Black and odorous water body identification method based on remote sensing spectrum

Abstract

The present invention provides a method for identifying black and odorous water bodies based on remote sensing spectra, which relates to the technical field of environmental monitoring, including: obtaining multispectral / hyperspectral satellite images of black and odorous water bodies and the attitude and position information when satellites shoot images; identifying water body images of black and odorous water body areas according to water body characteristic band data in satellite images; calculating the shooting angle information of each pixel in the water body image according to the attitude and position information when satellites shoot images; evaluating the interference degree of the shooting angle information on the quality of black and odorous water body images; detecting the surrounding images of black and odorous water body areas, and classifying the surrounding images; and correcting the spectral data of water body images of black and odorous water body areas according to the evaluation results and the results of the ground object classification. The present invention utilizes multispectral and hyperspectral satellite images, combined with accurate satellite position information, to propose targeted interference analysis and correction methods, thereby achieving high-precision black and odorous water body identification and interference elimination.

Description

Technical Field

[0001] The invention relates to the technical field of environmental monitoring, and in particular to a method for identifying black and odorous water bodies based on remote sensing spectroscopy. Background Art

[0002] In the identification and monitoring of black and odorous water bodies, spectral analysis is an important non-destructive detection method. The current technology uses vertical satellite images for analysis and determines whether it is a black and odorous water body through spectral data. However, when using satellites to identify black and odorous water bodies, during the satellite identification process (tilted), the reflection of plants on the shore will interfere with the water body spectrum, causing the problem of water surface reflection interference, affecting the accurate judgment of the water body.

[0003] The existing methods for identifying black and odorous water bodies mainly have the following problems: (1) Inadequate spectral interference processing: The existing water body identification methods do not adequately process the spectral interference of surrounding vegetation and buildings, which can easily lead to misidentification and reduced recognition accuracy. Especially in complex urban environments, the spectral mixing problem is obvious, making it difficult to accurately distinguish water bodies from other landforms. (2) The impact of shooting angle is not fully considered: Current technologies often ignore the impact of satellite shooting angles on image quality in water body identification, and fail to fully utilize angular information such as pitch, yaw and roll angles for interference analysis and correction. Different shooting angles will cause changes in shadows, reflections and other effects in the water area, affecting the accuracy of identification. (3) Inadequate intelligent processing of environmental complexity: For the complexity of the surrounding environment of water bodies, such as dense distribution of vegetation and buildings, existing technologies lack intelligent and fine-grained analysis methods. The accuracy and efficiency of traditional classification algorithms need to be improved when dealing with complex environments. (4) Low degree of automation: In the existing image processing process, many steps still require manual intervention, and the overall degree of automation is low, affecting processing efficiency and the stability of results. The inadequacy of automated algorithms and processes has limited large-scale applications. Summary of the invention

[0004] In view of this, an embodiment of the present application provides a method for identifying black and odorous water bodies based on remote sensing spectroscopy. The method utilizes multispectral and hyperspectral satellite images, combined with precise satellite position information, analyzes the shooting angle of satellite images, evaluates the impact of light incidence and shadow effects on image quality, and proposes targeted interference analysis and correction methods to achieve high-precision black and odorous water body identification and interference elimination.

[0005] The embodiment of the present application provides the following technical solution: a method for identifying black and odorous water bodies based on remote sensing spectroscopy, comprising:

[0006] Obtain multispectral / hyperspectral satellite images of black and odorous water bodies and the attitude and position information when the satellite images are taken;

[0007] Extracting water body information from satellite images based on water body characteristic band data in the multispectral / hyperspectral satellite images, and identifying water body images in black and odorous water areas;

[0008] According to the attitude and position information of the satellite when taking the image, the shooting angle data of the water body image of the black and odorous water body area is parsed, the shooting angle data is combined with the geographic coordinate system, and the shooting angle information of each pixel in the water body image of the black and odorous water body area is calculated;

[0009] Analyze the influence of the shooting angle information of each pixel in the water body image of the black and odorous water body area on the black and odorous water body image, and evaluate the interference degree of the shooting angle information on the quality of the black and odorous water body image;

[0010] Detecting the surrounding images of the black and odorous water area according to the multispectral / hyperspectral data in the multispectral / hyperspectral satellite image, and classifying the surrounding images of the black and odorous water area;

[0011] According to the evaluation result of the interference degree of the shooting angle information on the quality of the black and odorous water body image, and the ground object classification result of the surrounding image of the black and odorous water body area, the spectral data of the water body image of the black and odorous water body area is corrected.

[0012] According to one embodiment of the present application, the attitude and position information of the satellite when taking images includes pitch angle, yaw angle, roll angle, satellite orbit altitude information and position coordinates of the satellite in the geocentric coordinate system, as well as image metadata.

[0013] According to an embodiment of the present application, based on the water body characteristic band data in the multispectral / hyperspectral satellite image, a water body index algorithm is used to extract water body information in the satellite image, generate a water body mask, identify the water body image of the black and smelly water area, and mark it.

[0014] According to an embodiment of the present application, the shooting angle data is combined with a geographic coordinate system to calculate the shooting angle information of each pixel in the water body image of the black and odorous water body area, including:

[0015] Establishing a mapping relationship between a geographic coordinate system and a pixel coordinate system, mapping the position coordinates of each pixel in the water body image of the black and odorous water body area to geographic coordinates;

[0016] Convert the geographic coordinates of each pixel in the water body image into three-dimensional coordinates in the geocentric coordinate system;

[0017] According to the three-dimensional coordinates of each pixel in the water body image in the geocentric coordinate system, the shooting angle information of each pixel in the water body image of the black and smelly water body area is calculated.

[0018] According to an embodiment of the present application, the shooting angle information of each pixel in the water body image of the black and odorous water body area is calculated according to the three-dimensional coordinates of each pixel in the water body image in the geocentric coordinate system, including:

[0019] According to the three-dimensional coordinates of each pixel in the water body image in the geocentric coordinate system, the position vector of each pixel is determined, and according to the position coordinates of the satellite in the geocentric coordinate system, the position vector of the satellite is determined;

[0020] Calculate the sight vector of each pixel in the water body image according to the position vector of each pixel and the position vector of the satellite;

[0021] According to the sight line vector and the ground normal vector of each pixel, the shooting angle information of each pixel in the water body image of the black and smelly water body area is calculated.

[0022] According to an embodiment of the present application, calculating and obtaining the sight vector of each pixel in the water body image also includes:

[0023] Construct an attitude rotation matrix, and correct the sight line vector of each pixel according to the attitude rotation matrix to obtain a corrected sight line vector; calculate and obtain the shooting angle information of each pixel in the water image of the black and odorous water area according to the corrected sight line vector of each pixel and the ground normal vector.

[0024] According to an embodiment of the present application, the influence of the shooting angle information of each pixel in the water body image of the black and odorous water body area on the black and odorous water body image is analyzed, wherein the influence of the shooting angle information of each pixel on the black and odorous water body image includes the influence of the light incident angle and the shadow area, including:

[0025] Calculate the incident angle and observation angle of each pixel. If the following conditions are met, it is judged that the image quality of black and odorous water bodies is affected by the incident angle of light;

[0026] (180±5)°; where is the incident angle, is the observation angle, is the angle tolerance, is the azimuth of the sun, is the azimuth of the sensor;

[0027] If the incident angle exceeds a set threshold, it is determined that the pixel corresponding to the incident angle is in a shadow area, and the quality of the water body image is affected by the shadow area.

[0028] According to an embodiment of the present application, the interference degree of the shooting angle information on the image quality of the black and odorous water body is evaluated, including:

[0029] According to the calculated incident angle, observation angle of each pixel, and the azimuth angle difference between the sun and the sensor, the specular reflection interference function, shadow interference function and atmospheric interference function are calculated;

[0030] Set the interference degree of the shooting angle information on the image quality of black and odorous water bodies to score D: ;in, , , are weight coefficients respectively, satisfying =1; is the specular reflection interference function, is the shadow interference function, is the atmospheric disturbance function;

[0031] According to the interference degree score D, the quality of the black and odorous water body image is graded, and the spectral data of the black and odorous water body image that needs to be corrected is determined.

[0032] According to an embodiment of the present application, based on the evaluation result of the interference degree of the shooting angle information on the quality of the black and odorous water body image and the ground object classification result of the surrounding image of the black and odorous water body area, the spectral data of the water body image of the black and odorous water body area is corrected, including:

[0033] A spectral correction algorithm of least squares matched filtering is adopted to determine the type of interference to the corresponding pixel points according to the evaluation results and the ground object classification results, and to correct the spectral data of the interfered black and smelly water body image.

[0034] According to an embodiment of the present application, the method further includes:

[0035] Determine the ground sampling points, obtain the ground measured spectral data of the ground sampling points, and verify the spectral data of the corrected black and odorous water body image based on the ground measured spectral data.

[0036] Compared with the prior art, the beneficial effects that can be achieved by at least one of the above technical solutions adopted in the embodiments of this specification include at least the following: The embodiments of the present invention achieve high-precision water body identification through comprehensive processing of multi-source satellite images and multispectral / hyperspectral analysis, and can accurately distinguish water bodies from surrounding objects, especially in complex environments, thereby improving the accuracy of black and odorous water body identification. At the same time, based on the comprehensive analysis of the shooting angle and the surrounding environment, the embodiments of the present invention can accurately identify and eliminate the spectral interference caused by vegetation, buildings, etc., improve the authenticity and reliability of image data, effectively eliminate environmental interference, and further ensure the high-precision identification of black and odorous water bodies. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0038] Figure 1 It is a schematic flow chart of a method for identifying black and odorous water bodies based on remote sensing spectroscopy according to an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0040] The following describes the implementation methods of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The present application can also be implemented or applied through other different specific implementation methods, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that, in the absence of conflict, the following embodiments and the features in the embodiments can be combined with each other. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work belong to the scope of protection of the present application.

[0041] like Figure 1 As shown, the embodiment of the present invention provides a method for identifying black and odorous water bodies based on remote sensing spectroscopy, comprising:

[0042] S101. Acquire multispectral / hyperspectral satellite images of black and odorous water bodies and the attitude and position information of satellite images;

[0043] S102. Extracting water information from satellite images based on water feature band data in the multispectral / hyperspectral satellite images, and identifying water images of black and odorous water areas;

[0044] S103. According to the attitude and position information of the satellite when taking the image, the shooting angle data of the water body image of the black and odorous water body area is analyzed, the shooting angle data is combined with the geographic coordinate system, and the shooting angle information of each pixel in the water body image of the black and odorous water body area is calculated;

[0045] S104. Analyze the influence of the shooting angle information of each pixel in the water body image of the black and odorous water body area on the black and odorous water body image, and evaluate the interference degree of the shooting angle information on the quality of the black and odorous water body image;

[0046] S105. Detecting the surrounding images of the black and odorous water area according to the multispectral / hyperspectral data in the multispectral / hyperspectral satellite image, and classifying the surrounding images of the black and odorous water area;

[0047] S106. Correct the spectral data of the water body image of the black and odorous water body area according to the evaluation result of the interference degree of the shooting angle information on the quality of the black and odorous water body image and the object classification result of the surrounding image of the black and odorous water body area.

[0048] In the specific implementation of the embodiment of the present invention, in step S101, the multispectral / hyperspectral satellite image of the black and odorous water body and the attitude and position information of the satellite image are first obtained. Specifically, it includes:

[0049] 1. Satellite image acquisition

[0050] Use multispectral or hyperspectral imaging remote sensing technology to obtain high-resolution satellite images. The satellite image data should contain multiple band information for subsequent water body identification and interference elimination. (Gaofen satellite series)

[0051] 2. Obtaining attitude and position information when satellites take images

[0052] At the same time, the precise location information of the satellite when taking the image is obtained, including longitude, latitude, satellite orbit altitude, pitch angle, yaw angle, roll angle, and the satellite's position coordinates in the geocentric coordinate system, as well as image metadata, etc. The relevant information is recorded and stored in real time.

[0053] In the specific implementation of the embodiment of the present invention, in step S102, preliminary analysis is performed based on the water body characteristic band data in the multispectral / hyperspectral satellite image, such as blue, green and near-infrared bands. A water body index algorithm, such as NDWI (Normalized Difference Water Index), is used to extract water body information in the satellite image, generate a water body mask, identify the water body image of the black and odorous water area, and mark it for subsequent processing.

[0054] During the specific implementation of the embodiment of the present invention, in step S103, the shooting angle data of the water image of the black and odorous water area is parsed according to the posture and position information of the satellite when taking the image, and the shooting angle data is combined with the geographic coordinate system to calculate the shooting angle information of each pixel in the water image of the black and odorous water area.

[0055] In this embodiment, the shooting angle data is combined with the geographic coordinate system to calculate the shooting angle information of each pixel in the water body image of the black and odorous water body area, including: establishing a mapping relationship between the geographic coordinate system and the pixel coordinate system, mapping the position coordinates of each pixel in the water body image of the black and odorous water body area to the geographic coordinates; converting the geographic coordinates of each pixel in the water body image into three-dimensional coordinates in the geocentric coordinate system; and calculating the shooting angle information of each pixel in the water body image of the black and odorous water body area according to the three-dimensional coordinates of each pixel in the water body image in the geocentric coordinate system.

[0056] The method of calculating the shooting angle information of each pixel in the water body image of the black and odorous water body area according to the three-dimensional coordinates of each pixel in the water body image in the geocentric coordinate system includes:

[0057] According to the three-dimensional coordinates of each pixel in the water body image in the geocentric coordinate system, the position vector of each pixel is determined, and according to the position coordinates of the satellite in the geocentric coordinate system, the position vector of the satellite is determined; according to the position vector of each pixel and the position vector of the satellite, the line of sight vector of each pixel in the water body image is calculated; according to the line of sight vector of each pixel and the ground normal vector, the shooting angle information of each pixel in the water body image of the black and smelly water area is calculated.

[0058] In another embodiment, calculating the sight vector of each pixel in the water body image also includes: constructing a posture rotation matrix, correcting the sight vector of each pixel according to the posture rotation matrix, and obtaining a corrected sight vector; and calculating the shooting angle information of each pixel in the water body image of the black and smelly water area according to the corrected sight vector of each pixel and the ground normal vector.

[0059] In the specific implementation of step S103 of this embodiment, the following steps are included:

[0060] 1. Satellite image angle data analysis

[0061] The satellite position information is used to analyze the shooting angle data of the image, including the pitch angle, yaw angle and roll angle. The angle data is combined with the geographic coordinate system to calculate the specific angle information of each pixel in the image. The following is a detailed description of the angle data calculation of the embodiment of the present invention:

[0062] (1) Data preparation

[0063] Satellite attitude data acquisition: Get the attitude information at the time of shooting from the satellite telemetry data, including:

[0064] Pitch angle ( ): The satellite's rotation angle in the direction perpendicular to the orbital plane.

[0065] Yaw angle ( ): The rotation angle of the satellite around the vertical axis.

[0066] Roll Angle ( ): The rotation angle of the satellite around the forward direction axis.

[0067] Satellite orbit and position data: Get the satellite's orbital height H at the time of shooting, and its position in the geocentric coordinate system ( , , ).

[0068] Image metadata: Get basic information about the image, such as image resolution, size, and the geographic coordinates (longitude, latitude) corresponding to each pixel.

[0069] (2) Establishing the mapping between geographic coordinate system and pixel coordinate system

[0070] Geographic coordinate system (WGS84): A universal geographic coordinate system is used, expressed in longitude λ, latitude φ and altitude h.

[0071] Pixel coordinate system: The pixel position in the image is represented by row and column numbers (i, j).

[0072] Mapping relationship establishment: Through geometric correction and orthorectification, each pixel position (i, j) is mapped to geographic coordinates (λ, φ).

[0073] (3) Calculate the sight vector for each pixel

[0074] The geographic coordinates are converted to geocentric coordinates. For each pixel in the image, the geographic coordinates (λ, φ, h) are converted to three-dimensional coordinates in the geocentric coordinate system ( , , ).

[0075] Conversion formula:

[0076]

[0077]

[0078]

[0079]

[0080] Where, a: length of the Earth’s major axis, m; e: 2 : The square of the first eccentricity (about 0.00669437999014); h: The height from the ground, which can be set to 0 if unknown.

[0081] Calculate the view vector:

[0082] Satellite position vector: S = ( , , );

[0083] Target point position vector: G=( , , );

[0084] View vector: V -S

[0085] (4) Correction of the sight vector considering satellite attitude

[0086] Construct the posture rotation matrix: Rotation matrix around the X axis:

[0087]

[0088] Rotation matrix around the Y axis:

[0089]

[0090] Rotation matrix around the Z axis:

[0091]

[0092] The total rotation matrix: + +

[0093] Corrected sight vector:

[0094] The ground normal vector is:

[0095]

[0096] Angle of incidence ( ): The angle between the sight vector and the ground normal vector:

[0097]

[0098] Calculate the east component of the sight vector , North component . Azimuth ( ): The angle between the projection of the sight vector and the true north direction:

[0099]

[0100] In the specific implementation of the embodiment of the present invention, in step S104, the influence of the shooting angle information of each pixel in the water body image of the black and odorous water body area on the black and odorous water body image is analyzed, wherein the influence of the shooting angle information of each pixel on the black and odorous water body image includes the influence of the light incident angle and the shadow area, including:

[0101] Calculate the incident angle and observation angle of each pixel. If the following conditions are met, it is judged that the image quality of black and odorous water bodies is affected by the incident angle of light;

[0102] (180±5)°; where is the incident angle, is the observation angle, is the angle tolerance, is the azimuth of the sun, is the azimuth of the sensor;

[0103] If the incident angle exceeds a set threshold, it is determined that the pixel corresponding to the incident angle is in a shadow area, and the quality of the water body image is affected by the shadow area.

[0104] In the specific implementation, the influence of the shooting angle on the water body image is analyzed, including the light incident angle and the shadow effect. Combined with the geographical location and terrain data of the water body, the interference degree of the shooting angle on the image quality is evaluated. The evaluation method of the embodiment of the present invention is described in detail below:

[0105] (1) Data preparation

[0106] Satellite attitude and position data: Get the satellite’s attitude information at the time of shooting, including the pitch angle ( )、yaw angle( ) and the roll angle ( ), and the satellite's position coordinates ( , , ).

[0107] Sun position data: Get the sun's azimuth at the time of shooting ( ) and altitude angle ( ), used to analyze lighting conditions and shadow effects.

[0108] Digital Elevation Model (DEM): Obtain terrain elevation data of the study area to consider the impact of terrain undulation on the incident and observation angles.

[0109] Satellite imagery data: High-resolution images that have been geometrically and atmospherically corrected to ensure spatial and spectral accuracy.

[0110] (2) Calculate the incident angle and observation angle of each pixel

[0111] Ground normal vector calculation: For each pixel, the ground normal vector is calculated based on the DEM If the terrain is flat, the ground normal vector can be approximated as pointing vertically upward.

[0112] Solar vector calculation: Based on the azimuth of the sun and altitude angle , calculate the sun vector S:

[0113]

[0114] Angle of incidence calculate:

[0115]

[0116] Observation angle calculation (angle between the sensor line of sight and the ground normal): The sensor line of sight vector V for each pixel has been calculated in the previous step (shooting angle analysis).

[0117] Observation Angle calculate:

[0118]

[0119] In the specific implementation of the embodiment of the present invention, in step S104, the interference degree of the shooting angle information on the image quality of the black and odorous water body is evaluated, including:

[0120] According to the calculated incident angle, observation angle of each pixel, and the azimuth angle difference between the sun and the sensor, the specular reflection interference function, shadow interference function and atmospheric interference function are calculated;

[0121] Set the interference degree of the shooting angle information on the image quality of black and odorous water bodies to score D: ;in, , , are weight coefficients respectively, satisfying =1; is the specular reflection interference function, is the shadow interference function, is the atmospheric disturbance function;

[0122] According to the interference degree score D, the quality of the black and odorous water body image is graded, and the spectral data of the black and odorous water body image that needs to be corrected is determined.

[0123] In specific implementation, the mirror reflection effect: when mirror reflection occurs on the surface of the water, it will cause the sensor to receive excessively high reflection signals, affecting spectral analysis.

[0124] Condition judgment: If the following conditions are met, there may be mirror reflection interference.

[0125]

[0126] , : are the azimuths of the sun and the sensor respectively;

[0127] : Angle tolerance, generally set to 5°;

[0128] Shadow effect: When the incident angle is large or the terrain is blocked, a shadow area will be generated, resulting in missing or distorted spectral information. Shadow determination: If the incident angle exceeds a certain threshold (for example, 85°), the pixel is considered to be in the shadow area.

[0129] Atmospheric scattering and absorption: Larger observation angles may increase the atmospheric path length, leading to enhanced scattering and absorption.

[0130] (4) Setting evaluation criteria

[0131] 1. Interference degree classification: According to the calculated incident angle, observation angle and azimuth angle difference, set the interference degree score D:

[0132]

[0133] , , : Weight coefficient, satisfying The weight coefficient is set according to the degree of influence of interference factors on target detection. In water body recognition, specular reflection may cause water body highlighting and affect spectral characteristics, so it is given a higher weight. Shadows may cause the reflectivity of water bodies to decrease and the spectral characteristics to change. Atmospheric interference is only obvious when the observation angle is large, so the weight is relatively small.

[0134] : Specular reflection interference function;

[0135] : Shadow interference function;

[0136] : atmospheric interference function;

[0137]

[0138] is the shadow threshold angle, for example 85°.

[0139]

[0140] is the maximum observation angle, generally set to 30°.

[0141] 2. Determination of interference degree

[0142] The interference degree score D ranges from [0, 1], and a larger value indicates more severe interference.

[0143] According to the value of D, the pixels are divided into the following levels:

[0144] Low interference (0≤D<0.3): The image quality is good and can be used directly for analysis.

[0145] Moderate interference (0.3≤D<0.7): The image may have some interference and requires spectral correction.

[0146] High interference (0.7≤D≤1): The image has serious interference and needs to be used with caution or eliminated.

[0147] In the specific implementation of step S105 of this embodiment, multispectral / hyperspectral data in satellite images are used to detect vegetation and buildings around water bodies. Classification algorithms, such as support vector machines (SVM) or random forests (RandomForest), are used to classify objects in the images. It is analyzed to determine which pixels are plants or building shadows.

[0148] In step S106 of this embodiment, the spectral data of the water body image of the black and odorous water body area is corrected according to the evaluation result of the interference degree of the shooting angle information on the quality of the black and odorous water body image, and the object classification result of the surrounding image of the black and odorous water body area, including: using the spectral correction algorithm of least squares matched filtering, judging the type of interference suffered by the corresponding pixel point according to the evaluation result and the object classification result, and correcting the spectral data of the interfered black and odorous water body image.

[0149] In the specific implementation, the spectral correction algorithm of Least Squares Matching Filter (LSMF) is used. According to the evaluation results of the shooting angle and the surrounding environment, it is determined what kind of interference the pixel may be subjected to, and classified as plant interference or building interference. The information of different types of reflection spectra is found in the database, the color of the building is determined, and the spectral data of the disturbed water image is corrected to eliminate the spectral interference of vegetation and buildings. The specific formula is as follows:

[0150]

[0151] Among them, I measured (x,y) is the measured spectral data, a i is the interference coefficient, R i (λ) is the interference spectrum.

[0152] In one embodiment of the present invention, the method further includes: determining ground sampling points, obtaining ground measured spectral data of the ground sampling points, and verifying the spectral data of the corrected black and odorous water body image based on the ground measured spectral data.

[0153] In the specific implementation, the spectral unmixing technology is used to decompose the mixed spectral signal into pure spectral components to improve the accuracy of water body identification. The corrected image is verified to ensure that the spectral data of the water body area accurately reflects the actual situation. The verification method of the embodiment of the present invention is described in detail below:

[0154] (1) Obtaining ground-truth data or reference spectra

[0155] Carry out on-site data collection to ensure that the ground sampling time is consistent or close to the satellite image shooting time. Select a representative water area and record the precise GPS coordinates. Obtain the spectral reflectance and water quality indicators (such as suspended matter, chlorophyll concentration, etc.) of the water body. Establish a reference spectral library to collect standard spectral curves of pure water bodies and common interferences (vegetation, buildings, etc.). The spectral library should cover the band range involved in the image.

[0156] (2) Extracting the corresponding spectral data from the image

[0157] Using GPS coordinates, the ground sampling points are located on the corrected image. The spectral data of the corresponding pixels are extracted. Considering the spatial resolution of the satellite image, a certain range of pixels around the sampling point can be taken to calculate the average spectral value to reduce random errors.

[0158] (3) Spectral comparison and analysis

[0159] Plot the ground truth spectrum and the corrected image spectrum in the same coordinate system. Visually check whether the shape and peak position of the spectrum curve are consistent.

[0160] Calculate the error metric: Root Mean Square Error (RMSE)

[0161]

[0162] : the reflectivity of the i-th band measured on the ground;

[0163] : reflectance of the i-th band of the corrected image;

[0164] n: number of bands;

[0165] Correlation coefficient:

[0166]

[0167] : Average value of reflectivity

[0168] Spectral Angle Matching:

[0169]

[0170] : Spectral angle, in radians. The smaller it is, the higher the spectral match.

[0171] (4) Spatial consistency verification

[0172] Calculate the spectral mean and standard deviation for all pixels in the water area. Check if the standard deviation is within a reasonable range; a large standard deviation may indicate noise or improper calibration. Map the spatial distribution of a specific band or spectral index (such as NDWI). Observe if there are abnormally high or low values ​​in the water area.

[0173] (5) Verification result evaluation: Set the multi-level evaluation indicators of RMSE, correlation coefficient, and spectral angle according to the situation. If most indicators reach "low error" or "high correlation", the correction effect is considered to be good. If there are many "high errors" or "low correlations", it is necessary to readjust the spectral correction parameters or methods.

[0174] The embodiment of the present invention utilizes multispectral and hyperspectral satellite images, combined with precise satellite position information, analyzes the shooting angle of satellite images, evaluates the impact of light incidence and shadow effects on image quality, and proposes targeted interference analysis and correction methods, thereby achieving high-precision black and odorous water body identification and interference elimination.

[0175] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for identifying black and odorous water bodies based on remote sensing spectroscopy, characterized in that: include: Obtain multispectral or hyperspectral satellite images of black and odorous water bodies and the attitude and position information when satellite images are taken; Extracting water body information from satellite images based on water body characteristic band data in the multispectral or hyperspectral satellite images, and identifying water body images in black and odorous water areas; According to the attitude and position information of the satellite when taking the image, the shooting angle data of the water body image of the black and odorous water body area is parsed, the shooting angle data is combined with the geographic coordinate system, and the shooting angle information of each pixel in the water body image of the black and odorous water body area is calculated; Analyze the influence of the shooting angle information of each pixel in the water body image of the black and odorous water body area on the black and odorous water body image, and evaluate the interference degree of the shooting angle information on the quality of the black and odorous water body image; Detecting the surrounding images of the black and odorous water area according to the multispectral or hyperspectral data in the multispectral or hyperspectral satellite image, and classifying the surrounding images of the black and odorous water area; According to the evaluation result of the interference degree of the shooting angle information on the quality of the black and odorous water body image, and the ground object classification result of the surrounding image of the black and odorous water body area, the spectral data of the water body image of the black and odorous water body area is corrected.

2. The black and odorous water body identification method based on remote sensing spectrum according to claim 1 is characterized in that: The attitude and position information of the satellite when taking images includes pitch angle, yaw angle, roll angle, satellite orbit height information and position coordinates of the satellite in the geocentric coordinate system, as well as image metadata.

3. The black and odorous water body identification method based on remote sensing spectrum according to claim 1 is characterized in that: According to the water body characteristic band data in the multispectral or hyperspectral satellite image, a water body index algorithm is used to extract the water body information in the satellite image, generate a water body mask, identify the water body image of the black and odorous water area and mark it.

4. The black and odorous water body identification method based on remote sensing spectrum according to claim 2 is characterized in that: The shooting angle data is combined with the geographic coordinate system to calculate the shooting angle information of each pixel in the water body image of the black and odorous water body area, including: Establishing a mapping relationship between a geographic coordinate system and a pixel coordinate system, mapping the position coordinates of each pixel in the water body image of the black and odorous water body area to geographic coordinates; Convert the geographic coordinates of each pixel in the water body image into three-dimensional coordinates in the geocentric coordinate system; According to the three-dimensional coordinates of each pixel in the water body image in the geocentric coordinate system, the shooting angle information of each pixel in the water body image of the black and smelly water body area is calculated.

5. The black and odorous water body identification method based on remote sensing spectrum according to claim 4 is characterized in that: According to the three-dimensional coordinates of each pixel in the water body image in the geocentric coordinate system, the shooting angle information of each pixel in the water body image of the black and odorous water body area is calculated, including: According to the three-dimensional coordinates of each pixel in the water body image in the geocentric coordinate system, the position vector of each pixel is determined, and according to the position coordinates of the satellite in the geocentric coordinate system, the position vector of the satellite is determined; Calculate the sight vector of each pixel in the water body image according to the position vector of each pixel and the position vector of the satellite; According to the sight line vector and the ground normal vector of each pixel, the shooting angle information of each pixel in the water body image of the black and smelly water body area is calculated.

6. The method for identifying black and odorous water bodies based on remote sensing spectroscopy according to claim 5 is characterized in that: Calculate the sight vector of each pixel in the water body image, including: Construct an attitude rotation matrix, and correct the sight line vector of each pixel according to the attitude rotation matrix to obtain a corrected sight line vector; calculate and obtain the shooting angle information of each pixel in the water image of the black and odorous water area according to the corrected sight line vector of each pixel and the ground normal vector.

7. The method for identifying black and odorous water bodies based on remote sensing spectroscopy according to claim 1 is characterized in that: Analyze the influence of the shooting angle information of each pixel in the water body image of the black and odorous water body area on the black and odorous water body image, wherein the influence of the shooting angle information of each pixel on the black and odorous water body image includes the influence of the light incident angle and the shadow area, including: Calculate the incident angle and observation angle of each pixel. If the following conditions are met, it is judged that the image quality of black and odorous water bodies is affected by the incident angle of light; (180±5)°; where is the incident angle, is the observation angle, is the angle tolerance, is the azimuth of the sun, is the azimuth of the sensor; If the incident angle exceeds a set threshold, it is determined that the pixel corresponding to the incident angle is in a shadow area, and the quality of the water body image is affected by the shadow area.

8. The method for identifying black and odorous water bodies based on remote sensing spectroscopy according to claim 7 is characterized in that: The interference degree of the shooting angle information on the image quality of black and odorous water bodies is evaluated, including: According to the calculated incident angle, observation angle of each pixel, and the azimuth angle difference between the sun and the sensor, the specular reflection interference function, shadow interference function and atmospheric interference function are calculated; Set the interference degree of the shooting angle information on the image quality of black and odorous water bodies to score D: ;in, , , are weight coefficients respectively, satisfying =1; is the specular reflection interference function, is the shadow interference function, is the atmospheric disturbance function; According to the interference degree score D, the quality of the black and odorous water body image is graded, and the spectral data of the black and odorous water body image that needs to be corrected is determined.

9. The method for identifying black and odorous water bodies based on remote sensing spectroscopy according to claim 1, characterized in that: According to the evaluation result of the interference degree of the shooting angle information on the quality of the black and odorous water body image, and the ground object classification result of the surrounding image of the black and odorous water body area, the spectral data of the water body image of the black and odorous water body area is corrected, including: A spectral correction algorithm of least squares matched filtering is adopted to determine the type of interference to the corresponding pixel points according to the evaluation results and the ground object classification results, and to correct the spectral data of the interfered black and smelly water body image.

10. The method for identifying black and odorous water bodies based on remote sensing spectroscopy according to claim 1, characterized in that: The method further comprises: Determine the ground sampling points, obtain the ground measured spectral data of the ground sampling points, and verify the spectral data of the corrected black and odorous water body image based on the ground measured spectral data.

Images