Water body inherent optical quantity remote sensing inversion method, system, medium and equipment

Abstract

The invention discloses a water body inherent optical quantity remote sensing inversion method, system, medium and equipment, and relates to the technical field of water body inherent optical quantity measurement, and the method comprises the steps: obtaining inherent optical quantity data and sampling point coordinates of a water body sample; acquiring multispectral satellite remote sensing data according to the spatial and temporal distribution of the water body sample, preprocessing the multispectral satellite remote sensing data, extracting the reflectivity of each wave band of a corresponding pixel at a sampling point coordinate from the preprocessed data, and calculating a characteristic index; combining the inherent optical quantity data, the reflectivity of each wave band and the characteristic index into a data set A, and dividing the data set A into a training set and a test set; constructing a deep learning model which comprises an input layer, a feature extraction layer, a feature enhancement full-connection block and an output layer; and training and testing the model by using the training set and the test set, wherein the tested model is used for predicting the inherent optical quantity of the water body. The method for predicting the inherent optical quantity of the water body is high in universality and stable in inversion result.

Description

Technical Field

[0001] The present invention relates to the technical field of water body intrinsic optical quantity measurement, and in particular to a water body intrinsic optical quantity remote sensing inversion method, system, medium and equipment. Background Art

[0002] Inherent Optical Properties (IOPs) of water refer to the optical parameters of water (such as absorption coefficient, scattering coefficient, backscattering coefficient, and attenuation coefficient) that are directly related to the propagation and absorption characteristics of light. They are determined by the physical and chemical composition of the water, reflect the optical properties of the water itself, and are not affected by external light field conditions. IOPs are an important basis for understanding the optical environment of water, analyzing water quality, monitoring marine ecosystems, and performing water color remote sensing inversion.

[0003] Traditional measurement and analysis of the intrinsic optical quantities of water bodies usually involve laboratory and field measurements, combined with optical instruments and data analysis methods. Although accurate optical property data can be obtained to a certain extent, these methods usually require high human and material resources and are limited in space and time, so they may face some challenges in practical applications.

[0004] With the development of science and technology, remote sensing technology provides an effective means for the measurement and analysis of the inherent optical quantities of water bodies, and can achieve efficient monitoring in spatial and temporal dimensions, gradually supplementing and replacing these traditional methods. The total absorption coefficient and backscattering coefficient of water bodies are mainly inverted by using the band reflectance obtained by field measurements or satellites. Common inversion algorithms can be roughly divided into two categories: empirical algorithms and semi-empirical and semi-analytical algorithms. Empirical algorithms such as the total absorption coefficient of water bodies constructed by Lee et al., the absorption coefficient of phytoplankton and the spectral reflectance at 440nm have a good linear inversion model; Li et al. established an inversion model for the backscattering coefficient at 778nm and the band reflectance of the lower surface of the water body. The statistical relationship between band reflectance and IOPs is established based on simple or multiple regression methods. This algorithm is more dependent on data and has poor versatility.

[0005] The semi-empirical and semi-analytical algorithm obtains the approximate equation of water body radiation transfer through numerical simulation to clarify the relationship between water body IOPs and water body component content. Among them, Maritorena et al. developed the GSM algorithm based on Garver et al., obtained parameters such as chlorophyll concentration and backscattering coefficient, and applied it to the global IOPs inversion of MODIS. However, the algorithm requires the relevant parameters to be given in advance and the measured data range of a certain area to be set. Therefore, in actual applications, the inversion results in some nearshore or inland areas are not very stable, and the full-band IOPs inversion cannot be achieved. Summary of the invention

[0006] The purpose of the present invention is to solve the problem that the measurement method of the inherent optical quantity of water body has poor versatility, and propose a remote sensing inversion method of the inherent optical quantity of water body, comprising the following steps: S1. Obtain the intrinsic optical quantity data and sampling point coordinates of water samples; S2, obtaining multispectral satellite remote sensing data and preprocessing it, extracting the reflectance of each band of the corresponding pixel at the sampling point coordinates from the preprocessed data, and calculating the characteristic index; S3, combining the intrinsic optical quantity data, the reflectivity of each band, and the characteristic index into a data set A, and dividing the data set A into a training set and a test set; S4. Build a deep learning model, including: input layer, feature extraction layer, feature enhancement fully connected block, and output layer; Dataset A is used as input data. The input layer accepts the input data, and the feature extraction layer extracts the wavelength information in the input data. The feature enhancement fully connected block accepts input data and wavelength information, and after being processed by N feature enhancement fully connected blocks, the feature vector is passed through the output layer to output the prediction of the inherent optical quantity of the water body in a given band; Among them, the feature enhancement fully connected block includes: a fully connected layer and a splicing layer; S5. Use the training set and the test set to train and test the model. The tested model is used to predict the inherent optical quantity of the water body.

[0007] Furthermore, the multispectral satellite remote sensing data are subjected to radiation calibration, atmospheric correction, orthorectification, and water body mask processing to obtain preprocessed data.

[0008] Furthermore, the reflectivity of each band includes reflectivity of red, green, blue, near infrared, and short-wave infrared bands.

[0009] Furthermore, the characteristic indexes include: normalized difference water index, green difference water index, environmental water index, water relative index, phytoplankton index, normalized difference humidity index, land surface moisture index, turbidity index; The calculation formula is as follows: ; ; ; ; ; ; ; ; Among them, NDWI is the normalized difference water index, NDGI is the green difference water index, EWI is the environmental water index, WRI is the water relative index, ChI is the phytoplankton index, NDMI is the normalized difference moisture index, LSWI is the land surface moisture index, and TI is the turbidity index; represents the reflectivity of the green light band, represents the reflectivity in the near-infrared band, represents the reflectivity of the red light band, Represents the reflectivity of the blue light band, Indicates the reflectivity in the short-wave infrared light band.

[0010] Further, The first fully connected layer of the feature enhanced fully connected block processes the input data, and the first splicing layer of the feature enhanced fully connected block splices the output of the first fully connected layer and the wavelength information output by the feature extraction layer to obtain the output of the first feature enhanced fully connected block. The input of the second to Nth fully connected layers is the output of the previous feature enhanced fully connected block. The second to Nth splicing layers splice the wavelength information output by the feature extraction layer and the output of the current fully connected layer to obtain the output of the second to Nth feature enhanced fully connected blocks.

[0011] The present invention also proposes a remote sensing inversion system for inherent optical quantities of water bodies, comprising: The first data acquisition module is used to obtain the intrinsic optical quantity data and sampling point coordinates of the water sample; The second data acquisition module is used to acquire and preprocess multispectral satellite remote sensing data, extract the reflectance of each band of the corresponding pixel at the sampling point coordinates from the preprocessed data, and calculate the characteristic index; A data partitioning module is used to combine the intrinsic optical quantity data, the reflectivity of each band, and the characteristic index into a data set A, and divide the data set A into a training set and a test set; Model building module, used to build deep learning models, including: input layer, feature extraction layer, feature enhancement fully connected block, output layer; Dataset A is used as input data. The input layer accepts the input data, and the feature extraction layer extracts the wavelength information in the input data. The feature enhancement fully connected block accepts input data and wavelength information, and after being processed by N feature enhancement fully connected blocks, the feature vector is passed through the output layer to output the prediction of the inherent optical quantity of the water body in a given band; Among them, the feature enhancement fully connected block includes: a fully connected layer and a splicing layer; The model training and testing module is used to train and test the model using the training set and the test set. The tested model is used to predict the inherent optical quantity of the water body.

[0012] The present invention also proposes a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the remote sensing inversion method of the intrinsic optical quantity of a water body is implemented.

[0013] The present invention also proposes an electronic device, comprising a processor and a memory, wherein the processor and the memory are interconnected, wherein the memory is used to store a computer program, the computer program comprises computer-readable instructions, and the processor is configured to call the computer-readable instructions to execute the above-mentioned remote sensing inversion method of intrinsic optical quantities of water bodies.

[0014] The beneficial effects brought by the technical solution provided by the present invention are: The present invention introduces multiple exponential features, which reduce the influence of spatiotemporal factors on data analysis to a certain extent, suppress the noise that exists in each band, effectively improve the model's ability to analyze data at different times and locations, enhance the model's prediction performance, enhance the model's robustness, and stabilize the inversion results. In addition, through the feature-enhanced fully connected block, the band information is enhanced in the model training, effectively realizing the prediction of the inherent optical quantity of the water body in the whole band, and reducing the model's dependence on data and parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a flow chart of a method for remote sensing inversion of intrinsic optical quantities of water bodies according to an embodiment of the present invention; Figure 2 is a structural diagram of a deep learning model constructed in an embodiment of the present invention; Figure 3 is a block diagram of an electronic device in an exemplary embodiment of the present invention; Figure 4 These are the predicted value and the true value of the total particle backscattering coefficient obtained by the model of the embodiment of the present invention using the validation set. DETAILED DESCRIPTION

[0016] To make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0017] The flowchart of the remote sensing inversion method of the inherent optical quantity of water body according to the embodiment of the present invention is as follows: Figure 1 , specifically including the following steps: S1. Obtain the intrinsic optical quantity data and sampling point coordinates of the water sample. Select the experimental data of n water samples, and the sample collection location is , the corresponding intrinsic optical quantity of water is , the collection date is .in , Indicates the wavelength value.

[0018] Intrinsic optical quantities refer to quantities that are only related to the composition of the water body and do not change with changes in lighting conditions, including: (1) Absorption coefficient, scattering coefficient and scattering phase function of water molecules; (2) Absorption coefficient, unit absorption coefficient, scattering coefficient, unit scattering coefficient, backscattering coefficient, forward scattering coefficient, and scattering phase function of Chl-a; (3) Unit absorption coefficient of yellow substance; (4) Absorption and scattering characteristics of other components, including inorganic matter, debris, etc.

[0019] S2. Acquire multispectral satellite remote sensing data and perform preprocessing, extract the reflectance of each band of the corresponding pixel at the sampling point coordinates from the preprocessed data, and calculate the characteristic index.

[0020] According to the spatiotemporal distribution of the selected water samples, the multispectral images taken by the same multispectral remote sensing satellite at the corresponding date and location are obtained. ENVI5.3 or other remote sensing professional software is used to complete the radiation calibration, atmospheric correction, orthorectification, and water mask processing of the multispectral images; or water mask processing is added to the existing image preprocessing products through band analysis to obtain preprocessed data.

[0021] According to the geographical location of the water sample, find the corresponding pixel position in the image and read the pixel value of the red, green, blue, near infrared, and short-wave infrared bands of this pixel. Obtain the reflectivity of the red, green, blue, near infrared, and short-wave infrared bands.

[0022] The characteristic indexes include: normalized difference water index, green difference water index, environmental water index, water relative index, phytoplankton index, normalized difference humidity index, land surface moisture index, and turbidity index; The calculation formula is as follows: ; ; ; ; ; ; ; ; Among them, NDWI is the normalized difference water index, NDGI is the green difference water index, EWI is the environmental water index, WRI is the water relative index, ChI is the phytoplankton index, NDMI is the normalized difference moisture index, LSWI is the land surface moisture index, and TI is the turbidity index; represents the reflectivity of the green light band, represents the reflectivity in the near-infrared band, represents the reflectivity of the red light band, Represents the reflectivity of the blue light band, Indicates the reflectivity in the short-wave infrared light band.

[0023] S3. The intrinsic optical quantity data, the reflectivity of each band, and the characteristic index are combined into a data set A, and the data set A is divided into a training set and a test set.

[0024] Dataset ,in: .

[0025] S4. Build a deep learning model, including: input layer, feature extraction layer, feature enhancement fully connected block, and output layer.

[0026] Data set A is used as input data. The input layer accepts the input data, and the feature extraction layer extracts the wavelength information in the input data.

[0027] The feature enhancement fully connected block accepts input data and wavelength information, and passes through N The feature vector is processed by a feature enhancement fully connected block and passes through the output layer to output the prediction of the inherent optical quantity of the water body in a given band.

[0028] Among them, the feature enhancement fully connected block includes: a fully connected layer and a splicing layer.

[0029] The first fully connected layer of the feature enhanced fully connected block processes the input data, and the first splicing layer of the feature enhanced fully connected block splices the output of the first fully connected layer and the wavelength information output by the feature extraction layer to obtain the output of the first feature enhanced fully connected block. The input of the second to Nth fully connected layers is the output of the previous feature enhanced fully connected block. The second to Nth splicing layers splice the wavelength information output by the feature extraction layer and the output of the current fully connected layer to obtain the output of the second to Nth feature enhanced fully connected blocks.

[0030] Reference for building a deep learning model constructed by the present invention Figure 2 .

[0031] S5. Use the training set and the test set to train and test the model. The tested model is used to predict the inherent optical quantity of the water body.

[0032] The model was trained using the K-fold cross-validation method, where K=4, i.e. the training set was randomly divided into 4 subsets for multiple training and validation cycles. In each cycle, 25% of the subsets were used as validation sets, and a set of model results with the best evaluation indicators in the cycle were selected. During training, the model was optimized using the Adam optimizer, the loss function was the mean squared error function (MSE), and the evaluation indicator was the mean absolute error function (MAE). The formulas are as follows: ; ; Where k is the number of test samples, is the actual value of the test sample, is the predicted value of the test sample.

[0033] In an exemplary embodiment, a remote sensing inversion system for intrinsic optical quantities of a water body is provided, comprising: The first data acquisition module is used to obtain the intrinsic optical quantity data and sampling point coordinates of the water sample; The second data acquisition module is used to acquire and preprocess multispectral satellite remote sensing data, extract the reflectance of each band of the corresponding pixel at the sampling point coordinates from the preprocessed data, and calculate the characteristic index; A data partitioning module is used to combine the intrinsic optical quantity data, the reflectivity of each band, and the characteristic index into a data set A, and divide the data set A into a training set and a test set; Model building module, used to build deep learning models, including: input layer, feature extraction layer, feature enhancement fully connected block, output layer; Dataset A is used as input data. The input layer accepts the input data, and the feature extraction layer extracts the wavelength information in the input data. The feature enhancement fully connected block accepts input data and wavelength information, and after being processed by N feature enhancement fully connected blocks, the feature vector is passed through the output layer to output the prediction of the inherent optical quantity of the water body in a given band; Among them, the feature enhancement fully connected block includes: a fully connected layer and a splicing layer; The model training and testing module is used to train and test the model using the training set and the test set. The tested model is used to predict the inherent optical quantity of the water body.

[0034] In an exemplary embodiment, a computer-readable storage medium is included, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned remote sensing inversion method of intrinsic optical quantities of water bodies is implemented.

[0035] See also Figure 3 In an exemplary embodiment, an electronic device is also included, including at least one processor, at least one memory, and at least one communication bus.

[0036] Wherein, a computer program is stored in the memory, and the computer program includes computer-readable instructions. The processor calls the computer-readable instructions stored in the memory through a communication bus to execute the above-mentioned remote sensing inversion method for inherent optical quantities of water bodies.

[0037] In the embodiment of the present invention, 600 sets of water body intrinsic optical quantity sample data are selected, and the multispectral band images of Sentinel-2 satellite are collected according to the temporal and spatial information of the samples. (including blue band, green band, red band, near infrared band, and short-wave infrared).

[0038] The GEE (Google Earth Engine) platform is used to extract the reflectance of the red, green, blue, near-infrared, and short-wave infrared bands at the temporal and spatial matching pixels of the sampling points. .

[0039] And according to the formula, the index characteristics are calculated, and finally 200 data sets are obtained. .

[0040] in: .

[0041] The data set A is randomly divided into a training set and a validation set in a ratio of 8:2. The training set contains 480 samples and the test set contains 120 samples.

[0042] Define an input layer when building the model to accept the dataset ; Build a custom feature extraction layer to extract band information separately; build three feature enhancement fully connected blocks to extract deep spectral features, and build an output layer to predict the inherent optical quantity of water in a specified band, taking the total particle backscattering coefficient as an example.

[0043] 480 training samples were input into the model for training, using the K-fold cross validation method, where K=4. The batch size of the training process was 16, the number of iterations was 500, the initial learning rate was set to 0.001, and the learning rate self-adjustment strategy was enabled. The model was optimized using the Adam optimizer, the loss function was the mean square error, and the early stopping method was used to stop training in advance when the performance of the validation set no longer improved, to avoid overfitting caused by too many iterations.

[0044] 120 validation samples were input into the trained model for prediction, and the root mean square error was used to evaluate the prediction performance. The predicted value of the total particle backscattering coefficient obtained by the model of the embodiment of the present invention using the validation set was different from the true value. Figure 4shown. Figure 4 In the figure, the true value of the total particle backscattering coefficient is the horizontal coordinate, and the corresponding predicted value is the vertical coordinate. The ideal predicted value should be equal to the true value, so the ideal predicted value and the true value are a straight line with a slope of 1 in this coordinate system, such as Figure 4 In fact, when the true value is used as the horizontal axis, the straight line after the predicted value fitting has a certain deviation from the ideal, such as Figure 4 The solid line in .

[0045] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A remote sensing inversion method for intrinsic optical quantities of water bodies, characterized in that: The following steps are involved: S1. Obtain the intrinsic optical quantity data and sampling point coordinates of water samples; S2. Obtain multispectral satellite remote sensing data and preprocess them according to the temporal and spatial distribution of water samples, extract the reflectance of each band of the corresponding pixel at the sampling point coordinates from the preprocessed data, and calculate the characteristic index; S3, combining the intrinsic optical quantity data, the reflectivity of each band, and the characteristic index into a data set A, and dividing the data set A into a training set and a test set; S4. Build a deep learning model, including: input layer, feature extraction layer, feature enhancement fully connected block, and output layer; Dataset A is used as input data. The input layer accepts the input data, and the feature extraction layer extracts the wavelength information in the input data. The feature enhancement fully connected block accepts input data and wavelength information, and after being processed by N feature enhancement fully connected blocks, the feature vector is passed through the output layer to output the prediction of the inherent optical quantity of the water body in a given band; Among them, the feature enhancement fully connected block includes: a fully connected layer and a splicing layer; S5. Use the training set and the test set to train and test the model. The tested model is used to predict the inherent optical quantity of the water body.

2. A method for remote sensing inversion of intrinsic optical quantities of water bodies according to claim 1, characterized in that: The multispectral satellite remote sensing data is subjected to radiation calibration, atmospheric correction, orthorectification and water body mask processing to obtain preprocessed data.

3. A method for remote sensing inversion of intrinsic optical quantities of water bodies according to claim 1, characterized in that: The reflectivity of each band includes: reflectivity of red light, green light, blue light, near infrared light, and short-wave infrared light bands.

4. The method for remote sensing inversion of intrinsic optical quantities of water bodies according to claim 1, characterized in that: The characteristic indexes include: normalized difference water index, green difference water index, environmental water index, water relative index, phytoplankton index, normalized difference humidity index, land surface moisture index, and turbidity index; The calculation formula is as follows: ； ； ； ； ； ； ； ； Among them, NDWI is the normalized difference water index, NDGI is the green difference water index, EWI is the environmental water index, WRI is the water relative index, ChI is the phytoplankton index, NDMI is the normalized difference moisture index, LSWI is the land surface moisture index, and TI is the turbidity index; represents the reflectivity of the green light band, represents the reflectivity in the near-infrared band, represents the reflectivity of the red light band, Represents the reflectivity of the blue light band, Indicates the reflectivity in the short-wave infrared light band.

5. The method for remote sensing inversion of intrinsic optical quantities of water bodies according to claim 1, characterized in that: The first fully connected layer of the feature enhanced fully connected block processes the input data, and the first splicing layer of the feature enhanced fully connected block splices the output of the first fully connected layer and the wavelength information output by the feature extraction layer to obtain the output of the first feature enhanced fully connected block. The input of the second to Nth fully connected layers is the output of the previous feature enhanced fully connected block. The second to Nth splicing layers splice the wavelength information output by the feature extraction layer and the output of the current fully connected layer to obtain the output of the second to Nth feature enhanced fully connected blocks.

6. A remote sensing inversion system for intrinsic optical quantities of water bodies, characterized in that: include: The first data acquisition module is used to obtain the intrinsic optical quantity data and sampling point coordinates of the water sample; The second data acquisition module is used to acquire and preprocess multispectral satellite remote sensing data, extract the reflectance of each band of the corresponding pixel at the sampling point coordinates from the preprocessed data, and calculate the characteristic index; A data partitioning module is used to combine the intrinsic optical quantity data, the reflectivity of each band, and the characteristic index into a data set A, and divide the data set A into a training set and a test set; Model building module, used to build deep learning models, including: input layer, feature extraction layer, feature enhancement fully connected block, output layer; Dataset A is used as input data. The input layer accepts the input data, and the feature extraction layer extracts the wavelength information in the input data. The feature enhancement fully connected block accepts input data and wavelength information, and after being processed by N feature enhancement fully connected blocks, the feature vector is passed through the output layer to output the prediction of the inherent optical quantity of the water body in a given band; Among them, the feature enhancement fully connected block includes: a fully connected layer and a splicing layer; The model training and testing module is used to train and test the model using the training set and the test set. The tested model is used to predict the inherent optical quantity of the water body.

7. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

8. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the processor and the memory are interconnected, wherein the memory is used to store a computer program, the computer program comprises computer-readable instructions, and the processor is configured to call the computer-readable instructions to execute the method according to any one of claims 1 to 5.

Images