Underwater topographic data evolution analysis method, device and equipment and storage medium

Abstract

The invention discloses an underwater topographic data evolution analysis method, which comprises the following steps of: acquiring multi-modal original data of underwater topography, and preprocessing the original data; extracting features from the preprocessed data, and combining the extracted multi-modal features into a regression model in a polynomial form; using parameters of the polynomial regression model as optimization variables of a quantum annealing algorithm, defining an objective function, and minimizing the objective function through the quantum annealing algorithm; and predicting the input data based on the optimized polynomial regression model, and outputting a prediction result. The invention further discloses an underwater topographic data evolution analysis device, corresponding equipment and a storage medium. According to the underwater topography data evolution analysis method based on quantum annealing optimization provided by the embodiment of the invention, high-precision prediction of the underwater topography can be realized.

Description

Technical Field

[0001] The present application relates to the technical field of data processing and analysis, and more specifically, to a method, device, equipment and storage medium for analyzing underwater terrain data evolution. Background Art

[0002] The analysis of underwater terrain evolution plays a key role in fields such as marine engineering, geological exploration, and cross-sea infrastructure planning. Accurately predicting the evolution of underwater landforms is crucial for ensuring the stability of offshore infrastructure, optimizing marine resource development, and assessing natural disaster risks in coastal areas. The rapid development of modern ocean monitoring technology has made multimodal underwater data acquisition possible. However, extracting valuable information from this massive, heterogeneous, multimodal data and constructing high-precision prediction models remain core challenges in this field.

[0003] Currently, traditional underwater terrain prediction methods primarily rely on single-modal data, such as side-scan sonar imagery or multibeam bathymetry data. Due to limited information sources, these methods struggle to fully reflect the complex features of underwater terrain, severely limiting prediction accuracy. In recent years, multimodal data fusion technology has gained increasing attention. However, due to the high dimensionality and heterogeneity of data from different modalities, the fusion process is extremely complex, and feature extraction is extremely challenging. Furthermore, traditional regression models, such as linear regression, perform poorly when processing data with complex nonlinear relationships. While polynomial regression can theoretically fit nonlinear relationships, it can easily fall into local optimal solutions in high-dimensional multimodal data scenarios, resulting in inaccurate model parameter estimation and unstable prediction performance. Therefore, there is an urgent need for an advanced method that can effectively fuse multimodal data and simultaneously optimize model parameters to improve the accuracy and reliability of underwater terrain data evolution analysis. Summary of the Invention

[0004] In response to at least one defect or improvement need in the prior art, the present application provides a method, apparatus, device and storage medium for analyzing the evolution of underwater terrain data, which can solve at least one of the problems existing in the above-mentioned background technology.

[0005] To achieve the above objectives, according to a first aspect of the present application, a method for analyzing underwater terrain data evolution is provided, the method comprising: Collect multimodal raw data of underwater terrain and preprocess the raw data; Extract features from the preprocessed data and combine the extracted multimodal features into a polynomial regression model; Using the parameters of the polynomial regression model as optimization variables of the quantum annealing algorithm, defining an objective function, and minimizing the objective function using the quantum annealing algorithm; Predict the input data based on the optimized polynomial regression model and output the prediction results.

[0006] Furthermore, in the above-mentioned underwater terrain data evolution analysis method, the acquisition of multimodal raw data of underwater terrain includes: Acquiring underwater terrain image data by side-scan sonar, wherein the underwater terrain image data includes texture, landform, and water characteristics; Acquiring underwater terrain point cloud data using a multi-beam detection system, wherein the underwater terrain point cloud data includes depth and slope; Meteorological and hydrological structured data are acquired through meteorological and hydrological monitoring stations, wherein the meteorological and hydrological structured data include wind speed, tide, precipitation and flow velocity.

[0007] Furthermore, in the above-mentioned underwater terrain data evolution analysis method, the preprocessing of the original data specifically includes: Performing radiation correction and geometric correction on the underwater terrain image data to eliminate image noise and distortion; For the underwater terrain point cloud data, outliers and erroneous data are removed, and coordinate transformation is performed to unify the data with other data in a spatial coordinate system; For the meteorological and hydrological structured data, missing values ​​are filled and the data is normalized. Furthermore, in the above-mentioned underwater terrain data evolution analysis method, the feature extraction from the pre-processed data specifically includes: For the pre-processed underwater terrain image data, convolutional neural network image processing technology is used to extract the texture, landform and water body features in the image; Extract depth features and terrain relief features from the pre-processed underwater terrain point cloud data; For the pre-processed meteorological and hydrological structured data, variable characteristics related to water depth, tidal change characteristics, flow velocity, precipitation intensity and frequency characteristics are extracted.

[0008] Furthermore, in the above-mentioned underwater terrain data evolution analysis method, the step of combining the extracted multimodal features into a polynomial regression model specifically includes: The extracted underwater terrain image features are , the extracted underwater terrain point cloud features are , the extracted meteorological and hydrological characteristics are ; The independent variable X for constructing the polynomial regression model is

[0009] in, , , , is the coefficient to be optimized, 、 and are the number of extracted underwater terrain image features, underwater terrain point cloud features and meteorological and hydrological features respectively, and the high-order cross term is 、 and The cross product terms of .

[0010] Furthermore, the above underwater terrain data evolution analysis method defines the mean square error as the objective function, and the objective function for

[0011] in, is the sample size, is the actual observed value, is the predicted value of the polynomial regression model.

[0012] Furthermore, in the above-mentioned underwater terrain data evolution analysis method, minimizing the objective function by using the quantum annealing algorithm specifically includes: By simulating the annealing process of a quantum system from high temperature to low temperature, multiple parameter combinations are explored simultaneously based on the superposition and entanglement properties of quantum states; In each iteration, the parameters of the polynomial regression model are adjusted according to the current state of the quantum system, so that the objective function value gradually decreases, and finally the optimal parameter combination that minimizes the objective function is found.

[0013] According to a second aspect of the present application, there is also provided an underwater terrain data evolution analysis device, comprising: Data acquisition module, used to collect multi-modal raw data of underwater terrain and pre-process the raw data; The feature extraction module is used to extract features from the preprocessed data and combine the extracted multimodal features into a polynomial regression model; a parameter optimization module, configured to use the parameters of the polynomial regression model as optimization variables of a quantum annealing algorithm, define an objective function, and minimize the objective function using the quantum annealing algorithm; The prediction module is used to predict the input data based on the optimized polynomial regression model and output the prediction results.

[0014] According to the third aspect of the present application, an underwater terrain data evolution analysis device is also provided, which includes at least one processing unit and at least one storage unit, wherein the storage unit stores a computer program, and when the computer program is executed by the processing unit, the processing unit performs the steps of any one of the above methods.

[0015] According to the fourth aspect of the present application, a storage medium is also provided, which stores a computer program that can be executed by an underwater terrain data evolution analysis device. When the computer program runs on the underwater terrain data evolution analysis device, the underwater terrain data evolution analysis device executes the steps of any one of the above methods.

[0016] In general, the above technical solutions conceived by this application can achieve the following beneficial effects compared with the existing technology: The embodiment of the present application provides a method for analyzing the evolution of underwater terrain data. By fusing and preprocessing multimodal data, extracting features to construct a polynomial regression model, and optimizing model parameters using a quantum annealing algorithm, high-precision prediction of underwater terrain is achieved. The method can fully utilize the complementary information of different data sources and mine the complex nonlinear relationships between data, avoiding the defect of traditional optimization algorithms that are prone to falling into local optimal solutions. It effectively solves the problems of multimodal data fusion difficulties, inaccurate model parameter estimation, and low prediction accuracy in the existing technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 A flowchart of a method for analyzing underwater terrain data evolution provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solutions and advantages of this application more clearly understood, this application is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely for explaining this application and are not intended to limit this application. In addition, the technical features involved in the various embodiments of this application described below may be combined with each other as long as they do not conflict with each other.

[0020] The terms "first," "second," "third," and the like in the specification and claims of this application and the accompanying drawings are used to distinguish between different objects, not to describe a particular order. Furthermore, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or elements is not limited to the listed steps or elements, but may optionally include steps or elements not listed, or may optionally include other steps or elements inherent to the process, method, product, or apparatus.

[0021] Figure 1 A flow chart of a method for analyzing underwater terrain data evolution provided in an embodiment of the present application is shown as follows: Figure 1 As shown, an embodiment of the present application provides a method for analyzing underwater terrain data evolution, comprising: Collect multimodal raw data of underwater terrain and preprocess the raw data; Extract features from the preprocessed data and combine the extracted multimodal features into a polynomial regression model; Using the parameters of the polynomial regression model as optimization variables of the quantum annealing algorithm, defining an objective function, and minimizing the objective function using the quantum annealing algorithm; Predict the input data based on the optimized polynomial regression model and output the prediction results.

[0022] Specifically, side-scan sonar image data, multi-beam sonar detection data, and meteorological and hydrological data were collected from a specific sea area. Professional side-scan sonar image processing software was used to perform radiometric and geometric corrections on the images, removing noise and other interference to improve image quality. Multi-beam sonar equipment installed on a survey vessel was used to collect underwater depth data for the sea area. The sonar data was denoised to remove erroneous data caused by equipment failure or abnormal reflections, and its coordinates were converted to a unified spatial coordinate system. Wind speed, tide, precipitation, current, and other data for the sea area were obtained from nearby meteorological and hydrological monitoring stations. The meteorological and hydrological data was cleaned, missing values ​​were filled, and normalized to a value range between [0, 1].

[0023] For side-scan sonar imagery, a pre-trained convolutional neural network (CNN) model was used to extract image features, generating a set of feature vectors reflecting information such as texture, topography, and water characteristics. Features such as depth mean, standard deviation, and terrain relief were extracted from multibeam sonar data. For meteorological and hydrological data, features such as tidal amplitude, total precipitation, and average wind speed were extracted. Based on data distribution and correlation analysis, the independent variables of a polynomial regression model were constructed, including linear terms, cross terms, and higher-order cross terms.

[0024] A quantum annealing algorithm is used to optimize the parameters of a polynomial regression model. The coefficients of the polynomial are encoded as qubits, and the Hamiltonian of the quantum system is constructed. The objective function is defined as the mean squared error (MSE). By continuously adjusting the Hamiltonian of the quantum system, it is gradually cooled from a high temperature to a low temperature. During this process, the state of the quantum system continuously evolves, eventually converging to a state that minimizes the objective function. The corresponding qubit value at this point is the optimal parameter of the polynomial regression model.

[0025] The optimized polynomial regression model was applied to analyze the water depth evolution in this sea area. A portion of the data was used as a training set to train the model and optimize parameters; another portion of the data was used as a test set to evaluate the model's predictive performance. The model's prediction accuracy was evaluated by calculating metrics such as the mean squared error (MSE) and mean absolute error (MAE) on the test set.

[0026] The embodiment of the present application provides a method for analyzing the evolution of underwater terrain data. By fusing and preprocessing multimodal data, extracting features to construct a polynomial regression model, and optimizing model parameters using a quantum annealing algorithm, high-precision prediction of underwater terrain is achieved. The method can fully utilize the complementary information of different data sources and mine the complex nonlinear relationships between data, avoiding the defect of traditional optimization algorithms that are prone to falling into local optimal solutions. It effectively solves the problems of multimodal data fusion difficulties, inaccurate model parameter estimation, and low prediction accuracy in the existing technology.

[0027] Optionally, in the underwater terrain data evolution analysis method provided in an embodiment of the present application, the collecting of multimodal raw data of the underwater terrain includes: Acquiring underwater terrain image data by side-scan sonar, wherein the underwater terrain image data includes texture, landform, and water characteristics; Acquiring underwater terrain point cloud data using a multi-beam detection system, wherein the underwater terrain point cloud data includes depth and slope; Meteorological and hydrological structured data are acquired through meteorological and hydrological monitoring stations, wherein the meteorological and hydrological structured data include wind speed, tide, precipitation and flow velocity.

[0028] Specifically, underwater terrain image data is acquired through side-scan sonar. When operating, side-scan sonar emits a fan-shaped acoustic beam underwater and receives reflected signals, forming a two-dimensional image of the underwater terrain. This image data contains a rich collection of textures, landforms, and water characteristics. For example, different seabed geological structures such as rock, sand, and gravel appear in distinct texture patterns in the image; landforms are reflected in the undulations, gullies, and ridges of the terrain; and water characteristics can be reflected in the image through changes in turbidity and color.

[0029] A multibeam sounding system is used to acquire underwater topographic point cloud data. A multibeam bathymetric system can simultaneously transmit multiple acoustic beams, covering a wide underwater area. After receiving the returned acoustic signals, it generates a large number of depth measurement points, forming point cloud data. This data accurately records the depth and slope of the underwater terrain, depicting its three-dimensional form and providing high-precision spatial data support for subsequent terrain analysis.

[0030] Meteorological and hydrological structured data are obtained from meteorological and hydrological monitoring stations. These monitoring stations are typically equipped with a variety of sensors that monitor and record data such as wind speed, tides, precipitation, and current velocity in real time. Wind speed data reflects the effects of wind on the sea surface and has a significant impact on the deposition and erosion of underwater topography. Tidal data reflects the periodic rise and fall of seawater and is directly related to the short-term and long-term evolution of underwater topography. Precipitation data not only affects the salinity and density distribution of seawater but can also lead to increased sediment transport at river estuaries, thereby affecting offshore underwater topography. Current velocity data reflects the flow of seawater and plays a key role in the transport and deposition of underwater materials. These meteorological and hydrological data are stored in a structured format to facilitate integration and processing in subsequent data analysis, providing important environmental factors to supplement the evolution analysis of underwater topography data.

[0031] Optionally, in the underwater terrain data evolution analysis method provided in the embodiment of the present application, the preprocessing of the raw data specifically includes: Performing radiation correction and geometric correction on the underwater terrain image data to eliminate image noise and distortion; For the underwater terrain point cloud data, outliers and erroneous data are removed, and coordinate transformation is performed to unify the data with other data in a spatial coordinate system; For the meteorological and hydrological structured data, missing values ​​are filled and the data is normalized. Specifically, for underwater terrain image data obtained by side-scan sonar, radiation correction is first performed by adjusting the grayscale value of the image to compensate for energy attenuation and reflection differences during sonar transmission and reception. Then, geometric correction is performed to eliminate image distortion caused by factors such as equipment posture and water refraction, thereby improving image clarity and accuracy and providing high-quality basic data for subsequent feature extraction.

[0032] The preprocessing process for underwater terrain point cloud data acquired by the multibeam detection system involves removing outliers and erroneous data. These anomalies may be caused by equipment failure, signal interference, or reflections from non-target objects. After removing outliers, the remaining valid data undergoes coordinate transformation, unifying data collected at different times and spaces into the same spatial coordinate system. This ensures spatial alignment and consistency of the multimodal data for fusion analysis.

[0033] For structured meteorological and hydrological data collected by meteorological and hydrological monitoring stations, preprocessing involves filling missing values ​​and estimating missing data such as wind speed, tides, precipitation, and flow velocity using appropriate interpolation methods or time series-based forecasting models to ensure data integrity. Furthermore, data normalization is performed to convert data with different dimensions and value ranges to the same scale, such as [0, 1]. This eliminates dimensionality effects and improves the efficiency and stability of subsequent model training.

[0034] Optionally, in the underwater terrain data evolution analysis method provided in an embodiment of the present application, extracting features from the preprocessed data specifically includes: For the pre-processed underwater terrain image data, convolutional neural network image processing technology is used to extract the texture, landform and water body features in the image; Extract depth features and terrain relief features from the pre-processed underwater terrain point cloud data; For the pre-processed meteorological and hydrological structured data, variable characteristics related to water depth, tidal change characteristics, flow velocity, precipitation intensity and frequency characteristics are extracted.

[0035] Specifically, convolutional neural network (CNN) image processing technology is used to extract features from preprocessed underwater terrain image data. Specifically, the image data is fed into a pretrained CNN model, where the model's convolutional and pooling layers automatically learn and extract texture, landform, and water features from the image. For example, texture features can reflect the surface structure of different seabed geological types; landform features can depict topographic elements such as ridges, valleys, and plains; and water features can distinguish between different water bodies, such as boundaries and depth variations. These features are output as high-dimensional feature vectors, providing visual information for subsequent analysis.

[0036] Depth features and topographic relief features are extracted from preprocessed underwater terrain point cloud data. Depth features characterize the depth variations of underwater terrain by calculating the depth value of each point and depth statistics of its neighborhood, such as the mean, standard deviation, and range. Topographic relief features reflect the degree of topographic relief and its changing trends by calculating geometric quantities such as elevation difference, slope, and curvature between adjacent points. For example, the slope and curvature of each point are calculated using a moving window method to identify steep and flat areas of the terrain, providing a basis for analyzing the morphological characteristics of underwater terrain.

[0037] For the preprocessed meteorological and hydrological structured data, variable characteristics related to water depth, tidal variation characteristics, and flow velocity, precipitation intensity, and frequency characteristics are extracted. First, correlation analysis is used to identify meteorological and hydrological variables closely related to water depth, such as wind speed and precipitation, and these are used as part of the feature vector. Second, time series analysis is performed on the tidal data to extract characteristics such as tidal period, amplitude, and rate of change to reflect the short-term and long-term impact of tides on underwater topography. Finally, statistical analysis is performed on the flow velocity and precipitation data to extract characteristics such as the maximum value, average value, and duration of flow velocity, as well as the intensity, frequency, and cumulative precipitation. These characteristics can reflect the scouring effect of water flow and the indirect impact of precipitation on underwater topography, providing important environmental factors to support a comprehensive analysis of the evolution of underwater topography.

[0038] Optionally, in the underwater terrain data evolution analysis method provided in an embodiment of the present application, the step of combining the extracted multimodal features into a polynomial regression model specifically includes: The extracted underwater terrain image features are , the extracted underwater terrain point cloud features are , the extracted meteorological and hydrological characteristics are ; The independent variable X for constructing the polynomial regression model is

[0039] in, , , , is the coefficient to be optimized, 、 and are the number of extracted underwater terrain image features, underwater terrain point cloud features and meteorological and hydrological features respectively, and the high-order cross term is 、 and The cross product terms of . Specifically, the extracted underwater terrain image features are: , such as texture features, landform features, etc.; the extracted underwater terrain point cloud features are , such as depth mean, slope, etc.; the extracted meteorological and hydrological characteristics are , such as tidal changes, total precipitation, etc.

[0040] When constructing the independent variable X of the polynomial regression model based on the extracted features, the features of each mode are first linearly combined and assigned different coefficients. , , , to reflect the contribution of each feature to the evolution of underwater terrain. Specifically, for the image feature part, the weighted sum To integrate; for the feature part of the point cloud, through weighted sum To integrate; for the meteorological and hydrological characteristics, through weighted sum To integrate.

[0041] On this basis, cross terms are further introduced to capture the interaction between different modal features. For example, the cross term between image features and point cloud features is This cross-term reflects the coupling relationship between image features and point cloud features such as depth and slope, helping to more accurately depict the complex variations in underwater terrain. Similarly, cross-terms between image features and meteorological and hydrological characteristics, as well as cross-terms between point cloud features and meteorological and hydrological characteristics, can also be introduced. Furthermore, higher-order cross-terms, such as third-order cross-terms between image features, point cloud features, and meteorological and hydrological characteristics, are introduced to further explore deep nonlinear relationships between multimodal data.

[0042] Finally, the above parts are combined to form the independent variable X of the complete polynomial regression model, which is expressed as:

[0043] Among them, the coefficient , , , etc. are all parameters to be optimized, and their values ​​will be determined through the subsequent quantum annealing optimization process so that the model can fit the training data to the greatest extent and accurately predict the evolution trend of the underwater terrain. 、 and They represent the number of extracted underwater terrain image features, underwater terrain point cloud features and meteorological and hydrological features respectively. These parameters are determined according to actual data and feature extraction results.

[0044] Optionally, the underwater terrain data evolution analysis method provided in the embodiment of the present application defines the mean square error as the objective function, and the objective function for

[0045] in, is the sample size, is the actual observed value, is the predicted value of the polynomial regression model.

[0046] Specifically, the mean square error is defined as the objective function to measure the difference between the predicted value of the polynomial regression model and the actual observed value, thereby guiding the optimization process of the model parameters. The objective function is constructed to evaluate the prediction performance of the model. is the number of samples used to train the model, Indicates the The actual observation value of a sample, such as the actual water depth measurement at a certain point. This means that the polynomial regression model is The predicted value of a sample, such as the predicted water depth of the corresponding point calculated according to the regression model.

[0047] Objective function It is defined as the mean square error, and its calculation formula is

[0048] During the model training process, we use the quantum annealing algorithm to optimize the parameters of the polynomial regression model to minimize the objective function The quantum annealing algorithm simulates the annealing process of the quantum system from high temperature to low temperature, and uses the superposition and entanglement characteristics of quantum states to explore multiple parameter combinations at the same time, thereby effectively avoiding local optimal solutions and finding the optimal solution that makes the objective function The optimal parameter combination for minimization.

[0049] By minimizing the mean square error objective function, the model can fit the training data to the greatest extent and improve prediction accuracy.

[0050] Optionally, in the underwater terrain data evolution analysis method provided in an embodiment of the present application, minimizing the objective function by using a quantum annealing algorithm specifically includes: By simulating the annealing process of a quantum system from high temperature to low temperature, multiple parameter combinations are explored simultaneously based on the superposition and entanglement properties of quantum states; In each iteration, the parameters of the polynomial regression model are adjusted according to the current state of the quantum system, so that the objective function value gradually decreases, and finally the optimal parameter combination that minimizes the objective function is found.

[0051] Specifically, the quantum system is initialized, encoding the parameters of the polynomial regression model as quantum bits. The quantum annealing algorithm exploits the superposition and entanglement properties of quantum states to simultaneously explore multiple parameter combinations in the search space, significantly improving search efficiency. The initial temperature and annealing rate are set to control the evolution of the quantum system.

[0052] The iterative optimization phase then begins. In each iteration, the objective function value is calculated based on the current state of the quantum system. Through the quantum tunneling effect, the algorithm effectively avoids local optima, escapes local minima, and continues searching for the global optimal solution. The state of the qubits is adjusted, and the parameters of the polynomial regression model are updated to gradually reduce the objective function value. This process simulates the annealing process of a quantum system from high temperature to low temperature. As the temperature decreases, the system state gradually freezes, eventually converging to the optimal parameter combination that minimizes the objective function.

[0053] Finally, when the termination criteria are met (e.g., reaching the maximum number of iterations or convergence of the objective function), the optimized model parameters are output. These parameters represent the optimal solution to the polynomial regression model and can be used to predict and analyze underwater terrain data, achieving accurate predictions of underwater terrain evolution.

[0054] Optionally, an embodiment of the present application further provides an underwater terrain data evolution analysis device, comprising: Data acquisition module, used to collect multi-modal raw data of underwater terrain and pre-process the raw data; The feature extraction module is used to extract features from the preprocessed data and combine the extracted multimodal features into a polynomial regression model; a parameter optimization module, configured to use the parameters of the polynomial regression model as optimization variables of a quantum annealing algorithm, define an objective function, and minimize the objective function using the quantum annealing algorithm; The prediction module is used to predict the input data based on the optimized polynomial regression model and output the prediction results.

[0055] The present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above method. The computer-readable storage medium may include, but is not limited to, any type of disk, including a floppy disk, an optical disk, a DVD, a CD-ROM, a microdrive, a magneto-optical disk, a ROM, a RAM, an EPROM, an EEPROM, a DRAM, a VRAM, a flash memory device, a magnetic card or an optical card, a nanosystem (including a molecular memory IC), or any type of medium or device suitable for storing instructions and / or data.

[0056] It should be noted that for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application.

[0057] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0058] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interface, and the indirect coupling or communication connection of the device or unit can be electrical or other forms.

[0059] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0060] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0061] If the integrated unit 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 memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a memory and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned memory includes: U disk, read-only memory (ROM), random access memory (RAM), mobile hard disk, magnetic disk, or optical disk, etc., various media that can store program code.

[0062] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable memory, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0063] The above is only an exemplary embodiment of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the specification and practicing the disclosure herein, those skilled in the art will easily think of the implementation scheme of the present disclosure. This application is intended to cover any variation, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the art that are not recorded in the present disclosure. The description and examples are to be regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

[0064] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0065] It is easy for those skilled in the art to understand that the above is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A method for analyzing underwater terrain data evolution, characterized in that: include: Collect multimodal raw data of underwater terrain and preprocess the raw data; Extract features from the preprocessed data and combine the extracted multimodal features into a polynomial regression model; Using the parameters of the polynomial regression model as optimization variables of the quantum annealing algorithm, defining an objective function, and minimizing the objective function using the quantum annealing algorithm; Predict the input data based on the optimized polynomial regression model and output the prediction results.

2. The underwater terrain data evolution analysis method according to claim 1, characterized in that: The multimodal raw data of underwater terrain is collected, including: Acquiring underwater terrain image data by side-scan sonar, wherein the underwater terrain image data includes texture, landform, and water characteristics; Acquiring underwater terrain point cloud data using a multi-beam detection system, wherein the underwater terrain point cloud data includes depth and slope; Meteorological and hydrological structured data are acquired through meteorological and hydrological monitoring stations, wherein the meteorological and hydrological structured data include wind speed, tide, precipitation and flow velocity.

3. The underwater terrain data evolution analysis method according to claim 2, characterized in that: The preprocessing of the original data specifically includes: Performing radiation correction and geometric correction on the underwater terrain image data to eliminate image noise and distortion; For the underwater terrain point cloud data, outliers and erroneous data are removed, and coordinate transformation is performed to unify the data with other data in a spatial coordinate system; For the meteorological and hydrological structured data, missing values ​​are filled and the data is normalized.

4. The underwater terrain data evolution analysis method according to claim 3, wherein: Extracting features from the preprocessed data specifically includes: For the pre-processed underwater terrain image data, convolutional neural network image processing technology is used to extract the texture, landform and water body features in the image; Extract depth features and terrain relief features from the pre-processed underwater terrain point cloud data; For the pre-processed meteorological and hydrological structured data, variable characteristics related to water depth, tidal change characteristics, flow velocity, precipitation intensity and frequency characteristics are extracted.

5. The underwater terrain data evolution analysis method according to claim 4, characterized in that: The step of combining the extracted multimodal features into a polynomial regression model specifically includes: The extracted underwater terrain image features are , the extracted underwater terrain point cloud features are , the extracted meteorological and hydrological characteristics are ; The independent variable X for constructing the polynomial regression model is in, , , , is the coefficient to be optimized, 、 and are the number of extracted underwater terrain image features, underwater terrain point cloud features and meteorological and hydrological features respectively, and the high-order cross term is 、 and The cross product terms of .

6. The underwater terrain data evolution analysis method according to claim 5, characterized in that: Define mean square error as the objective function, the objective function for in, is the sample size, is the actual observed value, is the predicted value of the polynomial regression model.

7. The underwater terrain data evolution analysis method according to claim 1, wherein: The minimization of the objective function by the quantum annealing algorithm specifically includes: By simulating the annealing process of a quantum system from high temperature to low temperature, multiple parameter combinations are explored simultaneously based on the superposition and entanglement properties of quantum states; In each iteration, the parameters of the polynomial regression model are adjusted according to the current state of the quantum system, so that the objective function value gradually decreases, and finally the optimal parameter combination that minimizes the objective function is found.

8. An underwater terrain data evolution analysis device, characterized in that: include: Data acquisition module, used to collect multi-modal raw data of underwater terrain and pre-process the raw data; The feature extraction module is used to extract features from the preprocessed data and combine the extracted multimodal features into a polynomial regression model; a parameter optimization module, configured to use the parameters of the polynomial regression model as optimization variables of a quantum annealing algorithm, define an objective function, and minimize the objective function using the quantum annealing algorithm; The prediction module is used to predict the input data based on the optimized polynomial regression model and output the prediction results.

9. An underwater terrain data evolution analysis device, characterized in that: The method comprises at least one processing unit and at least one storage unit, wherein the storage unit stores a computer program, and when the computer program is executed by the processing unit, the processing unit executes the steps of the method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: It stores a computer program that can be executed by an underwater terrain data evolution analysis device. When the computer program runs on the underwater terrain data evolution analysis device, the underwater terrain data evolution analysis device executes the steps of the method described in any one of claims 1 to 7.

Images