Remote sensing shallow sea water depth inversion method and device, electronic equipment and storage medium

Abstract

The invention discloses a remote sensing shallow sea water depth inversion method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring a satellite remote sensing image and actually measured water depth data of a to-be-measured area; performing correction preprocessing on the satellite remote sensing image to obtain spectral data; dividing the actually measured water depth data based on a preset proportion to obtain a training set and a test set; constructing a partial least squares regression model according to the training set and the spectral data of each wave band corresponding to the training set, predicting the spectral data, obtaining a prediction result, and performing prediction verification; when the prediction verification result does not meet the preset condition, the training set and the test set are divided again; the previous steps are cycled until the prediction verification result meets the preset condition; a support vector machine model is constructed through a parrot optimization algorithm based on the training set and the test set; and performing water depth inversion of the target area by using the support vector machine model. The method can accurately perform shallow sea water inversion, and can be widely applied to the technical field of data inversion.

Description

Technical Field

[0001] The present invention relates to the technical field of data inversion, and in particular to a remote sensing shallow sea water depth inversion method, device, electronic equipment and storage medium. Background Art

[0002] Shallow sea depth is one of the important parameters of the coastal environment. It plays an important role in navigation safety, fishery farming, nearshore engineering and coastal ecological environment protection. Traditional depth measurement methods such as field measurements and ship-borne sonar are inefficient and have limited coverage. In recent years, airborne lidar has become a supplement to depth measurement methods, but it is expensive and cannot reach shallow sea areas with different and complex natural environments. Summary of the invention

[0003] The present invention aims to solve the problem of related technical limitations at least to a certain extent. To this end, the present invention proposes a remote sensing shallow sea water depth inversion method, device, electronic equipment and storage medium, which can accurately perform remote sensing shallow sea water depth inversion.

[0004] On the one hand, an embodiment of the present invention provides a remote sensing shallow sea water depth inversion method, comprising the following steps: Obtain satellite remote sensing images and measured water depth data of the area to be measured; Correct and preprocess satellite remote sensing images to obtain spectral data; The measured water depth data is divided into a training set and a test set based on a preset ratio; A partial least squares regression model is constructed based on the training set and the spectral data of each corresponding band; The spectral data were predicted using the partial least squares regression model to obtain the prediction results; Verify the predictions based on the measured values ​​of the training set and test set and their corresponding prediction results; When the prediction verification result does not meet the preset conditions, the measured water depth data is randomly re-divided based on the preset ratio to obtain the training set and the test set; Return to execute the step of constructing a partial least squares regression model based on the training set and the spectral data of each band corresponding to it, until the result of the prediction verification meets the preset conditions; The support vector machine model was constructed based on the training set and the test set using the Parrot optimization algorithm; The support vector machine model is used to invert the water depth of the target area.

[0005] Optionally, before the step of randomly re-dividing the measured water depth data based on a preset ratio to obtain a training set and a test set, the method further includes the following steps: Update and adjust the preset ratio.

[0006] Optionally, the satellite remote sensing image is corrected and preprocessed to obtain spectral data, including the following steps: The satellite remote sensing images are preprocessed by radiometric calibration, atmospheric correction and geometric correction in sequence to obtain water surface reflectivity data; The normalized water index is obtained by processing the green band and near-infrared band in the water surface reflectance data, and the area to be tested is divided into a water area and a non-water area based on the normalized water index; The water surface reflectivity data of the water area is processed with solar flare correction to obtain the off-water reflectivity data.

[0007] Optionally, before the step of dividing the measured water depth data based on a preset ratio to obtain a training set and a test set, the method further includes the following steps: Tidal correction is performed on the measured water depth data based on the pre-acquired tidal observation data values.

[0008] Optionally, constructing a partial least squares regression model based on the training set and the spectral data of each band corresponding to the training set includes the following steps: Extract the spectral values ​​of each band of the remote sensing image pixel corresponding to each water depth measurement point in the training set from the spectral data; The spectral values ​​of each band and the measured values ​​of the corresponding water depth measurement points are used as input and output in turn to construct a partial least squares regression model; Before constructing the partial least squares regression model, the spectral values ​​of each band and the measured values ​​of the corresponding water depth measurement points are normalized in advance.

[0009] Optionally, prediction verification is performed based on the measured values ​​of the training set and the test set and their corresponding prediction results, including the following steps: The first root mean square error is obtained by processing the measured values ​​of the training set and the corresponding prediction results; According to a first average value of all measured values ​​of the training set, the measured values ​​of the training set and their corresponding prediction results are processed to obtain a first determination coefficient; The second root mean square error is obtained based on the measured values ​​of the test set and the corresponding prediction results; The second determination coefficient is obtained according to the second average value of all the measured values ​​of the test set, combined with the measured values ​​of the test set and the corresponding prediction results.

[0010] Optionally, constructing a support vector machine model based on the training set and the test set by using the Parrot optimization algorithm includes the following steps: The initial support vector machine model is obtained by training with the training set; An initial population is randomly generated and used as the target population; the initial population includes a preset number of individuals, each of which includes a penalty coefficient and a kernel parameter; Apply the parameters of the individuals in the target population to the initial support vector machine model, use the preset objective function to evaluate the fitness of the initial support vector machine model applied to each individual through the test set, and obtain the optimal individual; The target population is updated by imitating the optimal individual through other individuals in the target population and then adjusting the target population in combination with the preset optimization strategy; Return the step of applying the parameters of the individuals of the target population to the initial support vector machine model until the maximum number of iterations is reached, and output the optimal individual of the last iteration; The support vector machine model is established using the penalty coefficient and kernel parameter in the optimal individual.

[0011] On the other hand, an embodiment of the present invention provides a remote sensing shallow sea water depth inversion device, comprising: The first module is used to obtain satellite remote sensing images and measured water depth data of the area to be measured; The second module is used to perform correction and preprocessing on satellite remote sensing images to obtain spectral data; The third module is used to divide the measured water depth data into a training set and a test set based on a preset ratio; The fourth module is used to construct a partial least squares regression model based on the training set and the spectral data of each corresponding band; The fifth module is used to predict the spectral data using the partial least squares regression model to obtain the prediction results; The sixth module is used to verify the prediction based on the measured values ​​of the training set and the test set and their corresponding prediction results; The seventh module is used to randomly re-divide the measured water depth data into training sets and test sets based on a preset ratio when the prediction verification result does not meet the preset conditions; The eighth module is used to return to execute the fourth module until the result of the prediction verification meets the preset conditions; The ninth module is used to construct a support vector machine model based on the training set and the test set through the Parrot optimization algorithm; The tenth module is used to perform water depth inversion in the target area using a support vector machine model.

[0012] Optionally, the device further comprises: The eleventh module is used to update and adjust the preset ratio.

[0013] Optionally, the device further comprises: The twelfth module is used to perform tidal correction on the measured water depth data based on the pre-acquired tidal observation data values.

[0014] On the other hand, an embodiment of the present invention provides an electronic device, including: a processor and a memory; the memory is used to store programs; the processor executes the program to implement the above-mentioned remote sensing shallow sea water depth inversion method.

[0015] On the other hand, an embodiment of the present invention provides a computer storage medium, in which a program executable by a processor is stored. When the program executable by the processor is executed by the processor, it is used to implement the above-mentioned remote sensing shallow sea water depth inversion method.

[0016] The embodiment of the present invention obtains satellite remote sensing images and measured water depth data of the area to be measured; performs correction preprocessing on the satellite remote sensing images to obtain spectral data; divides the measured water depth data based on a preset ratio to obtain a training set and a test set; constructs a partial least squares regression model based on the training set and the spectral data of each corresponding band; predicts the spectral data using the partial least squares regression model to obtain a prediction result; performs prediction verification based on the measured values ​​of the training set and the test set and the corresponding prediction results; when the result of the prediction verification does not meet the preset conditions, re-randomly divides the measured water depth data based on the preset ratio to obtain a training set and a test set; returns to execute the step of constructing a partial least squares regression model based on the training set and the spectral data of each corresponding band until the result of the prediction verification meets the preset conditions; constructs a support vector machine model based on the training set and the test set through the Parrot optimization algorithm; and uses the support vector machine model to perform water depth inversion of the target area. The present invention is based on satellite remote sensing data and uses a machine learning model to perform water depth inversion, which can significantly improve the accuracy of water depth results. In addition, the present invention can avoid overfitting or underfitting caused by random division of training sets and test sets by continuously iterating and verifying the prediction results of each division, thereby improving the accuracy of water depth inversion. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings are used to provide a further understanding of the technical solution of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the technical solution of the present invention and do not constitute a limitation on the technical solution of the present invention.

[0018] Figure 1 This is a schematic diagram of an implementation environment for remote sensing shallow sea water depth inversion provided by an embodiment of the present invention; Figure 2 It is a flow chart of a remote sensing shallow sea water depth inversion method provided by an embodiment of the present invention; Figure 3 A schematic diagram of the expanded flow of step S200 provided in an embodiment of the present invention; Figure 4 A schematic diagram of an example of a seafloor time structure provided by an embodiment of the present invention; Figure 5 A schematic diagram of the expanded flow of step S400 provided in an embodiment of the present invention; Figure 6 A schematic diagram of the expanded flow of step S600 provided in an embodiment of the present invention; Figure 7 A schematic diagram of the expansion process of constructing a support vector machine model using the Parrot algorithm provided in an embodiment of the present invention; Figure 8 A schematic diagram of the overall process of a specific application of the remote sensing shallow water depth inversion method provided by an embodiment of the present invention; Fig. 9 A schematic diagram of the structure of a remote sensing shallow sea water depth inversion device provided by an embodiment of the present invention; Fig.10 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0020] It should be noted that, although the functional modules are divided in the system schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the system or the order in the flowchart. The terms "first / S100", "second / S200", etc. in the specification, claims and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0021] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0022] It is understandable that the remote sensing shallow sea water depth inversion method provided in the embodiment of the present invention can be applied to any computer device with data processing and computing capabilities, and this computer device can be various types of terminals or servers. When the computer device in the embodiment is a server, the server is an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network, content distribution network), and big data and artificial intelligence platforms. Optionally, the terminal is a smart phone, a tablet computer, a laptop computer, a desktop computer, etc., but is not limited to this.

[0023] To facilitate the understanding of the technical solution of the present invention, the technical feature terms that may appear in the embodiments of the present invention are first explained: like Figure 1 FIG. 1 is a schematic diagram of an implementation environment provided by an embodiment of the present invention. Figure 1 The implementation environment includes at least one terminal 102 and a server 101. The terminal 102 and the server 101 can be connected to a network wirelessly or wired to complete data transmission and exchange.

[0024] Server 101 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers. It can also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), as well as big data and artificial intelligence platforms.

[0025] In addition, the server 101 can also be a node server in the blockchain network. Blockchain is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, and encryption algorithm.

[0026] The terminal 102 may be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, etc., but is not limited thereto. The terminal 102 and the server 101 may be directly or indirectly connected via wired or wireless communication, which is not limited in the embodiment of the present invention.

[0027] Based on the example Figure 1In the implementation environment shown, an embodiment of the present invention provides a remote sensing shallow sea water depth inversion method. The following is explained using the remote sensing shallow sea water depth inversion method applied to the server 101 as an example. It can be understood that the remote sensing shallow sea water depth inversion method can also be applied to the terminal 102.

[0028] Reference Figure 2 , Figure 2 The flowchart of the remote sensing shallow sea water depth inversion method applied to the server provided in the embodiment of the present invention, the execution subject of the remote sensing shallow sea water depth inversion method can be any of the aforementioned computer devices (including servers or terminals). Figure 2 , the method comprises the following steps: S100, obtaining satellite remote sensing images and measured water depth data of the area to be measured; S200, performing correction preprocessing on the satellite remote sensing image to obtain spectral data; It should be noted that in some embodiments, Figure 3 As shown, the satellite remote sensing image is corrected and preprocessed to obtain spectral data, which can include the following steps: S201, the satellite remote sensing image is sequentially preprocessed with radiation calibration, atmospheric correction and geometric correction to obtain water surface reflectance data; S202, the normalized water index is obtained according to the green band and near infrared band in the water surface reflectance data, and the area to be measured is divided into water area and non-water area based on the normalized water index; S203, the water surface reflectance data of the water area is corrected for solar flare to obtain off-water reflectance data.

[0029] For example, in some specific implementations, a multispectral or hyperspectral satellite remote sensing image of the area to be measured is obtained, and the image is sequentially preprocessed by radiometric calibration, atmospheric correction, and geometric correction to obtain water surface reflectance data ( R rs ).

[0030] The normalized water index NDWI was calculated. NDWI>0 was the water area, and NDWI<0 was the non-water area. The water area was extracted and retained, and the non-water area was removed.

[0031]

[0032] Where, NDWI is the normalized water index; R rs (green) is the water surface reflectivity in the green band; R rs (nir) is the water surface reflectivity in the near-infrared band.

[0033] The water surface reflectivity data of the water area is processed by solar flare correction to obtain the off-water reflectivity data.

[0034] In the formula, R w is the reflectivity off water; R rs is the water surface reflectivity; b i For deep water areas, the near-infrared band water surface reflectance of each pixel is the x-axis, and the corresponding visible light band water surface reflectance is the y-axis. The slope of the linear fitting of the scatter plot formed by the two (such as Figure 4 ), where deep water areas usually appear dark blue or black on remote sensing images, and can be further determined by setting the reflectance thresholds of the blue band and the near-infrared band. Usually, in areas with good water quality, the R b <0.04 and R nir <0.004, the threshold can be adjusted according to the actual situation of the image, and auxiliary judgment can be made by drawing a scatter plot of the water surface reflectivity in the blue band and near-infrared band; R b is the water surface reflectivity in the blue band; R nir is the water surface reflectivity in the near-infrared band; R min-nir It is the minimum value of water surface reflectance in the near-infrared band in the scatter plot.

[0035] S300, dividing the measured water depth data based on a preset ratio to obtain a training set and a test set; For example, in some specific implementations, the measured water depth data of the area to be measured can be randomly divided into a training set and a test set at a ratio of 85% and 15%. The ratio can be adjusted according to demand, and in the actual iteration process, the ratio is continuously adjusted with the prediction verification of each iteration.

[0036] It should be noted that, in some embodiments, before the step of dividing the measured water depth data based on a preset ratio to obtain a training set and a test set, the method may also include the following steps: performing tidal correction on the measured water depth data based on pre-acquired tidal observation data values.

[0037] For example, in some specific implementations, the measured water depth data of the measured area is obtained, and the measured water depth data is tidal corrected to ensure that the tidal height of the measured water depth data is consistent with the tidal height at the time of image capture. The measured water depth after tidal correction:

[0038] Where h is the measured water depth at the time of image capture, h met is the measured water depth value, T is the tidal height when obtaining the measured water depth data, h data The tidal height at the time the image was taken. The tidal height can be obtained from the tidal monitoring website, which provides the tidal observation value closest to the area to be measured.

[0039] S400, constructing a partial least squares regression model according to the training set and the spectral data of each corresponding band; It should be noted that in some embodiments, Figure 5 As shown, constructing a partial least squares regression model based on the spectral data of the training set and its corresponding bands can include the following steps: S401, extracting the spectral values ​​of each band of the remote sensing image pixel corresponding to each water depth measured point in the training set from the spectral data; S402, using the spectral values ​​of each band and the measured values ​​of the corresponding water depth measured points as input and output in sequence to construct a partial least squares regression model; wherein, before the step of constructing the partial least squares regression model, the spectral values ​​of each band and the measured values ​​of the corresponding water depth measured points are normalized in advance.

[0040] For example, in some specific implementations, first, the spectral values ​​of each band of the remote sensing image pixel corresponding to the measured water depth point in the training set are extracted. Then, the spectral values ​​of each band are used as input, and the corresponding measured water depth values ​​are used as output. After the data is normalized, a partial least squares regression model is constructed.

[0041] S500, predicting the spectral data using a partial least squares regression model to obtain a prediction result; Exemplarily, in some specific implementations, the constructed model is used to predict the training set and test set data respectively, and the prediction results are denormalized to obtain the original data range (ie, the prediction results).

[0042] S600, performing prediction verification based on the measured values ​​of the training set and the test set and their corresponding prediction results; It should be noted that in some embodiments, Figure 6 As shown, prediction verification based on the measured values ​​of the training set and the test set and their corresponding prediction results may include the following steps: S601, obtaining a first root mean square error based on the measured values ​​of the training set and their corresponding prediction results; S602, obtaining a first determination coefficient based on a first average value of all measured values ​​of the training set, combined with the measured values ​​of the training set and their corresponding prediction results; S603, obtaining a second root mean square error based on the measured values ​​of the test set and their corresponding prediction results; S604, obtaining a second determination coefficient based on a second average value of all measured values ​​of the test set, combined with the measured values ​​of the test set and their corresponding prediction results.

[0043] For example, in some specific implementations, the root mean square error (RMSE1) and the coefficient of determination (R1 2 ), calculate the root mean square error (RMSE2) and determination coefficient (R2) of the model prediction test set and the measured water depth data value 2 ).in:

[0044] Where RMSE is the root mean square error, n is the number of water depth points in the training set or test set, is the measured water depth value at the i-th point, Calculate the model value for the i-th point, i=1....n.

[0045]

[0046] In the formula, is the coefficient of determination, is the mean of the measured water depth values ​​in the training set or test set.

[0047] S700, when the result of the prediction verification does not meet the preset conditions, the measured water depth data is randomly re-divided based on the preset ratio to obtain a training set and a test set; Among them, in some embodiments, before the step of randomly re-dividing the measured water depth data based on the preset ratio to obtain the training set and the test set, the method may also include the following steps: updating and adjusting the preset ratio.

[0048] For example, in some specific implementations, if the error between RMSE1 and RMSE2 is less than or equal to 0.2, and R2 2 With R2 2 If the error is less than or equal to 0.2, keep the existing ratio and division scheme of the training set and test set, otherwise randomly divide the measured water depth data test set and training set, and / or, continue the original steps by redistributing the ratio of the measured water depth data training set and test set until the error is within 0.2. The error can be adjusted according to demand, but it should not be too large, otherwise the subsequent model will be overfitting or underfitting.

[0049] S800, returning to execute the step of constructing a partial least squares regression model according to the training set and the spectral data of each band corresponding to it, until the result of the prediction verification meets the preset conditions; S900, based on the training set and the test set, the support vector machine model was constructed by the Parrot optimization algorithm; It should be noted that in some embodiments, Figure 7As shown, the support vector machine model is constructed by the Parrot optimization algorithm based on the training set and the test set, which may include the following steps: S901, using the training set to train to obtain the initial support vector machine model; S902, randomly generating an initial population, and using the initial population as the target population; the initial population includes a preset number of individuals, each of which includes a penalty coefficient and a kernel parameter; S903, applying the parameters of the individuals of the target population to the initial support vector machine model, and using the preset objective function to evaluate the fitness of the initial support vector machine model applied to each individual through the test set to obtain the optimal individual; S904, imitating the optimal individual through other individuals in the target population, and then adjusting and updating the target population in combination with the preset optimization strategy; S905, returning to the step of applying the parameters of the individuals of the target population to the initial support vector machine model, until the maximum number of iterations is reached, and the optimal individual of the last round of iteration is output; S906, using the penalty coefficient and kernel parameter in the optimal individual to establish the support vector machine model.

[0050] For example, in some specific implementations, if the root mean square error and determination coefficient of the training set and the test set are slightly different, the training set and the test set data are normalized. Based on the data characteristics, the Parrot optimization algorithm is used to obtain the optimal values ​​of the penalty coefficient and the kernel parameter in the support vector machine model.

[0051] Among them, the Parrot Optimization (PO) algorithm is a population-based stochastic optimization algorithm inspired by the group behavior of parrots. It simulates the behavior of parrots searching for food and learning the environment in nature, and pays special attention to the social learning and memory ability of parrots. This algorithm is usually applied to the global optimization of high-dimensional complex problems. Through the Parrot Optimization Algorithm, the optimal parameters of the support vector machine model can be efficiently found to improve the prediction performance of the model. At the same time, it has global search capabilities and fast convergence characteristics, which is suitable for dealing with nonlinear complex optimization problems.

[0052] Specifically, the optimization steps can be implemented as follows: 1. Determine the objective function to evaluate the performance of a given parameter combination. The objective function can be defined using the following steps: (1) Use the training set to train the support vector machine model; (2) Use the test set to evaluate model performance, such as by mean square error.

[0053] In the formula, is the measured water depth value in the test set, is the calculated value of the support vector machine model, N is the number of measured water depth points in the test set, and i=1....N.

[0054] (3) The performance index is used as the fitness value, and the optimization goal is to minimize the error.

[0055] 2. Initialize the population: (1) Randomly generate an initial population, where each individual represents a penalty coefficient (C) and a kernel parameter (g); (2) Set the search range, for example, the lower limit of the penalty coefficient (C) and the kernel parameter (g) is 0.001 and the upper limit is 100; (3) Record the population size (e.g., 30 individuals) and the maximum number of iterations (e.g., 100).

[0056] 3. Fitness evaluation: (1) For each individual, call the objective function and calculate the fitness value.

[0057] (2) Sort by fitness value and determine the current optimal individual.

[0058] 4. Imitation and update: (1) Let the suboptimal individuals in the population imitate the behavior of the current optimal individual to narrow the search range; (2) Adjust the position of individuals according to the optimization strategy, including random perturbations and imitation guidance.

[0059] Update formula: ;in, is the position of the i-th individual at the t-th iteration; is the position of the current optimal individual; is a random factor; is a noise term to prevent falling into local optimum.

[0060] 5. Termination: When the maximum number of iterations is reached or the change in fitness value is less than the threshold, the current optimal penalty coefficient (C) and kernel parameter (g) are output.

[0061] The optimized penalty coefficient and kernel parameters are used to establish a support vector machine model, and the model is used to perform water depth inversion on the test area. In some specific application scenarios, the mean relative error (MRE) and root mean square error (RMSE) can also be used to evaluate the accuracy of the results. The smaller the MRE and RMSE values, the higher the accuracy.

[0062]

[0063] Where n is the number of measured water depth points in the test set; is the measured water depth value of the i-th point in the test set; Compute values ​​for the model; i=1...n.

[0064] S1000, using the support vector machine model to perform water depth inversion in the target area.

[0065] For example, in some specific implementations, water depth inversion can be achieved based on remote sensing images of the target area through the optimized support vector machine model.

[0066] In order to explain the principle of the technical solution of the present invention in detail, the overall process of the present invention is described below in combination with some specific embodiments. It is easy to understand that the following is an explanation of the technical principle of the present invention and cannot be regarded as a limitation of the present invention.

[0067] First of all, it should be noted that with the emergence of machine learning methods, support vector machines, random forests and other machine learning methods have been applied to water depth inversion. Compared with traditional empirical models and semi-analytical models, the accuracy of water depth results obtained by machine learning methods has been significantly improved.

[0068] However, in the related technologies, water depth inversion is performed using machine learning models based on satellite remote sensing data. The main purpose is to build a water depth inversion model between the spectral characteristics of satellite remote sensing image pixels and the characteristic relationship between the corresponding water depth measured data, and then obtain a wider range of water depth values ​​based on remote sensing images. In this process, the water depth measured data needs to be divided into training sets and test sets, and the training set is used to build the model, while the test set is used to evaluate the accuracy of the water depth inversion results. Currently, the training set and the test set are divided by random methods, but the random division method often leads to overfitting and underfitting, which ultimately reduces the accuracy of water depth inversion. Overfitting refers to the phenomenon that the machine learning model performs well on the training data but performs poorly on the test data. This means that the model has learned too many specific details on the training data set, so that it cannot generalize on new and unseen data. Underfitting refers to the phenomenon that the machine learning model performs poorly on both the training data and the test data. This means that the model does not have enough learning ability to capture the key features and patterns in the data.

[0069] In view of the above problems, the present invention provides a shallow sea water depth inversion method based on satellite remote sensing images, which can solve the overfitting or underfitting phenomenon caused by random division of training sets and test sets, and improve the water depth inversion accuracy. Figure 8 As shown, the present invention can be implemented as follows: Obtain multispectral or hyperspectral satellite remote sensing images of the area to be measured, perform radiometric calibration, atmospheric correction and geometric correction preprocessing on the images in turn, and obtain water surface reflectance data ( R rs ).

[0070] The normalized water index NDWI was calculated. NDWI>0 was the water area, and NDWI<0 was the non-water area. The water area was extracted and retained, and the non-water area was removed.

[0071]

[0072] Where, NDWI is the normalized water index; R rs (green) is the water surface reflectivity in the green band; R rs (nir) is the water surface reflectivity in the near-infrared band.

[0073] The water surface reflectivity data of the water area is processed by solar flare correction to obtain the off-water reflectivity data.

[0074] In the formula, R w is the reflectivity off water; R rs is the water surface reflectivity; b i For deep water areas, the near-infrared band water surface reflectance of each pixel is the x-axis, and the corresponding visible light band water surface reflectance is the y-axis. The slope of the linear fitting of the scatter plot formed by the two (such as Figure 4 ), where deep water areas usually appear dark blue or black on remote sensing images, and can be further determined by setting the reflectance thresholds of the blue band and the near-infrared band. Usually, in areas with good water quality, the R b <0.04 and R nir <0.004, the threshold can be adjusted according to the actual situation of the image, and auxiliary judgment can be made by drawing a scatter plot of the water surface reflectivity in the blue band and near-infrared band; R b is the water surface reflectivity in the blue band; R nir is the water surface reflectivity in the near-infrared band; R min-nir It is the minimum value of water surface reflectance in the near-infrared band in the scatter plot.

[0075] Obtain the measured water depth data of the area to be measured, and perform tidal correction on the measured water depth data to ensure that the tidal height of the measured water depth data is consistent with the tidal height at the time of image shooting. Measured water depth after tidal correction:

[0076] Where h is the measured water depth at the time of image capture, h met is the measured water depth value, T is the tidal height when obtaining the measured water depth data, h data The tidal height at the time the image was taken. The tidal height can be obtained from the tidal monitoring website, which provides the tidal observation value closest to the area to be measured.

[0077] The measured water depth data of the area to be tested are randomly divided into training set and test set at a ratio of 85% and 15%. The ratio can be adjusted according to demand, and in the actual iteration process, the ratio is continuously adjusted with the prediction verification of each iteration.

[0078] Extract the spectral values ​​of each band of the remote sensing image pixels corresponding to the measured water depth points in the training set.

[0079] The spectral values ​​of each band are taken as input and the corresponding measured water depth values ​​are taken as output. After the data are normalized, a partial least squares regression model is constructed.

[0080] Calculate the root mean square error (RMSE1) and determination coefficient (R1) between the model prediction training set results and the measured water depth data values. 2 ), calculate the root mean square error (RMSE2) and determination coefficient (R2) of the model prediction test set and the measured water depth data value 2 ).in:

[0081] Where RMSE is the root mean square error, n is the number of water depth points in the training set or test set, is the measured water depth value at the i-th point, Calculate the model value for the i-th point, i=1....n.

[0082]

[0083] In the formula, is the coefficient of determination, is the mean of the measured water depth values ​​in the training set or test set.

[0084] If the error between RMSE1 and RMSE2 is less than or equal to 0.2, and R1 2 With R2 2 If the error is less than or equal to 0.2, the existing ratio and division scheme of the training set and the test set are retained. Otherwise, the ratio of the measured water depth data training set and the test set is redistributed, or the measured water depth data test set and the training set are randomly divided again, and the original steps are continued until the error is within the range of 0.2. The error can be adjusted according to the needs, but it should not be too large, otherwise the subsequent model will be overfitting or underfitting.

[0085] If the root mean square error and determination coefficient of the training set and the test set are slightly different, the training set and test set data are normalized.

[0086] Based on the data characteristics, the Parrot optimization algorithm is used to obtain the optimal values ​​of the penalty coefficient and kernel parameters in the support vector machine model.

[0087] Among them, the Parrot Optimization (PO) algorithm is a population-based stochastic optimization algorithm inspired by the group behavior of parrots. It simulates the behavior of parrots searching for food and learning the environment in nature, and pays special attention to the social learning and memory ability of parrots. This algorithm is usually applied to the global optimization of high-dimensional complex problems. Through the Parrot Optimization Algorithm, the optimal parameters of the support vector machine model can be efficiently found to improve the prediction performance of the model. At the same time, it has global search capabilities and fast convergence characteristics, which is suitable for dealing with nonlinear complex optimization problems.

[0088] Specifically, the optimization steps can be implemented as follows: 1. Determine the objective function to evaluate the performance of a given parameter combination. The objective function can be defined using the following steps: (1) Use the training set to train the support vector machine model; (2) Use the test set to evaluate model performance, such as by mean square error.

[0089] In the formula, is the measured water depth value in the test set, is the calculated value of the support vector machine model, N is the number of measured water depth points in the test set, and i=1....N.

[0090] (3) The performance index is used as the fitness value, and the optimization goal is to minimize the error.

[0091] 2. Initialize the population: (1) Randomly generate an initial population, where each individual represents a penalty coefficient (C) and a kernel parameter (g); (2) Set the search range, for example, the lower limit of the penalty coefficient (C) and the kernel parameter (g) is 0.001 and the upper limit is 100; (3) Record the population size (e.g., 30 individuals) and the maximum number of iterations (e.g., 100).

[0092] 3. Fitness evaluation: (1) For each individual, call the objective function and calculate the fitness value.

[0093] (3) Sort by fitness value and determine the current optimal individual.

[0094] 4. Imitation and update: (1) Let the suboptimal individuals in the population imitate the behavior of the current optimal individual to narrow the search range; (2) Adjust the position of individuals according to the optimization strategy, including random perturbations and imitation guidance.

[0095] Update formula: ;in, is the position of the i-th individual at the t-th iteration; is the position of the current optimal individual; is a random factor; is a noise term to prevent falling into local optimum.

[0096] 5. Termination: When the maximum number of iterations is reached or the change in fitness value is less than the threshold, the current optimal penalty coefficient (C) and kernel parameter (g) are output.

[0097] The support vector machine model was established using the optimized penalty coefficient and kernel parameters, and the model was used to invert the water depth of the test area. Finally, the map was drawn and the mean relative error (MRE) and root mean square error (RMSE) were used to evaluate the accuracy of the results. The smaller the MRE and RMSE values, the higher the accuracy.

[0098]

[0099] Where n is the number of measured water depth points in the test set; is the measured water depth value of the i-th point in the test set; Compute values ​​for the model; i=1...n.

[0100] In summary, the present invention solves the overfitting and underfitting problems that may be caused by random partitioning by applying an adaptive training set and test set partitioning strategy. In the process of remote sensing inversion of water depth, it performs iterative cycles according to whether the error (RMSE) and determination coefficient (R²) of the training set and the test set are within the threshold range. By iteratively optimizing the data partitioning scheme, the error is ensured to be within the set threshold range, thereby improving the generalization ability and robustness of the model. In addition, the present invention optimizes the hyperparameters of the support vector machine based on the Parrot optimization algorithm. Specifically, the present invention applies the Parrot optimization algorithm to shallow water depth inversion for the first time. Through a mechanism combining global search with local development, the penalty coefficient and kernel parameters of the support vector machine (SVM) are dynamically optimized, thereby improving the model's adaptability to nonlinear characteristics. The Parrot optimization algorithm designs an adaptive convergence mechanism, which combines the specific data distribution characteristics of shallow water depth inversion to further enhance the algorithm efficiency and the accuracy of the solution.

[0101] Compared with the prior art, the present invention has at least the following beneficial effects: Effectively solve the overfitting and underfitting problems caused by improper division of training and test sets in the process of water depth inversion. Aiming at the shortcomings of the existing methods that random division of training and test sets easily leads to overfitting or underfitting of the model, the present invention ensures the generalization ability and robustness of the model by dynamically adjusting the division ratio of training and test sets and the error evaluation mechanism, and significantly improves the overall accuracy of water depth inversion.

[0102] Combined with intelligent optimization algorithm to improve the performance of water depth inversion model. This paper introduces the Parrot optimization algorithm into the shallow water depth inversion field for the first time, optimizing the penalty coefficient and kernel parameters of the support vector machine. Through the global search and local development capabilities of the Parrot algorithm, the model parameters are dynamically adjusted to make it more adaptable to the nonlinear characteristics of the complex shallow sea environment, and improve the stability and accuracy of the inversion results.

[0103] On the other hand, Fig. 9 As shown, an embodiment of the present invention provides a remote sensing shallow sea water depth inversion device 900, which may include: The first module 901 is used to obtain satellite remote sensing images and measured water depth data of the area to be measured; The second module 902 is used to perform correction preprocessing on the satellite remote sensing image to obtain spectral data; The third module 903 is used to divide the measured water depth data into a training set and a test set based on a preset ratio; The fourth module 904 is used to construct a partial least squares regression model based on the training set and the spectral data of each band corresponding to it; The fifth module 905 is used to predict the spectral data using a partial least squares regression model to obtain a prediction result; The sixth module 906 is used to perform prediction verification based on the measured values ​​of the training set and the test set and their corresponding prediction results; The seventh module 907 is used for randomly re-dividing the measured water depth data into a training set and a test set based on a preset ratio when the prediction verification result does not meet the preset conditions; The eighth module 908 is used to return to execute the fourth module until the result of the prediction verification meets the preset condition; A ninth module 909 is used to construct a support vector machine model based on the training set and the test set by using the Parrot optimization algorithm; The tenth module 910 is used to perform water depth inversion of the target area using a support vector machine model.

[0104] In some embodiments, the apparatus may further include: The eleventh module is used to update and adjust the preset ratio.

[0105] In some embodiments, the apparatus may further include: The twelfth module is used to perform tidal correction on the measured water depth data based on the pre-acquired tidal observation data values.

[0106] The contents of the method embodiments of the present invention are all applicable to the device embodiments. The functions specifically implemented by the device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0107] On the other hand, an embodiment of the present invention further provides an electronic device, the electronic device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the above remote sensing shallow sea depth inversion method when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a car computer, etc.

[0108] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0109] like Fig.10 As shown, Fig.10 The hardware structure of an electronic device 1000 of another embodiment is illustrated. The electronic device 1000 includes: The processor 1001 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (application-specific integrated circuit, aSIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present invention; The memory 1002 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 1002 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 1002, and the processor 1001 calls and executes the network node population optimization method of the embodiment of the present invention; Input / output interface 1003, used to implement information input and output; Communication interface 1004, used to realize communication interaction between the device and other devices, which can be realized through wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WIFI, Bluetooth, etc.); A bus 1005 , which transmits information between various components of the device (e.g., the processor 1001 , the memory 1002 , the input / output interface 1003 , and the communication interface 1004 ); The processor 1001 , the memory 1002 , the input / output interface 1003 and the communication interface 1004 are connected to each other in communication within the device via the bus 1005 .

[0110] The electronic device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0111] The contents of the method embodiments of the present invention are all applicable to the electronic device embodiments. The functions specifically implemented by the electronic device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0112] Another aspect of an embodiment of the present invention further provides a computer-readable storage medium, wherein the storage medium stores a program, and the program is executed by a processor to implement the above method.

[0113] It should be noted that the computer-readable medium shown in the embodiment of the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (Compact Disc Read to Only Memory, CD to ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer readable signal media may also be any computer readable medium other than computer readable storage media, which may send, propagate, or transmit programs for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0114] The contents of the method embodiments of the present invention are all applicable to the computer-readable storage medium embodiments. The functions specifically implemented by the computer-readable storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0115] The embodiment of the present invention also discloses a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. A processor of a computer device can read the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the above method.

[0116] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0117] It should be noted that, although several modules of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to an embodiment of the present invention, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into being embodied by multiple modules or units.

[0118] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD to ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the implementation of the present invention.

[0119] In some selectable embodiments, the function / operation mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the function / operation involved, the two boxes shown in succession can actually be executed substantially simultaneously or the boxes can sometimes be executed in reverse order. In addition, the embodiment presented and described in the flow chart of the present invention is provided by way of example, for the purpose of providing a more comprehensive understanding of technology. The disclosed method is not limited to the operation and logic flow presented herein. Selectable embodiments are expected, wherein the order of various operations is changed and the sub-operation of a part for which is described as a larger operation is performed independently.

[0120] In addition, although the present invention is described in the context of functional modules, it should be understood that, unless otherwise specified, one or more of the functions and / or features can be integrated into a single physical device and / or software module, or one or more functions and / or features can be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the present invention. More specifically, in view of the properties, functions and internal relationships of the various functional modules in the device disclosed herein, the actual implementation of the module will be understood within the conventional skills of the engineer. Therefore, those skilled in the art can implement the present invention set forth in the claims without excessive experimentation using ordinary techniques. It is also understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present invention, which is determined by the full scope of the appended claims and their equivalents.

[0121] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0122] The logic and / or steps represented in the flowchart or otherwise described herein, for example, may be considered as an ordered list of executable instructions for implementing logical functions, and may be embodied in any computer-readable medium for use by an instruction execution device, apparatus or device (such as a computer-based device, a device including a processor, or other device that can fetch instructions from an instruction execution device, apparatus or device and execute the instructions), or in conjunction with such instruction execution device, apparatus or device. For purposes of this specification, "computer-readable medium" may be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution device, apparatus or device, or in conjunction with such instruction execution device, apparatus or device.

[0123] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or processing in another suitable manner as necessary, and then stored in a computer memory.

[0124] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution device. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0125] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0126] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

[0127] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of the present invention.

Claims

1. A remote sensing shallow sea water depth inversion method, characterized in that: The following steps are involved: Obtain satellite remote sensing images and measured water depth data of the area to be measured; Performing correction preprocessing on the satellite remote sensing image to obtain spectral data; Dividing the measured water depth data based on a preset ratio to obtain a training set and a test set; Constructing a partial least squares regression model based on the spectral data of the training set and its corresponding bands; Using the partial least squares regression model to predict the spectral data, to obtain a prediction result; Performing prediction verification based on the measured values ​​of the training set and the test set and the corresponding prediction results; When the result of the prediction verification does not meet the preset condition, re-randomly dividing the measured water depth data based on the preset ratio to obtain the training set and the test set; Returning to the step of constructing a partial least squares regression model based on the spectral data of the training set and its corresponding bands, until the result of the prediction verification meets the preset condition; A support vector machine model is constructed based on the training set and the test set by using the Parrot optimization algorithm; The support vector machine model is used to perform water depth inversion in the target area.

2. The remote sensing shallow sea water depth inversion method according to claim 1 is characterized in that: Before the step of randomly re-dividing the measured water depth data based on the preset ratio to obtain the training set and the test set, the method further includes the following steps: The preset ratio is updated and adjusted.

3. The remote sensing shallow sea water depth inversion method according to claim 1 is characterized in that: The process of performing correction preprocessing on the satellite remote sensing image to obtain spectral data comprises the following steps: The satellite remote sensing image is sequentially subjected to radiometric calibration, atmospheric correction and geometric correction preprocessing to obtain water surface reflectivity data; Obtaining a normalized water index according to the green band and the near-infrared band in the water surface reflectance data, and dividing the area to be tested into a water area and a non-water area based on the normalized water index; The water surface reflectivity data of the water body area is subjected to solar flare correction processing to obtain off-water reflectivity data.

4. The remote sensing shallow sea water depth inversion method according to claim 1 is characterized in that: Before the step of dividing the measured water depth data based on a preset ratio to obtain a training set and a test set, the method further includes the following steps: The measured water depth data is tidal corrected based on the pre-acquired tidal observation data values.

5. The remote sensing shallow sea water depth inversion method according to claim 1 is characterized in that: The partial least squares regression model is constructed according to the spectral data of the training set and its corresponding bands, comprising the following steps: Extracting spectral values ​​of each band of the remote sensing image pixel corresponding to each water depth measurement point in the training set from the spectral data; The spectral values ​​of each band and the measured values ​​of the corresponding water depth measured points are sequentially used as input and output to construct the partial least squares regression model; Before the step of constructing the partial least squares regression model, the spectral values ​​of each band and the measured values ​​of the corresponding water depth measurement points are normalized in advance.

6. The remote sensing shallow sea water depth inversion method according to claim 1 is characterized in that: The prediction verification based on the measured values ​​of the training set and the test set and the corresponding prediction results comprises the following steps: Based on the measured value of the training set and the corresponding prediction result, a first root mean square error is obtained; According to a first average value of all the measured values ​​of the training set, the measured values ​​of the training set and their corresponding prediction results are processed to obtain a first determination coefficient; Obtaining a second root mean square error based on the measured value of the test set and the corresponding prediction result; A second determination coefficient is obtained according to a second average value of all the measured values ​​of the test set, combined with the measured values ​​of the test set and the corresponding prediction results.

7. The remote sensing shallow sea water depth inversion method according to claim 1 is characterized in that: The method of constructing a support vector machine model based on the training set and the test set by using the Parrot optimization algorithm comprises the following steps: Using the training set to train an initial support vector machine model; An initial population is randomly generated and used as a target population; the initial population includes a preset number of individuals, each of which includes a penalty coefficient and a kernel parameter; Applying the parameters of the individuals of the target population to the initial support vector machine model, and using a preset objective function to evaluate the fitness of the initial support vector machine model applied to each individual through the test set to obtain the optimal individual; The target population is adjusted and updated by imitating the optimal individual through other individuals in the target population and combining with a preset optimization strategy; Returning to the step of applying the parameters of the individual of the target population to the initial support vector machine model until a maximum number of iterations is reached, and outputting the optimal individual of the last round of iteration; The support vector machine model is established by using the penalty coefficient and the kernel parameter in the optimal individual.

8. A remote sensing shallow sea water depth inversion device, characterized in that: include: The first module is used to obtain satellite remote sensing images and measured water depth data of the area to be measured; The second module is used to perform correction preprocessing on the satellite remote sensing image to obtain spectral data; A third module is used to divide the measured water depth data into a training set and a test set based on a preset ratio; The fourth module is used to construct a partial least squares regression model based on the spectral data of the training set and its corresponding bands; The fifth module is used to predict the spectral data using the partial least squares regression model to obtain a prediction result; A sixth module is used to perform prediction verification based on the measured values ​​of the training set and the test set and the corresponding prediction results; The seventh module is used for re-randomly dividing the measured water depth data based on the preset ratio to obtain the training set and the test set when the result of the prediction verification does not meet the preset condition; An eighth module is used to return to execute the fourth module until the result of the prediction verification meets the preset condition; A ninth module is used to construct a support vector machine model based on the training set and the test set by using the Parrot optimization algorithm; The tenth module is used to perform water depth inversion of the target area using the support vector machine model.

9. An electronic device, characterized in that: including a processor and a memory; The memory is used to store programs; The processor executes the program to implement the method according to any one of claims 1 to 7.

10. A computer storage medium storing a program executable by a processor, characterized in that: The program executable by the processor is used to implement the method according to any one of claims 1 to 7 when executed by the processor.

Images