Water depth inversion method and system based on multispectral remote sensing image

Abstract

The invention relates to the technical field of exploration, and discloses a water depth inversion method and system based on a multispectral remote sensing image, which utilizes an independent verification sample set to carry out hierarchical precision evaluation, and analyzes model performance differences according to dimensions such as a water depth range and a substrate type. Error propagation of links such as atmospheric correction and water level correction is quantified through full-chain uncertainty analysis, an uncertainty quantification model of ensemble learning is constructed, and a pixel-level precision distribution diagram is generated. Therefore, according to the technical scheme, a closed-loop dynamic adaptation framework is formed by systematically integrating multi-temporal data dynamic modeling and an uncertainty quantification mechanism, a spatio-temporal variation compensation mechanism is embedded in the whole process from data acquisition to result verification, the interference of spatial-temporal heterogeneity of water optical characteristics on water depth inversion is effectively dealt with, and the accuracy of water depth inversion is improved. The problems of parameter mismatch and precision reduction of the water depth inversion model caused by dynamic change of optical characteristics of a coastal water body in seasonal and tidal scales are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of surveying, in particular to a water depth inversion method and system based on multi-spectral remote sensing images. BACKGROUND

[0002] Remote sensing science and technology, as an important part of spatial information science, has been widely applied in the field of resource and environment monitoring and spatial data acquisition. As an important branch of remote sensing technology, marine remote sensing focuses on the inversion of key parameters such as ocean color, water temperature and water depth from multi-spectral and hyperspectral image data obtained by satellite or aerial platforms. In shallow topographic mapping and seabed geomorphology research, the water depth inversion technology based on multi-spectral remote sensing images analyzes the water body transmission characteristics of blue and green bands, and establishes an empirical or semi-analytical model of spectral radiation and water depth. The implementation of this technology relies on the accurate modeling of the water body radiation transfer process, in which the inherent optical properties of the water body, including the total suspended matter backscattering coefficient, phytoplankton chlorophyll absorption coefficient and the exponential decay absorption coefficient of colored dissolved organic matter (CDOM), directly determine the light attenuation law in the water body and the transmission efficiency of the bottom reflectance signal.

[0003] The current mainstream multi-spectral water depth inversion method has significant limitations in dealing with the spatial and temporal heterogeneity of water body optical properties. Most of the inversion algorithms used in business applications use a single time-phase set of water body optical parameters for global modeling, ignoring the systematic drift of nearshore water body optical properties on the inter-monthly and seasonal scales caused by land-based material input and biogeochemical processes. For example, a large amount of land-based suspended sediment carried by runoff in the rainy season can increase the attenuation coefficient of coastal water by 2-5 times in a short period of time; the outbreak of diatom blooms in spring can cause the chlorophyll concentration to jump from the winter baseline level (<1mg / m³) to 10-30mg / m³; and the CDOM concentration can differ by several times on both sides of the estuary salinity front. These dynamic changes are systematically ignored or averaged in the existing static modeling framework, resulting in long-term mismatch between model parameters and actual water body conditions.

[0004] The contradiction between the spatio-temporal variation of the water body optical characteristics and the static assumption of the model causes multiple precision degradation phenomena in practical applications. In the monsoon climate dominated estuary and bay area, when the training model in the dry season is applied in the wet season, the increased light attenuation caused by high suspended solids concentration will be misjudged as an increase in water depth, resulting in a systematic positive deviation of 1-3 meters. In the eutrophic lake and reservoir, the strong absorption of chlorophyll to blue light band during algal bloom in spring and summer causes the spectral response of the same water depth to be significantly reduced, resulting in a "false shallowing" phenomenon in the inversion result. More seriously, in the intertidal zone with strong tidal power, the periodic fluctuation of suspended solids concentration (up to 5-10 times) caused by the rising and falling tides results in a difference of 40%-60% in the inversion water depth at the same location at different tidal times, completely losing the reliability of the topographic monitoring, and seriously restricting the practical value of the technology in the dynamic water environment.

[0005] Therefore, a water depth inversion method and system based on multispectral remote sensing images are proposed to solve the above problems. SUMMARY

[0006] The purpose of the present application is to provide a water depth inversion method and system based on multispectral remote sensing images to solve the problem that most of the inversion algorithms for business applications ignore the systematic drift of the optical characteristics of nearshore water bodies caused by land-based material input and biogeochemical processes on the inter-monthly and seasonal scales.

[0007] To achieve the above purpose, the present application provides the following technical solution: a water depth inversion method based on multispectral remote sensing images, the specific steps are as follows: S1: Obtain multi-temporal multispectral remote sensing images of the target water area in different seasons and tidal states, perform radiation calibration, atmospheric correction and geometric registration on the multi-temporal multispectral remote sensing images, perform dynamic water level correction based on the astronomical tide model, and unify the multi-temporal multispectral remote sensing images to the average sea level reference surface; S2: Extract the water body optical characteristic parameters of the multi-temporal multispectral remote sensing images, divide the optical dynamic zones using a time series clustering algorithm, establish an optical parameter time series database for each optical dynamic zone, and establish an optical parameter prediction model according to the environmental driving factors; S3: Design a sampling scheme according to the optical dynamic zones, collect measured data of water depth, suspended solids concentration and chlorophyll concentration in each optical dynamic zone, and supplement sampling points in high uncertainty areas using an active learning strategy; S4: Establish a semi-analytical water depth inversion model, express the water body attenuation coefficient as a function of the optical component concentration and the spatio-temporal factor, construct an optical parameter time series prediction module of the long short-term memory network, and adjust the parameters of the semi-analytical water depth inversion model according to the output of the optical parameter time series prediction module. S5: Establishing a substrate type classification system, using a spectral unmixing algorithm to separate water body signals and substrate signals, constructing a deep learning decoupling network, and inverting substrate reflectivity and water depth through the deep learning decoupling network; S6: Using a Bayesian fusion method to spatiotemporally fuse multi-period water depth inversion results, introducing a topographic continuity constraint condition, calculating water depth changes between adjacent periods and performing a significance test, and identifying topographic evolution areas; S7: Using an independent validation sample set for hierarchical precision evaluation and full-chain uncertainty analysis, constructing an uncertainty quantification model of ensemble learning, and generating a pixel-level precision distribution map.

[0008] Preferably, the step S1 is specifically as follows: S1.1: Obtain multi-temporal and multi-spectral remote sensing images of the target water area in the wet season, dry season and transition period, record the acquisition time and satellite overpass time of each image, perform radiation calibration, atmospheric correction and geometric registration on the multi-temporal and multi-spectral remote sensing images in turn, convert the original DN value to apparent reflectance by radiation calibration, use a water body self-adaptive correction algorithm based on the dark pixel principle for atmospheric correction, and register all images to a unified geographic coordinate system by geometric registration. Select one image as the radiation reference image, normalize other images to the radiation reference image by histogram matching method, and obtain a multi-temporal standardized image dataset; S1.2: Obtain the astronomical tide harmonic constant of the target water area, which includes the amplitude and phase angle of each component tide. Calculate the instantaneous tide level value according to the astronomical tide harmonic constant and the acquisition time of each image using the astronomical tide prediction formula. Set the mean sea level as the water level reference surface, calculate the water level deviation, and establish the water level correction amount for each pixel in the multi-temporal standardized image dataset. Subtract the water level deviation from the apparent water depth observation value to obtain the true water depth value corrected to the mean sea level reference surface.

[0009] Preferably, the step S2 is specifically as follows: S2.1: Extract blue, green and red band reflectance from the multi-temporal standardized image dataset, calculate the blue-green band ratio, normalized suspended sediment index and chlorophyll fluorescence peak index as water body optical characteristic parameters, arrange the water body optical characteristic parameters in time sequence for each pixel to construct a time series vector, calculate the similarity distance between time series vectors of pixels using dynamic time warping algorithm to form a similarity distance matrix, divide the pixels into optical dynamic partitions based on the similarity distance matrix using K-means clustering algorithm, and calculate the average value of water body optical characteristic parameters of each optical dynamic partition at each time phase to form an optical characteristic time variation curve, which is stored as an optical parameter time series database; S2.2: Obtain the environmental driving factor data corresponding to the multi-temporal normalized image dataset, the environmental driving factor data including daily rainfall, daily runoff, daily average wind speed and daily average water temperature, correspond the environmental driving factor data to each optical dynamic partition through a spatial interpolation method, convert the water body optical characteristic parameters into suspended matter concentration values and chlorophyll concentration values as the main control optical parameters through an empirical formula, use the environmental driving factor data as the independent variable and the main control optical parameters as the dependent variable, and use a multiple linear regression method to establish an optical parameter prediction model, and establish an independent optical parameter prediction model for each optical dynamic partition.

[0010] Preferably, the step S3 is specifically as follows: S3.1: Read the optical dynamic partition result, calculate the area proportion and fluctuation amplitude of each optical dynamic partition, the fluctuation amplitude being represented by the standard deviation of the water body optical characteristic parameters, calculate the initial sampling point number allocation weight according to the area proportion and fluctuation amplitude, distribute the preset total sampling point number to each optical dynamic partition according to the weight, determine the initial sampling point position by using a stratified random sampling method, measure the water depth value using a depth sounder, collect water samples and measure the suspended matter concentration value and chlorophyll concentration value, and establish an initial sampling point data table containing geographic coordinates, sampling time and measured parameters, perform spatio-temporal matching of the sampling point data and the multi-temporal normalized image dataset, extract the band reflectance and water body optical characteristic parameters of the corresponding pixels to form an initial training sample set; S3.2: Divide the initial training sample set into a training subset and a validation subset, train the water depth inversion model and the optical parameter inversion model using the training subset, apply the trained model to the pixels not participating in the training to calculate the predicted values, calculate the prediction uncertainty of each pixel by using a Gaussian process regression method, sort the pixels according to the prediction uncertainty, select the first N pixels as candidate supplementary sampling points, perform spatial clustering on the candidate sampling points, select the pixel with the highest prediction uncertainty in each clustering cluster as the actual supplementary sampling point, repeat the measurement process of step S3.1 to obtain supplementary sampling data, and add the supplementary data to the initial training sample set to form an expanded training sample set.

[0011] Preferably, the step S4 is specifically as follows: S4.1: A semi-analytical water depth inversion model is established based on radiation transfer theory, the water surface radiance is expressed as the sum of the water scattering contribution and the bottom reflection contribution, the water attenuation coefficient is expressed as a linear combination of the suspended matter absorption coefficient, the chlorophyll absorption coefficient and the colored dissolved organic matter absorption coefficient, each absorption coefficient has a power function relationship with the corresponding concentration, a seasonal correction factor and a tidal correction factor are introduced to modify the baseline value of the absorption coefficient, the measured data is extracted from the extended training sample set, the nonlinear least squares method is used to solve the minimum value of the objective function to obtain the optimal parameter value, the parameters are set for each optical dynamic partition, and a correspondence table of the optical dynamic partition number and the model parameter group is established; S4.2: An optical parameter time series prediction module of a long short-term memory network is constructed, the long short-term memory network includes an input layer, 2 to 4 LSTM hidden layers and a fully connected output layer, the input layer receives historical 3 to 12 time phase water optical characteristic parameters and environmental driving factors, LSTM training data set is generated by a time sliding window method, the model is trained by using an Adam optimization algorithm, historical data before the target time phase is input into the trained model to obtain the suspended matter concentration prediction value and the chlorophyll concentration prediction value, the dynamic water attenuation coefficient is calculated according to the prediction value and the space-time correction factor, and the semi-analytical water depth inversion model is substituted to perform water depth inversion calculation on the pixel.

[0012] Preferably, the step S5 is specifically as follows: S5.1: A bottom type classification system including a sandy bottom, a muddy bottom, a rock bottom and a seagrass bed bottom is established, bottom samples are collected at sampling points in a shallow water area with a water depth less than 5 m, a laboratory spectrometer is used to measure spectral reflectivity in a 400 nm to 900 nm wave band, a bottom type spectral library is established, a pure water body spectral model is established for pixels in a deep water area with a water depth greater than 20 m, a linear spectral unmixing algorithm is used to decompose an observation spectrum vector into a linear weighted combination of a pure water body endmember spectrum and a bottom endmember spectrum, a least squares method is used to solve a mixing proportion coefficient, the mixing proportion coefficient is decomposed into a water body spectral contribution vector and a bottom spectral contribution vector, and a spectral angle mapping method is used for bottom type identification; S5.2: A deep learning decoupling network including a water body attenuation feature extraction branch and a bottom reflection feature extraction branch is constructed, the two branches respectively include 3 to 5 convolution layers and pooling layers to output feature vectors, the feature vectors are weighted and modulated through an attention mechanism module, a feature fusion layer is used for splicing the feature vectors to form a fusion feature vector, a fully connected decoding layer is used to output a water depth prediction value, a bottom reflectivity prediction value and a bottom type probability distribution, a composite loss function including a data-driven loss term and a physical constraint loss term is constructed, an Adam optimization algorithm is used to train the network, and the trained network is applied to the shallow water area pixels for water depth inversion and bottom type identification.

[0013] Preferably, the step S6 is specifically as follows: S6.1: Collect multi-period water depth inversion results, extract the water depth inversion value of each time phase for each pixel to form time series data, establish a Bayesian fusion framework, take the multi-period water depth inversion value as observation data, construct an observation model and a state transition model, introduce spatial smoothness constraints and slope rationality constraints as Bayesian prior distribution, encode the spatial smoothness constraints as Markov random field prior distribution, and encode the slope rationality constraints as slope prior distribution, solve the posterior probability distribution by Markov chain Monte Carlo method or variational Bayesian method, extract the posterior mean value from the posterior distribution as the fused water depth value, and extract the posterior standard deviation as the fusion uncertainty index; S6.2: Select two adjacent fusion water depth products in time interval, calculate the water depth change for each pixel, calculate the uncertainty of water depth change according to the error propagation law, construct the significance test statistic as the ratio of water depth change and its uncertainty, set the significance level to 0.05 or 0.01 to determine the critical value, and determine the significant change pixel when the absolute value of the test statistic is greater than the critical value, perform spatial clustering analysis on the significance test results, set the minimum cluster area threshold to remove fragmented areas, calculate the area, average water depth change and change type of each topographic evolution area, and establish a topographic evolution area database.

[0014] Preferably, the step S7 is specifically as follows: S7.1: Collect independent validation sample set, the number of samples in the independent validation sample set is 20% to 30% of the total sample size, stratify the independent validation sample set according to water depth range, bottom type and optical dynamic partition, extract the water depth inversion value of the pixel corresponding to the validation sample point position from the fusion water depth product, calculate the root mean square error, mean absolute error, mean deviation and determination coefficient of each stratified subset, decompose the water depth inversion process into five links of atmospheric correction, water level correction, optical parameter estimation, model structure selection and parameter setting, calculate the uncertainty of each link, synthesize the overall uncertainty according to the error propagation theory, and calculate the contribution rate of each link to the overall uncertainty; S7.2: Construct an ensemble learning uncertainty quantification model including 5 to 20 base models, the base model types include semi-analytical model, deep learning network, random forest regression, support vector regression and gradient boosting regression, apply the trained base model to each pixel to calculate the water depth prediction value, calculate the mean value of the base model prediction value as the ensemble prediction value, calculate the standard deviation as the prediction uncertainty, use the independent validation sample set to establish an uncertainty calibration function, use piecewise linear interpolation or cubic spline interpolation method, apply the calibration function to each pixel to generate calibrated prediction error estimate value, generate pixel-level precision distribution map, calculate confidence interval according to confidence level, and generate confidence interval width distribution map.

[0015] The application also comprises a water depth inversion system based on multispectral remote sensing images, comprising an image preprocessing module, an optical zoning module, a sample collection module, a model inversion module, a bottom material decoupling module, a Bayesian fusion module and an accuracy evaluation module. The image preprocessing module is used for obtaining multi-temporal multispectral remote sensing images in different seasons and tidal states, performing radiation calibration, atmospheric correction, geometric registration and water level dynamic correction on the images, and unifying the images to the average sea level reference. The optical zoning module is used for extracting water body optical characteristic parameters, dividing the optical dynamic zones by using a time series clustering algorithm, establishing a time series database of the optical parameters of the zones and constructing an optical parameter prediction model. The sample collection module is used for designing a sampling scheme according to the optical zoning, collecting water depth and optical component measured data, and supplementing sampling points in high uncertainty areas by using an active learning strategy. The model inversion module is used for establishing a semi-analytical water depth inversion model, combining the time series prediction to output dynamic adjustment of the model parameters and realizing water depth calculation. The bottom material decoupling module is used for establishing a bottom material classification system, separating the water body and bottom material signals by using a spectral unmixing algorithm, and realizing joint inversion of the bottom material reflectivity and water depth by using a deep learning network. The Bayesian fusion module is used for spatio-temporal fusion of multi-period water depth inversion results, introducing spatial smoothing and slope constraints, calculating water depth changes and performing significance test to identify topographic evolution areas. The accuracy evaluation module is used for carrying out hierarchical accuracy evaluation and uncertainty analysis based on an independent verification sample set, constructing an integrated learning uncertainty quantification model to generate a pixel-level accuracy distribution and a confidence interval map.

[0016] Compared with the prior art, the application has the following beneficial effects: 1. Hierarchical accuracy evaluation is carried out by using an independent verification sample set, and model performance differences are analyzed in dimensions such as water depth range and bottom material type. The whole-chain uncertainty analysis quantifies the error propagation of atmospheric correction, water level correction and other links, and an integrated learning uncertainty quantification model is constructed to generate a pixel-level accuracy distribution map. Therefore, the technical scheme forms a closed-loop dynamic adaptation framework by systematically integrating multi-temporal data dynamic modeling and uncertainty quantification mechanisms, and embeds a spatio-temporal change compensation mechanism throughout the whole process from data acquisition to result verification, effectively deals with the interference of spatio-temporal heterogeneity of water body optical characteristics on water depth inversion, and solves the problem of model parameter mismatch and accuracy decline caused by the dynamic change of water body optical characteristics in seasonal and tidal scales.

[0017] 2、By dividing the initial training sample set into training subsets and validation subsets, independent data support is provided for model evaluation. The prediction model established by the training subsets can identify potential knowledge blind spots. The Gaussian process regression method is used to calculate the prediction uncertainty, which can effectively quantify the prediction confidence and focus on the most uncertain areas of the model. By spatial clustering of candidate sampling points and selecting the most informative points in each cluster, the efficiency of supplementary sampling is maximized. By iteratively integrating new data, the sample set gradually adapts to the complex evolution of water optical properties, significantly improving the representativeness of the training data and the model generalization ability. The whole scheme solves the problem of insufficient sample representativeness in optical dynamic zoning through dynamic optimization of sampling resource allocation and active learning mechanism, ensuring that the measured data can accurately capture the spatio-temporal variation of water optical properties, providing a high-quality training basis for water depth inversion models. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 A method step diagram of the present application.

[0019] Figure 2 A system flowchart of the present application. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0021] Embodiment one: please refer to Figure 1 A water depth inversion method based on multi-spectral remote sensing images, the specific steps are as follows: S1: Obtain multi-temporal multi-spectral remote sensing images of the target water area in different seasons and tidal states, perform radiation calibration, atmospheric correction and geometric registration on the multi-temporal multi-spectral remote sensing images, perform water level dynamic correction based on the astronomical tide model, and unify the multi-temporal multi-spectral remote sensing images to the average sea level reference surface; S2: Extract the water optical characteristic parameters of the multi-temporal multi-spectral remote sensing images, divide the optical dynamic zoning by using the time series clustering algorithm, establish the optical parameter time series database of each optical dynamic zoning, and establish the optical parameter prediction model according to the environmental driving factor; S3: According to the optical dynamic zoning, design a sampling scheme, collect measured data of water depth, suspended matter concentration and chlorophyll concentration in each optical dynamic zoning, and supplement sampling points in high uncertainty areas by using an active learning strategy; S4: Establish a semi-analytical water depth inversion model, express the water body attenuation coefficient as a function of optical component concentration and space-time factor, construct an optical parameter time series prediction module of long short-term memory network, and adjust the parameters of the semi-analytical water depth inversion model according to the output of the optical parameter time series prediction module; S5: Establish a bottom type classification system, separate the water signal and the bottom signal by using a spectral unmixing algorithm, construct a deep learning decoupling network, and invert the bottom reflectivity and water depth through the deep learning decoupling network; S6: Temporal and spatial fusion of multi-period water depth inversion results is performed by using a Bayesian fusion method, a topographic continuity constraint condition is introduced, a water depth change quantity of adjacent periods is calculated and a significance test is performed, and a topographic evolution area is identified; S7: Hierarchical precision evaluation and whole-chain uncertainty analysis are performed by using an independent verification sample set, an uncertainty quantification model of ensemble learning is constructed, and a pixel-level precision distribution map is generated.

[0022] In the embodiment: The multi-temporal and multi-spectral remote sensing image can be understood as image data with multiple band information obtained from a satellite or an aerial platform, which can be obtained by a Landsat series satellite, a Sentinel series satellite or other remote sensing devices with multi-spectral imaging capability. The process of radiation calibration can convert the original digital value into a physical quantity, such as apparent reflectance, by using the sensor parameters calibrated in the laboratory. The specific implementation of atmospheric correction includes but is not limited to correction algorithms based on the dark pixel principle, MODTRAN radiation transfer model simulation method or 6S model correction method. The implementation of the astronomical tide model can calculate the tidal components by harmonic analysis method, or predict the tidal level change by numerical simulation method.

[0023] The division of optical dynamic zoning can be realized by various clustering algorithms, such as hierarchical clustering algorithm, density clustering algorithm or fuzzy C-means clustering algorithm, the purpose of which is to divide the water area according to the time series homology of water optical characteristics. The establishment of the optical parameter prediction model can use machine learning methods such as support vector machine regression, random forest regression or artificial neural network, combined with environmental driving factor data to predict the change trend of optical parameters.

[0024] The implementation of active learning strategy can be completed in many ways, such as selecting the area with larger prediction error as the supplementary sampling point based on uncertainty sampling method, or using diversity-based sampling method to ensure the spatial distribution uniformity of sampling points. The water body attenuation coefficient in the semi-analytical water depth inversion model can be described by various mathematical expressions, such as polynomial fitting, exponential function fitting or piecewise linear interpolation method. The training of long short-term memory network can optimize the model performance by adjusting the number of hidden layers, learning rate or regularization parameter.

[0025] The specific implementation of the spectral unmixing algorithm can adopt the non-negative matrix factorization method, independent component analysis method or physical model-based unmixing algorithm to separate the water body signal and the substrate signal. The construction of the deep learning decoupling network can optimize the network structure by adjusting the convolution kernel size, pooling layer step or activation function type. The implementation of the Bayesian fusion method can solve the posterior probability distribution through Markov chain Monte Carlo sampling method or variational inference method. The encoding of the terrain continuity constraint condition can be realized by the neighborhood pixel difference minimization method or the slope gradient smoothing method.

[0026] By systematically integrating multi-temporal data dynamic modeling and uncertainty quantification mechanism, the interference of water body optical property spatial and temporal heterogeneity on water depth inversion is effectively addressed. The core lies in building a closed-loop dynamic adaptation framework, which embeds the temporal and spatial variation compensation mechanism throughout the process from data acquisition to result verification, avoiding the systematic bias caused by ignoring the optical parameter drift in traditional static models. By introducing optical dynamic zoning, optical parameter prediction model and active learning strategy, the model's adaptability to dynamic water environment is significantly improved. At the same time, through the Bayesian fusion method and the uncertainty quantification model of ensemble learning, the spatio-temporal consistency and reliability evaluation of multi-period water depth inversion results are realized, providing technical support for precise application in dynamic water environment.

[0027] By obtaining multi-temporal multi-spectral remote sensing images of the target water area in different seasons and tidal states, first, the images are processed for radiation calibration, atmospheric correction and geometric registration to eliminate the interference caused by imaging condition differences. Based on the astronomical tide model, the water level is dynamically corrected, and all images are unified to the mean sea level reference surface, ensuring the comparability of multi-temporal data in time and space, and providing consistent basic data source for subsequent analysis. On this basis, the water body optical characteristic parameters of multi-temporal images are extracted, and the time series clustering algorithm is used to divide the optical dynamic zoning, and the optical parameter time series database of each zoning is established. Combined with environmental driving factors such as rainfall, wind speed and other variables, an optical parameter prediction model is constructed, which can dynamically adjust the predicted value of optical parameters according to external environmental changes.

[0028] In the optical dynamic zoning, a sampling scheme is designed, the initial number of sampling points is allocated according to the area proportion and fluctuation amplitude of the zoning, and the active learning strategy is used to supplement the sampling points in the high uncertainty area, so as to improve the coverage ability of the measured data to the dynamic process. Subsequently, based on the radiation transfer theory, a semi-analytical water depth inversion model is established, which expresses the water body attenuation coefficient as a function of optical component concentrations such as suspended solids and chlorophyll and spatio-temporal factors, and introduces seasonal and tidal correction factors to make the model adapt to the dynamic changes of water body state. At the same time, a long short-term memory network optical parameter time series prediction module is constructed, which predicts the concentrations of suspended solids and chlorophyll using historical optical parameter sequences, and adjusts the model parameters in real time according to the prediction results to realize the immediate response to the optical state of the water body.

[0029] To further separate the water body signal from the substrate signal, a substrate type classification system is established, and a spectral unmixing algorithm is used to decompose the observed spectrum into a linear combination of pure water body endmember spectrum and substrate endmember spectrum. By constructing a deep learning decoupling network, the attention mechanism module is used to weight and modulate the water body attenuation characteristics and substrate reflection characteristics, and finally output the water depth prediction value and substrate reflectivity prediction value. In addition, a Bayesian fusion method is used to spatio-temporally fuse the multi-period water depth inversion results, and a topographic continuity constraint condition is introduced to ensure that the fusion results conform to the topographic evolution law. By calculating the water depth change between adjacent periods and performing significance test, the real topographic evolution area is identified, and the environmental noise and actual change are effectively distinguished.

[0030] Finally, the independent verification sample set is used for stratified precision evaluation, and the model performance difference is analyzed in the dimensions of water depth range and substrate type. The whole-chain uncertainty analysis quantifies the error propagation of atmospheric correction, water level correction, etc., and constructs an uncertainty quantification model of ensemble learning to generate a pixel-level precision distribution map. Thus, through the systematic integration of multi-temporal data dynamic modeling and uncertainty quantification mechanism, the technical scheme forms a closed-loop dynamic adaptation framework, which embeds the spatio-temporal change compensation mechanism throughout the data acquisition to result verification process, effectively deals with the interference of spatio-temporal heterogeneity of nearshore water optical properties on water depth inversion, and solves the problem of parameter mismatch and precision decline of water depth inversion model caused by the dynamic changes of nearshore water optical properties in seasonal and tidal scales.

[0031] Embodiment two: please refer to Figure 1 , the specific way of step S1 is as follows: S1.1: Obtain multi-temporal multi-spectral remote sensing images of the target water area in the wet season, dry season and transition period, record the acquisition time and satellite overpass time of each image, and sequentially perform radiation calibration, atmospheric correction and geometric registration on the multi-temporal multi-spectral remote sensing images. Radiation calibration converts the original DN value to apparent reflectance, atmospheric correction uses a water self-adaptive correction algorithm based on the dark pixel principle, and geometric registration registers all images to a unified geographic coordinate system. Select one image as the radiation reference image, normalize other images to the radiation reference image through histogram matching method, and obtain the multi-temporal standardized image dataset; S1.2: Obtain the astronomical tide harmonic constant of the target water area, which includes the amplitude and phase angle of each component. According to the astronomical tide harmonic constant and the acquisition time of each image, the instantaneous tide level value is calculated using the astronomical tide prediction formula. The average sea level is set as the water level reference surface, and the water level deviation is calculated. For each pixel in the multi-temporal standardized image dataset, the pixel-level water level correction amount is established, and the apparent water depth observation value is subtracted by the water level deviation to obtain the true water depth value corrected to the average sea level reference surface.

[0032] In this embodiment: multi-temporal multi-spectral remote sensing image refers to multi-spectral image data of the target water area obtained in different time periods, which can be obtained by remote sensing platforms such as Landsat series satellites or Sentinel-2 satellites. Radiometric calibration refers to the process of converting the digital quantization value (DN value) recorded by the sensor into apparent reflectance with physical meaning, which aims to eliminate the influence of sensor response difference on image data. Atmospheric correction refers to the process of removing atmospheric scattering and absorption effects by a specific algorithm to obtain the true reflectance of the ground surface, wherein the water body adaptive correction algorithm based on the dark pixel principle can optimize the correction effect for water body characteristics. Geometric registration refers to the process of unifying image data obtained at different times to the same geographic coordinate system, which aims to eliminate geometric distortion caused by imaging angle, terrain undulation and other factors. Histogram matching method is a radiometric consistency correction technology, which adjusts the brightness distribution of the image to be consistent with the reference image, thereby ensuring the comparability of multi-temporal image data.

[0033] The astronomical tide harmonic constant refers to a set of parameters describing the law of tidal variation, which can be obtained by analyzing long-term tidal level observation data. The astronomical tide prediction formula is a mathematical model based on the principle of harmonic analysis, which can accurately calculate the tidal level at any time according to the harmonic constant and time variable. The pixel-level water level correction amount refers to the tidal level correction value calculated for each pixel position, which aims to reflect the influence of spatial position on tidal level and solve the problem of tidal flat area caused by terrain undulation.

[0034] This scheme realizes dynamic water level correction through a systematic operation process. First, multi-temporal multi-spectral remote sensing images of the target water area in the wet season, dry season and transition period are obtained, and the acquisition time and satellite overflight time of each image are recorded. This operation ensures that the image data covers the key hydrological period and provides a time dimension basis for capturing the seasonal changes in water optical properties. Subsequently, the multi-temporal multi-spectral remote sensing images are sequentially subjected to radiometric calibration, atmospheric correction and geometric registration, which eliminates sensor response difference, atmospheric scattering effect and geometric distortion, ensuring the consistency of image data in radiation and space. The water body adaptive correction algorithm based on the dark pixel principle optimizes the atmospheric influence correction for water body characteristics, avoiding the lack of applicability of general algorithms in nearshore waters. Select one image as the radiometric reference image and normalize other images to the radiometric reference image by histogram matching method, which ensures the comparability of reflectance data between different images and lays a reliable foundation for subsequent time series analysis.

[0035] On this basis, the instantaneous tide level value is calculated by acquiring the astronomical tide harmonic constants of the target water area and using the astronomical tide prediction formula in combination with the acquisition time of each image, so that the tide level prediction is strictly synchronized with the image acquisition time. The average sea level is set as the water level reference surface, a stable water depth reference standard is established, and the reference surface drift caused by tidal fluctuations is eliminated. The pixel-level water level correction amount is established for each pixel, the influence of the spatial position on the tide level is considered, and the problem of spatial heterogeneity of the tide level caused by the terrain undulation in the tidal flat area is solved. Finally, the apparent water depth observation value is subtracted by the water level deviation amount to obtain the true water depth value, so that the water depth data is accurately converted to the unified reference surface, and the reliability of the terrain monitoring in the dynamic water environment is significantly improved.

[0036] Through the above technical scheme, the problem of non-uniform water depth reference surface caused by tidal fluctuations is effectively solved, especially in the area with strong tidal power. Through the pixel-level tide level accurate calculation mechanism, the tide level deviation caused by ignoring the time details and spatial heterogeneity in the traditional method is avoided, and accurate input is provided for subsequent water depth inversion.

[0037] Embodiment three: please refer to Figure 1 , the specific way of step S2 is as follows: S2.1: Extract the blue light band reflectivity, green light band reflectivity and red light band reflectivity from the multi-temporal standardized image data set, calculate the blue-green band ratio, normalized suspended matter index and chlorophyll fluorescence peak index as water body optical characteristic parameters, arrange the water body optical characteristic parameters in time sequence for each pixel to construct a time sequence vector, calculate the similarity distance between the time sequence vectors of the pixels by using the dynamic time warping algorithm to form a similarity distance matrix, divide the pixels into optical dynamic partitions based on the similarity distance matrix by using the K-means clustering algorithm, and form the optical characteristic time variation curve by statistically averaging the water body optical characteristic parameters of each optical dynamic partition at each phase, and store it as an optical parameter time sequence database; S2.2: Obtain the environmental driving factor data corresponding to the multi-temporal standardized image data set, the environmental driving factor data including daily rainfall, daily runoff, daily average wind speed and daily average water temperature, correspond the environmental driving factor data to each optical dynamic partition by using the spatial interpolation method, convert the water body optical characteristic parameters into suspended matter concentration and chlorophyll concentration by using the empirical formula as the main control optical parameters, and use the environmental driving factor data as the independent variable and the main control optical parameters as the dependent variable to establish an optical parameter prediction model by using the multiple linear regression method. The optical parameter prediction model is established independently for each optical dynamic partition.

[0038] In this embodiment: the optical dynamic partition refers to the spatial unit divided based on the time series clustering results of water body optical characteristic parameters, which can be realized by K-means clustering algorithm or hierarchical clustering algorithm, aiming to reflect the spatial heterogeneity of water body optical characteristics. The fluctuation amplitude refers to the dynamic change intensity index quantified by the standard deviation of water body optical characteristic parameters, which can be realized by the standard deviation calculation method or the coefficient of variation analysis method in statistics, aiming to identify the key areas with intense dynamic changes. The initial sampling point number allocation weight refers to the sampling resource allocation basis calculated by combining the area proportion and the fluctuation amplitude, which can be realized by the weighted average method or the normalization processing method, aiming to ensure higher sampling density in high dynamic areas. The stratified random sampling method refers to the sampling strategy of random point distribution in each partition, which can be realized by the systematic sampling method or the stratified proportional sampling method, aiming to avoid human selection bias while ensuring the uniformity of spatial distribution.

[0039] This scheme first reads the optical dynamic partition results divided in the early stage, providing a spatial framework for targeted sampling. On this basis, the area proportion and the fluctuation amplitude of each partition are calculated to quantify the dynamic change intensity of each partition, which enables the identification of key areas with small area but intense changes. Then, the fluctuation amplitude is included in the weight calculation system to realize the optimal allocation of limited sampling resources, making the sampling resources preferentially cover the areas with intense changes and improving the overall sample's representation ability of water body optical characteristics. The stratified random sampling method is used to determine the initial sampling point position, which not only avoids human selection bias, but also ensures the uniformity of spatial distribution and reduces the influence of local anomalies on the sample set. Key measured parameters are obtained by depth sounder and water sample analysis, and a data table containing spatial and temporal dimension information is established, which facilitates subsequent accurate matching with remote sensing images. The sampling point data and the multi-temporal normalized image data set are spatio-temporally matched to extract the band reflectance and water body optical characteristic parameters of the corresponding pixels, forming a high-quality initial training sample set.

[0040] In the subsequent link, the initial training sample set is divided into training subset and validation subset, which provides independent data support for model evaluation. The prediction model established by the training subset can identify potential knowledge blind area. The prediction uncertainty is calculated by Gaussian process regression method, which can effectively quantify the prediction confidence and focus on the most uncertain area of the model. Through spatial clustering of candidate sampling points and selecting the most informative points in each cluster, the efficiency of supplementary sampling is maximized. By iteratively integrating new data, the sample set gradually adapts to the complex evolution of water body optical properties, significantly improving the representativeness of training data and the generalization ability of the model. The whole scheme solves the problem of insufficient sample representativeness in optical dynamic zoning through dynamic optimization of sampling resource allocation and active learning mechanism, ensuring that the measured data can accurately capture the spatio-temporal variation of water body optical properties, and providing a high-quality training basis for water depth inversion model.

[0041] Embodiment four: please refer to Figure 1 , the specific way of step S3 is as follows: S3.1: read the optical dynamic zoning result, calculate the area proportion and fluctuation amplitude of each optical dynamic zoning, the fluctuation amplitude is represented by the standard deviation of water body optical characteristic parameters, calculate the initial sampling point number allocation weight according to the area proportion and fluctuation amplitude, allocate the preset total sampling point number to each optical dynamic zoning according to the weight, determine the initial sampling point position by using stratified random sampling method, measure the water depth value by using the depth sounder, collect water sample and measure the suspended matter concentration value and chlorophyll concentration value, establish the initial sampling point data table containing geographic coordinates, sampling time and measured parameters, match the sampling point data with the multi-temporal standardized image data set in time and space, extract the band reflectance and water body optical characteristic parameters of corresponding pixels to form the initial training sample set; S3.2: divide the initial training sample set into training subset and validation subset, train the water depth inversion model and optical parameter inversion model by using the training subset, apply the trained model to the pixels not participating in the training to calculate the prediction value, calculate the prediction uncertainty of each pixel by using Gaussian process regression method, sort the pixels according to the prediction uncertainty, select the first N pixels as candidate supplementary sampling points, spatially cluster the candidate sampling points, select the pixel with the highest prediction uncertainty in each cluster as the actual supplementary sampling point, repeat the measurement process of step S3.1 to obtain supplementary sampling data, add the supplementary data to the initial training sample set to form the expanded training sample set.

[0042] In this embodiment: the optical dynamic zoning refers to the region divided based on the change law of water body optical characteristics in time and space, which can be realized by time series clustering algorithm or spatial clustering algorithm. The purpose of introducing this feature is to provide a scientific basis for the subsequent optimization of the distribution of sampling points, ensuring that the sampling points can cover the main change mode of water body optical characteristics. The fluctuation amplitude refers to the degree of change of water body optical characteristic parameters in the region, which can be measured by standard deviation, variance or other statistical indicators, for quantifying the instability of optical characteristics in the region. The initial sampling point number allocation weight refers to the sampling priority index calculated comprehensively according to the zoning area and fluctuation amplitude, which can be generated by linear weighting, nonlinear function or other mathematical methods to realize the reasonable allocation of sampling resources. The hierarchical random sampling method is a hybrid sampling strategy combining stratified sampling and random sampling, which aims to ensure the uniform distribution of sampling points while taking into account the characteristic differences of different partitions.

[0043] The above scheme realizes the goal of dynamic optimization sampling strategy through a series of ordered steps. First, based on the optical dynamic zoning results, the importance of each partition is quantified by calculating the area ratio and fluctuation amplitude, providing an objective basis for the allocation of sampling points. This process not only considers the spatial range of the partition, but also pays special attention to the change intensity of the optical characteristics, so that the high fluctuation region can obtain higher sampling density. Second, the hierarchical random sampling method is used to determine the initial sampling point position, which not only ensures the spatial uniformity of the sampling points, but also avoids the potential bias of pure random sampling. Then, the key measured parameters are obtained by using the depth sounder and water sample analysis, and are spatio-temporally matched with the multi-temporal standardized image dataset to form a high-quality initial training sample set. On this basis, the sample set is divided into training subset and validation subset, which are used for model training and performance evaluation respectively, ensuring the generalization ability of the model. The prediction uncertainty of each pixel is calculated using the Gaussian process regression method, and the candidate supplementary sampling points are selected based on this, which embodies the core idea of the active learning strategy. By spatial clustering of the candidate sampling points and selecting the pixel with the highest prediction uncertainty in each clustering cluster as the actual supplementary sampling point, the problem of excessive concentration of sampling points is effectively avoided, and the diversity and representativeness of spatial coverage are improved. Finally, by iteratively supplementing the sampling data, the training sample set is gradually optimized, significantly enhancing the adaptability and precision of the model in dynamic water body environment.

[0044] The above scheme solves the problem that the initial sampling points cannot fully cover the high fluctuation region of optical characteristics by dynamically adjusting the sampling strategy, and supplements the sampling in the region with high model prediction uncertainty, thereby significantly improving the precision and reliability of the water depth inversion model in the spatio-temporally heterogeneous water body environment.

[0045] Embodiment five: please refer to Figure 1 , the specific way of step S4 is as follows: S4.1: Establish a semi-analytical water depth inversion model based on radiation transfer theory, express the water surface off-water radiation as the sum of water scattering contribution and bottom reflection contribution, express the water body attenuation coefficient as the linear combination of suspended matter absorption coefficient, chlorophyll absorption coefficient and colored dissolved organic matter absorption coefficient, each absorption coefficient has a power function relationship with the corresponding concentration, introduce seasonal correction factor and tidal correction factor to modify the absorption coefficient baseline value, extract the measured data from the extended training sample set, solve the minimum value of the objective function to obtain the optimal parameter value by using nonlinear least squares method, set parameters for each optical dynamic partition, and establish the correspondence table of optical dynamic partition number and model parameter group; S4.2: Construct an optical parameter time series prediction module of long short-term memory network, which includes an input layer, 2 to 4 layers of LSTM hidden layer and a fully connected output layer, the input layer receives historical 3 to 12 time phase water optical characteristic parameters and environmental driving factors, generates LSTM training data set by time sliding window method, trains the model by using Adam optimization algorithm, inputs the historical data before the target time phase into the trained model to obtain the prediction value of suspended matter concentration and chlorophyll concentration, calculates the dynamic water body attenuation coefficient according to the prediction value and the space-time correction factor, and substitutes it into the semi-analytical water depth inversion model to calculate the water depth of the pixel.

[0046] In this embodiment: the water surface off-water radiation refers to the water surface reflection signal obtained through remote sensing image, which can be realized by using multispectral sensor to capture reflectivity of different wave bands. The water scattering contribution refers to the light scattering effect caused by the particles in the water body, and the bottom reflection contribution refers to the reflection signal from the water bottom, which together constitute the main part of the water surface off-water radiation, and the purpose is to accurately separate the water optical process and the physical action of bottom reflection, and lay a theoretical foundation for water depth calculation.

[0047] The water body attenuation coefficient refers to the degree of absorption and scattering of light in the water body during the propagation process, which can be quantified by the linear combination of suspended matter absorption coefficient, chlorophyll absorption coefficient and colored dissolved organic matter absorption coefficient. This expression can reflect the independent contribution of different optical components to light attenuation, and the purpose is to make the model respond to the complex changes of water composition. In addition, the absorption coefficient has a power function relationship with the corresponding concentration, which reflects the nonlinear relationship between concentration and optical characteristics, and is more in line with the actual water behavior.

[0048] The seasonal correction factor and the tidal correction factor are the key parameters for dynamic compensation of the absorption coefficient baseline value, which can be realized by statistical analysis or empirical formula based on historical observation data. The purpose is to dynamically adjust the seasonal land source input and the suspended matter fluctuation caused by tides, and avoid systematic deviation caused by static parameters.

[0049] The extended training sample set refers to a collection of measured data containing multiple time phases and multiple regions obtained through preliminary sampling and supplementary sampling. The sample distribution can be optimized by supplementing sampling points in high uncertainty areas using an active learning strategy. The purpose is to optimize model accuracy using measured information and ensure that the parameters match the actual water body state.

[0050] The correspondence table between the optical dynamic zoning number and the model parameter group refers to assigning independent model parameter groups to different optical dynamic zones according to their characteristics. It can be stored and called in the form of a database or a mapping table. The purpose is to consider the spatial heterogeneity of water body optical characteristics, prevent errors at the partition boundary when using global unified parameters, and improve inversion efficiency.

[0051] The long short-term memory network is a deep learning model suitable for time series prediction. It captures long-term temporal dependencies through multiple LSTM hidden layers. The input layer receives historical 3 to 12 time phase water body optical characteristic parameters and environmental driving factors. These data can be used to generate continuous time series samples using a time sliding window method to enhance the model's fitting ability for dynamic processes. The Adam optimization algorithm is used to accelerate model convergence and stabilize the training process. The purpose is to improve prediction accuracy and training efficiency.

[0052] The above technical solutions achieve dynamic expression of water body attenuation coefficient through deep integration of physical models and time series prediction. First, the semi-analytical water depth inversion model based on radiation transfer theory decomposes the water surface off-water radiation into water scattering contribution and bottom reflection contribution, accurately separating the physical effects of water optical process and bottom reflection. Second, by expressing the water body attenuation coefficient as a linear combination of suspended matter absorption coefficient, chlorophyll absorption coefficient, and colored dissolved organic matter absorption coefficient, and introducing a power function relationship, the independent contribution of different optical components to light attenuation is quantified. On this basis, the introduction of seasonal correction factor and tidal correction factor effectively compensates for the seasonal land source input and the suspended matter fluctuation caused by tides, avoiding systematic bias caused by static parameters.

[0053] The measured data of the extended training sample set optimizes the objective function through nonlinear least squares method to obtain the optimal parameter value, ensuring that the model parameters match the actual water body state. Parameter setting is performed for each optical dynamic zone, and a correspondence table between the optical dynamic zoning number and the model parameter group is established, which not only considers the spatial heterogeneity of water body optical characteristics, but also improves the inversion efficiency. In terms of optical parameter time series prediction, the long short-term memory network integrates historical data and environmental driving factors to mine the time series dependency of optical parameters, generating dynamic water body attenuation coefficient to adapt to water body state changes in real time.

[0054] By the technical solution, the model mismatch problem caused by the space-time variation of the optical characteristics of the water body is solved, and the adaptability of the model in the dynamic water body environment is fundamentally improved. For example, during the season replacement or tidal fluctuation, the introduction of the dynamic water body attenuation coefficient effectively avoids the phenomenon that the high suspended substance concentration in the flood season is misjudged as the increase of water depth or the false shallow phenomenon caused by the increase of chlorophyll absorption during the algal bloom, and significantly improves the accuracy and reliability of the water depth inversion.

[0055] Embodiment six: please refer to Figure 1 , the specific way of step S5 is as follows: S5.1: a bottom type classification system including sandy bottom, muddy bottom, rocky bottom and seagrass bed bottom is established, bottom samples are collected at sampling points in shallow water areas with a water depth less than 5 meters, a spectral reflectance in a wave band of 400 nanometers to 900 nanometers is measured by a laboratory spectrometer, a bottom type spectral library is established, a pure water body spectral model is established for pixels in deep water areas with a water depth greater than 20 meters, a linear spectral unmixing algorithm is used to decompose the observed spectral vector into a linear weighted combination of the pure water body endmember spectrum and the bottom endmember spectrum, a least squares method is used to solve the mixing proportion coefficient, the mixing proportion coefficient is decomposed into a water body spectral contribution vector and a bottom spectral contribution vector, and a spectral angle mapping method is used for bottom type identification; S5.2: a deep learning decoupling network including a water body attenuation feature extraction branch and a bottom reflectance feature extraction branch is constructed, the two branches respectively include 3 to 5 layers of convolution layers and pooling layers to output feature vectors, the feature vectors are weighted and modulated through an attention mechanism module, a feature fusion layer is used to splice the feature vectors to form a fused feature vector, a fully connected decoding layer is used to output a water depth prediction value, a bottom reflectance prediction value and a bottom type probability distribution, a composite loss function including a data-driven loss term and a physical constraint loss term is constructed, an Adam optimization algorithm is used to train the network, and the trained network is applied to the pixels in the shallow water area for water depth inversion and bottom type identification.

[0056] In the embodiment: the bottom type classification system refers to a classification framework designed according to the spectral reflectance characteristics difference of different bottoms in the visible light to near-infrared wave band, which can be realized by field sampling combined with remote sensing image analysis, and the purpose is to customize the process for each bottom type to reduce systematic bias. The shallow water sampling point refers to the area with a water depth less than 5 meters, because the contribution of the bottom reflectance signal to the water surface spectrum cannot be ignored in this depth range, it can be determined by combining field sampling with remote sensing image, and the purpose is to ensure the accuracy of the model input data.

[0057] Spectral reflectance refers to the reflectivity of the substrate in a specific waveband range, which can be measured by a laboratory spectrometer or a high-precision field spectrometer, aiming to comprehensively capture the spectral information of the substrate characteristics. The substrate type spectral library refers to a database that stores standardized substrate spectral data, which can be achieved by spectral measurement and sorting of multiple substrate samples, aiming to provide reliable endmember reference for subsequent spectral unmixing.

[0058] The pure water spectral model refers to a water optical property benchmark model based on deep water pixels, which can be achieved by statistical modeling of pixels with water depth greater than 20 meters, aiming to provide a pure water optical property reference. The linear spectral unmixing algorithm refers to a technique for decomposing shallow water spectra into water transmission and substrate reflection signals, which can be solved by least squares method or other optimization algorithms, aiming to separate the two signal sources to improve the inversion accuracy.

[0059] The spectral angle mapping method refers to a classification method based on the similarity measurement of spectral vector angles, which can be used to identify substrate types by calculating the angles between spectral vectors, aiming to reduce misjudgment. The deep learning decoupling network refers to a dual-branch network structure including water attenuation feature extraction branch and substrate reflection feature extraction branch, which can be implemented by multi-layer convolutional neural network, aiming to model water optical properties and substrate reflection properties independently.

[0060] The composite loss function refers to an optimization objective function combining data-driven loss term and physical constraint loss term, which can be achieved by embedding prior knowledge such as radiative transfer equation, aiming to prevent the model from deviating from the physical law.

[0061] The above technical solutions solve the key problem of substrate reflection interference in shallow water depth inversion by systematically separating water and substrate signals. First, establishing a substrate type classification system can classify the spectral reflectance characteristics of different substrate types such as sandy bottom, muddy bottom, rocky bottom and seagrass bed bottom, avoiding systematic bias caused by substrate signal mixing in a unified model.

[0062] Selecting sampling points in shallow water area and collecting substrate samples, combined with laboratory spectrometer measurement of spectral reflectance in 400-900 nm waveband, can comprehensively capture the spectral information of substrate characteristics, providing reliable reference for subsequent spectral unmixing.

[0063] The deep water pixel is identified to establish a pure water body spectrum model, which provides a pure water body optical property benchmark for spectral unmixing in shallow water areas. In addition, the linear spectral unmixing algorithm is used to decompose the shallow water spectrum into water body and bottom signal, and the mixing proportion coefficient is solved by the least square method, which can effectively separate the two signal sources. Finally, a deep learning decoupling network is constructed, which independently models the water body attenuation and bottom reflectance characteristics through a double-branch design, and combines a composite loss function for optimization, which can significantly improve the accuracy and reliability of water depth inversion and bottom classification.

[0064] Through the above technical solutions, not only the problem of bottom reflectance interference in shallow water depth inversion is solved, but also the synchronous inversion of water depth and bottom type is realized, which improves the practicality and scientificity of the overall solution.

[0065] Embodiment seven: please refer to Figure 1 , the specific way of step S6 is as follows: S6.1: Collect multi-period water depth inversion results, extract the water depth inversion value of each time phase for each pixel to form time series data, establish a Bayesian fusion framework, take the multi-period water depth inversion value as observation data, construct an observation model and a state transition model, introduce spatial smoothness constraint and slope rationality constraint as Bayesian prior distribution, encode the spatial smoothness constraint as the prior distribution of Markov random field, encode the slope rationality constraint as the slope prior distribution, solve the posterior probability distribution by Markov chain Monte Carlo method or variational Bayesian method, extract the posterior mean value from the posterior distribution as the fused water depth value, and extract the posterior standard deviation as the fusion uncertainty index; S6.2: Select two adjacent time interval fused water depth products, calculate the water depth change for each pixel, calculate the uncertainty of water depth change according to the error propagation law, construct the significance test statistic as the ratio of water depth change to its uncertainty, set the significance level to 0.05 or 0.01 to determine the critical value, and determine the significant change pixel when the absolute value of the test statistic is greater than the critical value, perform spatial clustering analysis on the significance test results, set the minimum cluster area threshold to remove fragmented areas, calculate the area, average water depth change and change type of each topographic evolution area, and establish a topographic evolution area database.

[0066] In this embodiment: the Bayesian fusion framework refers to a probability inference method based on Bayes theorem, which can be implemented in various mathematical modeling ways, such as constructing a complex probabilistic graph model or using approximate inference algorithms. The purpose of introducing this framework is to integrate multi-period water depth inversion results, thereby effectively suppressing noise interference in single-period inversion and improving estimation stability.

[0067] The spatial smoothness constraint can be understood as a prior assumption that reflects the continuity of natural terrain, which can be expressed by a probability model such as a Markov random field, and the purpose is to eliminate the influence of random noise on the inversion result. The slope rationality constraint is to ensure that the terrain change conforms to the natural law and avoid unreasonable abrupt changes, which can be achieved by defining a reasonable slope distribution function.

[0068] The significance test statistic refers to a quantitative index for evaluating whether the water depth change has statistical significance, which can be constructed by various statistical methods such as t-test or z-test. The purpose of introducing this statistic is to distinguish between real terrain evolution and random fluctuations, thereby improving the reliability of terrain monitoring.

[0069] The above scheme provides a time series basis for subsequent analysis by collecting multi-period water depth inversion results, making the historical change pattern of each spatial location individualized, thereby reflecting the influence of dynamic drift of water body optical properties on the inversion result. A Bayesian fusion framework is established to use multi-period water depth inversion values as observation data, fully utilize all available information to avoid single-period limitations, construct observation model and state transition model to describe the relationship between inversion values and real water depth and the evolution law of water depth over time, and enhance the physical rationality of the model. Spatial smoothness constraint and slope rationality constraint are introduced as Bayesian prior distribution, which effectively models the spatial correlation between adjacent pixels to suppress random noise and ensure that the terrain change conforms to the natural slope law and avoids unreasonable abrupt changes.

[0070] The Markov Chain Monte Carlo method or the variational Bayesian method is used to solve the posterior probability distribution, which efficiently processes high-dimensional probability distribution to accurately estimate the fusion result, provides optimal estimation to reduce bias, and quantifies the estimation reliability of each pixel.

[0071] The two-period fusion water depth products selected with adjacent time intervals focus on the changes in the continuous period, avoiding the interference of long-term cumulative errors, and calculating the water depth change for each pixel directly measures the terrain evolution amplitude, ensuring that the change evaluation considers the transmission of original inversion errors.

[0072] The significance test statistic is constructed as the ratio of water depth change and its uncertainty, which effectively distinguishes between real terrain evolution and random fluctuations. Spatial clustering analysis is performed on the significance test results, merging adjacent significant change pixels into regions, which conforms to the characteristics of terrain evolution usually occurring in patches, eliminating small-scale noise points to focus on meaningful evolution regions. The area, average water depth change and change type of each terrain evolution region are calculated to quantify the evolution characteristics to support subsequent analysis, and a terrain evolution region database system is established to store information, facilitating long-term dynamic monitoring and management, thereby realizing reliable identification and quantitative evaluation of terrain evolution.

[0073] The technical scheme solves the problem in the process that a Bayesian fusion framework of multi-period water depth inversion results is used to integrate space-time information and improve the stability of water depth estimation. The scheme effectively integrates multi-period water depth inversion results, solves the problem that single-period inversion results are unstable in a dynamic water environment, and realizes reliable identification and quantitative evaluation of topographic evolution.

[0074] Embodiment eight: please refer to Figure 1 , the specific way of step S7 is as follows: S7.1: Collect an independent verification sample set, the number of samples in the independent verification sample set is 20% to 30% of the total sample size, layer the independent verification sample set according to the water depth range, bottom type and optical dynamic partition, extract the water depth inversion value of the pixel corresponding to the verification sample point position from the fusion water depth product, calculate the root mean square error, mean absolute error, mean deviation and determination coefficient of each layer subset, divide the water depth inversion process into five links: atmospheric correction, water level correction, optical parameter estimation, model structure selection and parameter setting, calculate the uncertainty of each link, synthesize the overall uncertainty according to the error propagation theory, and calculate the contribution rate of each link to the overall uncertainty; S7.2: Build an ensemble learning uncertainty quantification model including 5 to 20 base models, the base model types include semi-analytical model, deep learning network, random forest regression, support vector regression and gradient boosting regression, apply the trained base model to each pixel to calculate the water depth prediction value, calculate the mean value of the base model prediction value as the ensemble prediction value, calculate the standard deviation as the prediction uncertainty, use the independent verification sample set to establish an uncertainty calibration function, use piecewise linear interpolation or cubic spline interpolation method, apply the calibration function to each pixel to generate calibrated prediction error estimate value, generate a pixel-level precision distribution map, calculate the confidence interval according to the confidence level, and generate a confidence interval width distribution map.

[0075] In this embodiment: the independent verification sample set refers to a set of independent data used to evaluate the accuracy of water depth inversion results, which can be realized by using field measurement data or high-precision reference data, and the purpose is to verify the generalization ability of the model through independent data.

[0076] Layered processing refers to classifying the verification sample set according to environmental characteristics such as water depth range, bottom type and optical dynamic partition, which can be realized by clustering algorithm or rule division, and the purpose is to perform differential precision evaluation for different environmental conditions.

[0077] Error propagation theory refers to a mathematical method for describing the error transmission law in a multi-step processing process, which can be realized based on Taylor expansion or Monte Carlo simulation, and the purpose is to reveal the influence path of errors in each link on the final result.

[0078] The integrated learning uncertainty quantification model refers to capturing the uncertainty of a complex system through the combination of multiple base models, which can be realized in a weighted average or Bayesian inference manner, aiming to enhance the robustness of uncertainty estimation.

[0079] The technical solution above effectively solves the error distribution misalignment problem caused by environmental dynamics through a systematic hierarchical verification and full-chain error analysis mechanism. First, the hierarchical processing of the independent verification sample set can conduct differentiated precision evaluation for specific environmental conditions, avoiding the global average from masking local precision degradation, especially adapting to the drift of water optical properties caused by tides and seasonal changes.

[0080] Secondly, the water depth inversion process is divided into five key links and the uncertainty is calculated respectively, which reveals the error transmission path of each link based on the radiation transfer theory and error propagation law, so that the uncertainty quantification covers the complete chain from the original data to the final product. On this basis, the total uncertainty is synthesized through the error propagation theory to ensure the physical consistency of the uncertainty estimation and avoid the error amplification or cancellation caused by simple superposition. In addition, the integrated learning uncertainty quantification model uses the average and standard deviation of the prediction values of multiple base models to capture the complex influence of dynamic changes of water optical parameters, enhancing the robustness of uncertainty estimation. Finally, the calibration function is applied through piecewise linear interpolation or cubic spline interpolation method to finely adjust the nonlinear error characteristics under different confidence levels, ensuring that the calibration process adapts to the local characteristics of water depth and bottom material changes.

[0081] Through the technical solution above, the pixel-level precision distribution map and confidence interval width distribution map are generated to provide users with location-dependent reliability evaluation and support for accurately identifying high-uncertainty areas in dynamic water environments, thereby significantly improving the practical value of water depth inversion results in topography monitoring.

[0082] The present application also includes a water depth inversion system based on multi-spectral remote sensing images, please refer to Figure 2 The system includes an image preprocessing module, an optical zoning module, a sample collection module, a model inversion module, a bottom material decoupling module, a Bayesian fusion module, and a precision evaluation module. The image preprocessing module is used to obtain multi-temporal multi-spectral remote sensing images in different seasons and tidal states, and to perform radiation calibration, atmospheric correction, geometric registration, and water level dynamic correction on the images to unify the images to the average sea level reference. The optical zoning module is used to extract water optical characteristic parameters, divide the optical dynamic zones using a time series clustering algorithm, establish a time series database of zone optical parameters, and construct an optical parameter prediction model. The sample collection module is used for collecting water depth and optical component measured data according to the optical zoning design sampling scheme, and supplementing sampling points in a high uncertainty area by using an active learning strategy; The model inversion module is used for establishing a semi-analytical water depth inversion model, combining a time series prediction to output dynamic adjustment of model parameters to realize water depth calculation; The bottom decoupling module is used for establishing a bottom classification system, separating water body and bottom signals by using a spectral unmixing algorithm, and realizing joint inversion of bottom reflectivity and water depth by using a deep learning network; The Bayesian fusion module is used for spatio-temporal fusion of multi-period water depth inversion results, introducing spatial smoothing and slope constraints, calculating water depth changes, and performing significance test to identify topographic evolution areas; The precision evaluation module is used for carrying out hierarchical precision evaluation and uncertainty analysis based on an independent verification sample set, constructing an integrated learning uncertainty quantification model to generate a pixel-level precision distribution and a confidence interval map.

[0083] The contents not described in detail in the specification belong to the prior art known to those skilled in the art.

[0084] Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for part of the technical features, and any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for water depth inversion based on multispectral remote sensing imagery, characterized in that: The specific steps are as follows: S1: Acquire multi-temporal multispectral remote sensing images of the target water area under different seasons and tidal conditions, perform radiometric calibration, atmospheric correction and geometric registration on the multi-temporal multispectral remote sensing images, perform dynamic water level correction based on astronomical tide model, and unify the multi-temporal multispectral remote sensing images to mean sea level datum. S2: Extract water body optical feature parameters from multi-temporal multispectral remote sensing images, use time series clustering algorithm to divide optical dynamic zones, establish time series database of optical parameters for each optical dynamic zone, and establish optical parameter prediction model based on environmental driving factors; S3: Based on the optical dynamic zoning design sampling scheme, the measured data of water depth, suspended matter concentration and chlorophyll concentration are collected in each optical dynamic zoning, and an active learning strategy is used to supplement sampling points in areas with high uncertainty. S4: Establish a semi-analytical water depth inversion model, express the water body attenuation coefficient as a function of optical component concentration and spatiotemporal factor, construct a long short-term memory network optical parameter time series prediction module, and adjust the parameters of the semi-analytical water depth inversion model according to the output of the optical parameter time series prediction module; S5: Establish a classification system for sediment types, use a spectral unmixing algorithm to separate water signals and sediment signals, construct a deep learning decoupling network, and use the deep learning decoupling network to retrieve sediment reflectivity and water depth; S6: The Bayesian fusion method is used to perform spatiotemporal fusion of multi-period water depth inversion results, introduce topographic continuity constraints, calculate water depth changes in adjacent periods and perform significance tests to identify topographic evolution areas. S7: Utilize independent validation sample sets to perform hierarchical accuracy assessment and full-chain uncertainty analysis, construct an ensemble learning uncertainty quantification model, and generate a pixel-level accuracy distribution map.

2. The water depth inversion method based on multispectral remote sensing imagery according to claim 1, characterized in that: The specific method of step S1 is as follows: S1.1: Acquire multi-temporal multispectral remote sensing images of the target water area during the high-water season, low-water season, and transition period. Record the acquisition time and satellite transit time of each image. Perform radiometric calibration, atmospheric correction, and geometric registration on the multi-temporal multispectral remote sensing images in sequence. Radiometric calibration converts the original DN values ​​into apparent reflectance. Atmospheric correction adopts a water body adaptive correction algorithm based on the dark pixel principle. Geometric registration registers all images to a unified geographic coordinate system. Select one image as the radiometric reference image and normalize the other images to the radiometric reference image through histogram matching method to obtain a multi-temporal standardized image dataset. S1.2: Obtain the astronomical tidal harmonic constant of the target water area. The astronomical tidal harmonic constant includes the amplitude and lag angle of each tidal component. Calculate the instantaneous tidal level using the astronomical tidal forecast formula based on the astronomical tidal harmonic constant and the acquisition time of each image. Set the mean sea level as the water level reference surface and calculate the water level deviation. Establish a pixel-level water level correction for each pixel in the multi-temporal standardized image dataset. Subtract the water level deviation from the apparent water depth observation value to obtain the true water depth value corrected to the mean sea level reference surface.

3. The water depth inversion method based on multispectral remote sensing imagery according to claim 2, characterized in that: The specific method of step S2 is as follows: S2.1: Extract blue band reflectance, green band reflectance, and red band reflectance from the multi-temporal standardized image dataset. Calculate the blue-green band ratio, normalized suspended matter index, and chlorophyll fluorescence peak index as optical feature parameters of the water body. For each pixel, arrange the optical feature parameters of the water body in chronological order to construct a time series vector. Use a dynamic time warping algorithm to calculate the similarity distance between the time series vectors of pixels to form a similarity distance matrix. Based on the similarity distance matrix, use the K-means clustering algorithm to divide the pixels into optical dynamic partitions. Statistically calculate the average value of the optical feature parameters of the water body in each optical dynamic partition at each time phase to form the optical feature time change curve, and store it as an optical parameter time series database. S2.2: Obtain environmental driving factor data corresponding to the multi-temporal standardized image dataset. The environmental driving factor data includes daily rainfall, daily runoff, daily average wind speed, and daily average water temperature. The environmental driving factor data is mapped to each optical dynamic zone through spatial interpolation. The optical characteristic parameters of the water body are converted into suspended solids concentration and chlorophyll concentration values ​​as master optical parameters through empirical formulas. The environmental driving factor data is used as independent variables and the master optical parameters are used as dependent variables. The optical parameter prediction model is established using the multiple linear regression method. An independent optical parameter prediction model is established for each optical dynamic zone.

4. The water depth inversion method based on multispectral remote sensing imagery according to claim 3, characterized in that: The specific method of step S3 is as follows: S3.1: Read the results of optical dynamic zoning, calculate the area ratio and fluctuation amplitude of each optical dynamic zoning, the fluctuation amplitude is characterized by the standard deviation of the water body optical characteristic parameters, calculate the weight of the initial sampling point number allocation based on the area ratio and fluctuation amplitude, allocate the preset total number of sampling points to each optical dynamic zoning according to the weight, use a stratified random sampling method to determine the location of the initial sampling points, use a depth sounder to measure the water depth, collect water samples and measure the suspended solids concentration and chlorophyll concentration, establish an initial sampling point data table containing geographic coordinates, sampling time and measured parameters, perform spatiotemporal matching of the sampling point data with the multi-temporal standardized image dataset, extract the band reflectance and water body optical characteristic parameters of the corresponding pixels to form the initial training sample set; S3.2: Divide the initial training sample set into a training subset and a validation subset. Use the training subset to train the water depth inversion model and the optical parameter inversion model. Apply the trained model to the pixels that did not participate in the training to calculate the predicted values. Use the Gaussian process regression method to calculate the prediction uncertainty of each pixel. Sort the pixels according to the prediction uncertainty and select the top N pixels as candidate supplementary sampling points. Perform spatial clustering on the candidate sampling points. Select the pixel with the highest prediction uncertainty in each cluster as the actual supplementary sampling point. Repeat the measurement process of step S3.1 to obtain supplementary sampling data. Add the supplementary data to the initial training sample set to form an extended training sample set.

5. The water depth inversion method based on multispectral remote sensing imagery according to claim 4, characterized in that: The specific method of step S4 is as follows: S4.1: A semi-analytical water depth inversion model is established based on radiative transfer theory. The surface radiation is expressed as the sum of the water scattering contribution and the bottom sediment reflection contribution. The water attenuation coefficient is expressed as a linear combination of the suspended matter absorption coefficient, chlorophyll absorption coefficient, and colored dissolved organic matter absorption coefficient. Each absorption coefficient has a power function relationship with the corresponding concentration. Seasonal correction factors and tidal correction factors are introduced to correct the baseline value of the absorption coefficient. Measured data are extracted from the extended training sample set. The minimum value of the objective function is solved by nonlinear least squares method to obtain the optimal parameter value. Parameters are set for each optical dynamic zone. A correspondence table between the optical dynamic zone number and the model parameter group is established. S4.2: Construct a temporal prediction module for optical parameters of a Long Short-Term Memory (LSTM) network. The LSTM network includes an input layer, 2 to 4 LSTM hidden layers, and a fully connected output layer. The input layer receives water optical characteristic parameters and environmental driving factors from 3 to 12 historical time phases. An LSTM training dataset is generated using a time sliding window method. The model is trained using the Adam optimization algorithm. Historical data before the target time phase is input into the trained model to obtain predicted values ​​for suspended solids concentration and chlorophyll concentration. The dynamic water attenuation coefficient is calculated based on the predicted values ​​and the spatiotemporal correction factor. The coefficient is then substituted into the semi-analytical water depth inversion model to perform water depth inversion calculations on the pixels.

6. The water depth inversion method based on multispectral remote sensing imagery according to claim 5, characterized in that: The specific method of step S5 is as follows: S5.1: Establish a classification system for sediment types including sandy bottom, muddy bottom, rocky bottom, and seagrass bed bottom. Collect sediment samples from sampling points in shallow water areas with a depth of less than 5 meters. Measure the spectral reflectance in the 400 nm to 900 nm band using a laboratory spectrometer. Establish a sediment type spectral library. Identify pixels in deep water areas with a depth greater than 20 meters to establish a pure water spectral model. For pixels in shallow water areas, use a linear spectral unmixing algorithm to decompose the observed spectral vector into a linear weighted combination of the pure water endmember spectrum and the sediment endmember spectrum. Solve for the mixing ratio coefficient using the least squares method. Based on the mixing ratio coefficient, decompose it into a water spectral contribution vector and a sediment spectral contribution vector. Use the spectral angle mapping method to identify the sediment type. S5.2: Construct a deep learning decoupled network including a water body attenuation feature extraction branch and a bottom sediment reflectance feature extraction branch. The two branches each include 3 to 5 convolutional layers and pooling layers to output feature vectors. The feature vectors are weighted and modulated through an attention mechanism module. The feature fusion layer concatenates the feature vectors to form a fused feature vector. The fully connected decoding layer outputs the predicted water depth, the predicted bottom sediment reflectance, and the probability distribution of bottom sediment type. Construct a composite loss function including a data-driven loss term and a physical constraint loss term. Train the network using the Adam optimization algorithm. Apply the trained network to shallow water pixels for water depth inversion and bottom sediment type identification.

7. The water depth inversion method based on multispectral remote sensing imagery according to claim 6, characterized in that: The specific method of step S6 is as follows: S6.1: Collect water depth inversion results from multiple periods, extract water depth inversion values ​​for each time phase for each pixel to form time series data, establish a Bayesian fusion framework, use the water depth inversion values ​​from multiple periods as observation data, construct an observation model and a state transition model, introduce spatial smoothness constraints and slope rationality constraints as Bayesian prior distributions, encode the spatial smoothness constraints as Markov random field prior distributions, encode the slope rationality constraints as slope prior distributions, use Markov chain Monte Carlo method or variational Bayesian method to solve the posterior probability distribution, extract the posterior mean from the posterior distribution as the fused water depth value, and extract the posterior standard deviation as the fusion uncertainty index; S6.2: Select two adjacent time intervals of fused water depth products, calculate the water depth change for each pixel, calculate the uncertainty of the water depth change according to the error propagation law, construct a significance test statistic as the ratio of the water depth change to its uncertainty, set the significance level to 0.05 or 0.01 to determine the critical value, and judge the pixel as a significantly changed pixel when the absolute value of the test statistic is greater than the critical value. Perform spatial cluster analysis on the significance test results, set a minimum cluster area threshold to remove fragmented areas, calculate the area, average water depth change and change type of each terrain evolution area, and establish a terrain evolution area database.

8. The water depth inversion method based on multispectral remote sensing imagery according to claim 7, characterized in that: The specific method of step S7 is as follows: S7.1: Collect an independent validation sample set, with the number of samples in the independent validation sample set being 20% ​​to 30% of the total sample size. The independent validation sample set is stratified according to water depth range, seabed type, and optical dynamic zoning. The water depth inversion values ​​of the corresponding pixels of the validation sample point locations are extracted from the fused water depth product. The root mean square error, mean absolute error, mean deviation, and coefficient of determination of each stratified subset are calculated. The water depth inversion process is decomposed into five stages: atmospheric correction, water level correction, optical parameter estimation, model structure selection, and parameter setting. The uncertainty of each stage is calculated separately. The overall uncertainty is synthesized according to the error propagation theory, and the contribution rate of each stage to the overall uncertainty is calculated. S7.2: Construct an ensemble learning uncertainty quantification model comprising 5 to 20 base models. Base model types include semi-analytical models, deep learning networks, random forest regression, support vector regression, and gradient boosting regression. Apply the trained base models to each pixel to calculate the water depth prediction value. Calculate the average of the base model prediction values ​​as the ensemble prediction value and the standard deviation as the prediction uncertainty. Establish an uncertainty calibration function using an independent validation sample set. Use piecewise linear interpolation or cubic spline interpolation methods to apply the calibration function to each pixel to generate a calibrated prediction error estimate. Generate a pixel-level accuracy distribution map. Calculate the confidence interval based on the confidence level and generate a confidence interval width distribution map.

9. A water depth inversion system based on multispectral remote sensing imagery, characterized in that: The water depth inversion system based on multispectral remote sensing images is used to execute the water depth inversion method based on multispectral remote sensing images as described in any one of claims 1 to 8. The system includes an image preprocessing module, an optical partitioning module, a sample acquisition module, a model inversion module, a seabed decoupling module, a Bayesian fusion module, and an accuracy evaluation module. The image preprocessing module is used to acquire multi-temporal multispectral remote sensing images of different seasons and tidal states, perform radiometric calibration, atmospheric correction, geometric registration and water level dynamic correction on the images, and unify the images to mean sea level reference. The optical partitioning module is used to extract optical feature parameters of water bodies, and uses a time series clustering algorithm to divide the dynamic optical partitions, establish a time series database of partition optical parameters, and construct an optical parameter prediction model. The sample acquisition module is used to design sampling schemes based on optical zoning, collect measured data on water depth and optical components, and use an active learning strategy to supplement sampling points in areas with high uncertainty. The model inversion module is used to establish a semi-analysis water depth inversion model, and dynamically adjusts the model parameters in combination with time series prediction output to realize water depth calculation; The sediment decoupling module is used to establish a sediment classification system. It uses a spectral unmixing algorithm to separate water and sediment signals and uses a deep learning network to achieve joint inversion of sediment reflectance and water depth. The Bayesian fusion module is used to perform spatiotemporal fusion of multi-period water depth inversion results, introduce spatial smoothing and slope constraints, calculate water depth changes and perform significance tests to identify areas of topographic evolution; The accuracy assessment module is used to conduct hierarchical accuracy assessment and uncertainty analysis based on independent validation sample sets, and to build an integrated learning uncertainty quantification model to generate pixel-level accuracy distribution and confidence interval plots.

Images