beach, L max Indicates the length of the high water mark, L min Indicates the length of the low water mark. Indicates the width of the strip of beach.
9. The method for identifying and estimating the area of a shoal by remote sensing according to claim 8, characterized in that: The high water mark and low water mark are combined with the beach identification model to obtain the beach category and determine the beach area, including: Obtain the radius of curvature of each point on the low water line, and establish a function of the water line and the radius of curvature, R = p(x), where R represents the radius of curvature and x represents the distance from a point on the low water line to the source of the river; When R=0 and p(x)≤r0, the corresponding point is x g ; Create a point x g As the origin, point x g The tangent line of is the x-axis, with the counterclockwise direction as the positive direction, and the point x g The normal of the plane is the y-axis, and the direction pointing to the shore is the positive direction; The boundary of the shoal is fitted to a quadratic curve, and the shoal category is determined as the convex bank shoal of the water body. Let p(x g ) = r0 and the coordinates of the point with the smallest |x dg | on the plane rectangular coordinate system be (x d0 , y d0 ). Then the quadratic curve equation is: The area of the convex bank and beach of the water body is determined according to the quadratic curve equation: Among them, S3 represents the area of the convex bank and beach of the water body, g represents the vertex number, and G represents the maximum value of g.
10. The method for identifying and estimating the area of a shoal by remote sensing according to claim 9, characterized in that: The high water level line and the low water level line are combined with the beach identification model to obtain the beach category and determine the beach area, including: the beach area is the sum of the area of the center of the lake beach, the area of the strip beach and the area of the convex bank beach, that is, S = S1 + S2 + S3.
Citation Information
Patent Citations
Beach vegetation quantity dynamic interpretation method and device based on multi-source remote sensing data
CN115019188A
Method and system for constructing relationship between shallow lake terrain and water level storage capacity by considering influence of high-intensity human activities
CN117726764A
Water body area extraction method based on remote sensing pure water pixels
CN118155091A
