Shipborne-submarine marine geophysical data real-time processing and analyzing system

Abstract

The invention relates to the technical field of ocean exploration, in particular to a shipborne-submarine ocean geophysical data real-time processing and analysis system, which comprises a data acquisition module, a data preprocessing module, an ocean geologic feature recognition module and a data analysis module, the data acquisition module is used for acquiring ocean geophysical data; the data preprocessing module is used for preprocessing the collected ocean geophysical data; the marine geologic feature recognition module is used for recognizing and classifying seabed geology, automatically recognizing geologic features of a marine bottom layer and dividing different geologic types according to the seabed geologic features; and the data analysis module analyzes the change trend of the marine geological environment. According to the method, different geological types can be efficiently identified, the precise geological map is generated, personal errors are greatly reduced, the precision of submarine geological exploration is improved, and the data analysis speed is increased.

Description

Technical Field

[0001] The present invention relates to the field of ocean exploration technology, and in particular to a shipborne or submarine-borne ocean geophysical data real-time processing and analysis system. Background Art

[0002] With the increasing demand for the development of marine resources and the protection of the marine environment, real-time processing and analysis of marine geophysical data has become one of the key technologies. Marine geophysical data, including seismic waves, gravity, electromagnetic, sonar, etc., can provide valuable information for seabed geological exploration, resource assessment and environmental monitoring. By acquiring and analyzing these data in real time, we can effectively evaluate the seabed geological structure, resource distribution, environmental changes, etc., and support marine scientific research and resource development. However, due to the complexity of the marine environment and the huge amount of data, traditional data acquisition and processing methods are often difficult to cope with the complex marine geological environment and real-time requirements.

[0003] The existing marine geophysical data processing systems have the following deficiencies in data collection and analysis. First, many systems rely on a single data source or sensor, resulting in limited dimensions and accuracy of the acquired data, which cannot fully reflect the complex characteristics of the seabed geology. Second, data processing and analysis usually rely on manual intervention, which is prone to errors and has low efficiency, making it difficult to process large-scale data in real time. In addition, existing technologies are insufficient in data fusion, automatic identification of geological features, and spatial analysis, and fail to effectively integrate multi-source data, resulting in low accuracy in the identification and classification of seabed geological features, and unable to provide accurate geological maps and trend predictions. Therefore, how to improve the accuracy and efficiency of data collection and how to achieve efficient real-time data analysis and geological feature identification have become problems that need to be urgently solved in current technology.

[0004] The purpose of the present invention is to provide a shipborne and submarine-borne marine geophysical data real-time processing and analysis system, which can significantly improve the accuracy and real-time performance of data acquisition, reduce human errors, and improve the recognition accuracy and analysis efficiency of seabed geological features. In addition, the system can also predict changes in the marine geological environment through trend analysis, providing accurate technical support and decision-making basis for marine resource exploration, environmental monitoring and disaster prevention. Summary of the invention

[0005] The present invention provides a shipborne and submarine-borne ocean geophysical data real-time processing and analysis system to solve the problems raised by the above-mentioned background technology.

[0006] The shipborne and submarine-borne marine geophysical data real-time processing and analysis system includes a data acquisition module, a data preprocessing module, a marine geological feature recognition module and a data analysis module, wherein; The data acquisition module collects marine geophysical data, including seismic wave data, gravity data, electromagnetic data, and sonar data, through a variety of sensors on ship-borne and submarine-borne equipment; The data preprocessing module preprocesses the collected marine geophysical data, including denoising, filtering and standardization; The marine geological feature recognition module recognizes and classifies the seabed geology based on the preprocessed marine geophysical data, automatically recognizes the geological features of the ocean bottom, and divides different geological types according to the seabed geological features; The data analysis module integrates the pre-processed marine geophysical data into comprehensive data, and analyzes the changing trend of the marine geological environment in combination with the identification results of the marine bottom geological characteristics.

[0007] Optionally, the data acquisition module includes: Seismic wave data acquisition: The seismic wave signals are transmitted and received by ship-borne or submarine-borne seismograph sensors, and the time delay between reflected and transmitted waves is recorded to obtain information on the geological structure of the bottom layer of the ocean; Gravity data collection: Through ship-borne and submarine-borne gravimeter sensors, the gravity field changes in the ocean area are monitored in real time to collect accurate ocean gravity data; Electromagnetic data acquisition: Using ship-borne or submarine-borne electromagnetic sensors to measure changes in the electromagnetic field at the bottom of the ocean and obtain conductivity and resistivity data of the seabed; Sonar data collection: Sonar equipment emits sound waves, receives their echoes, and collects the time and intensity changes of the echoes.

[0008] Optionally, the data preprocessing module includes: De-noising: The collected marine geophysical data are de-noised using the wavelet threshold denoising algorithm; Filtering: Use bandpass filtering algorithm to remove high-frequency noise and low-frequency drift in marine geophysical data; Standardization processing: The standard deviation normalization method is used to normalize the marine geophysical data.

[0009] Optionally, the marine geological feature identification module includes: Geological feature extraction: Based on the preprocessed marine geophysical data, a convolutional neural network (CNN) model is used to extract seabed geological features, including sediment type, rock layer thickness, wave velocity, and density; Geological feature classification: The extracted seabed geological features are classified through the seabed geological classification model, and the seabed geological features are automatically divided into different geological types, including soft sediments, hard rock layers, sand layers, and mud layers; Geological distribution and spatial analysis: Conduct spatial distribution analysis on classified seabed geological features, evaluate the distribution patterns and spatial relationships of seabed geological structures, use the K-means clustering algorithm to automatically identify the distribution patterns of seabed geology, and generate geological maps.

[0010] Optionally, the convolutional neural network (CNN) model includes: Multi-channel input processing: Use multi-channel input to process marine geophysical data of different characteristics and scales; Multi-scale convolution: For the marine geophysical data with multi-channel input and output, convolution kernels of various sizes are used for processing, and the results of multi-scale convolution are fused to generate feature maps; Spatial self-attention mechanism: Introduce the spatial self-attention mechanism to enhance the fused feature map by calculating the weight of each pixel position; Output layer and classification: Through the fully connected layer, the enhanced feature map is mapped to different geological categories, including sediment type, rock thickness, wave velocity, and density.

[0011] Optionally, the seabed geological classification model adopts a deep neural network (DNN) model, and the deep neural network (DNN) model includes: Receiving extracted seabed geological features: The input layer receives extracted seabed geological features, including sediment type, rock layer thickness, wave velocity, and density; Feature mapping and nonlinear transformation: The hidden layer maps the input seabed geological features to a high-dimensional feature space, and gradually extracts the seabed geological features through multiple hidden layers; Generate category predictions: The output layer uses the Softmax activation function to map the seabed geological features extracted by the hidden layer to the probability distribution of each geological category; Determine the final geological type: According to the output probability of the Softmax activation function, select the geological category with the highest probability as the final prediction result .

[0012] Optionally, the geological distribution and spatial analysis includes: Input data preparation: Collect the result data of geological feature classification. Each data point of the result data includes the category label of seabed geology (soft sediment, hard rock layer, sand layer, mud layer) and its corresponding spatial position (3D coordinates); Spatial distribution analysis: Use the K-means clustering algorithm to perform cluster analysis on the data points, automatically dividing the data points into K different clusters, each cluster represents a geological type with similar spatial distribution; Generate geological distribution map: Based on the results of K-means clustering algorithm, generate the spatial distribution map of seabed geology, showing the distribution patterns of different geological types (soft sediments, hard rock layers, sand layers, mud layers) in space, and divide each data point into Classification results Visualization, generation of geological maps.

[0013] Optionally, the data analysis module includes: Data fusion: Fusion processing of pre-processed marine geophysical data to generate comprehensive data; Trend analysis: Based on the generated comprehensive data, combined with the different geological types identified from the seabed geological features, the changing trend of the marine geological environment is identified.

[0014] Optionally, the data fusion includes: Data normalization: Use the minimum-maximum normalization method to normalize the marine geophysical data from different sources; Weighted fusion: Perform weighted fusion processing on the normalized marine geophysical data to generate comprehensive data .

[0015] Optionally, the trend analysis includes: Input data: The integrated data Integration with geological types as input data ; Trend analysis: Based on the integrated input data, the relationship between geological types and marine geophysical data is analyzed through regression models.

[0016] Beneficial effects of the present invention: The present invention collects marine geophysical data in real time through a variety of sensors, including multi-dimensional data such as seismic waves, gravity, electromagnetic, and sonar. The system ensures high accuracy and time synchronization of data through an efficient data acquisition module, and is combined with the collaborative work of multiple sensors. It has powerful data processing capabilities and can obtain comprehensive seabed geological information in real time in complex marine environments, significantly improving the accuracy and efficiency of marine resource exploration and environmental monitoring.

[0017] The present invention, by adopting a convolutional neural network model, can automatically extract complex geological features from preprocessed marine geophysical data and perform high-precision geological classification. Through multi-scale convolution and spatial self-attention mechanism, the system can efficiently identify different geological types and generate accurate geological maps. The automated recognition and classification function greatly reduces human errors, improves the accuracy of seabed geological exploration, and speeds up data analysis.

[0018] The present invention combines the pre-processed marine geophysical data with the geological type and adopts the regression analysis model to perform trend analysis, which can identify the changing trend of the marine geological environment and accurately predict future geological changes. Through weighted fusion processing, the quality of the comprehensive data is significantly improved, and the impact of different geological types on the physical characteristics of the ocean can be comprehensively analyzed. It not only helps scientists and engineers to better understand the dynamic changes of marine geology, but also provides a scientific basis and decision-making support for the exploration of marine resources, environmental protection and disaster prevention. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 A schematic diagram of system function modules according to an embodiment of the present invention; Figure 2 Schematic diagram of a marine geological feature identification module according to an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0022] like Figure 1-Figure 2 As shown, the shipborne and submarine-borne marine geophysical data real-time processing and analysis system includes a data acquisition module, a data preprocessing module, a marine geological feature recognition module and a data analysis module, wherein; The data acquisition module collects marine geophysical data through various sensors on ship-borne and submarine-borne equipment, including seismic wave data, gravity data, electromagnetic data, and sonar data; The data preprocessing module preprocesses the collected marine geophysical data, including denoising, filtering and standardization; The marine geological feature recognition module identifies and classifies the seabed geology based on the pre-processed marine geophysical data, automatically identifies the geological features of the ocean floor, and divides different geological types according to the seabed geological features; The data analysis module integrates the pre-processed marine geophysical data into comprehensive data, and analyzes the changing trend of the marine geological environment in combination with the identification results of the marine bottom geological characteristics; Through the above content, efficient processing and accurate analysis of marine geophysical data are ensured. Through multi-source data fusion, the system can identify seabed geological features in real time, improve the accuracy and efficiency of marine resource exploration and environmental monitoring. At the same time, the collaborative work and adaptive processing capabilities between modules enable the system to operate stably in complex marine environments and provide efficient decision support and trend prediction.

[0023] The data acquisition module includes: Seismic wave data acquisition: The seismic wave signals are transmitted and received by ship-borne or submarine-borne seismograph sensors, and the time delay between reflected and transmitted waves is recorded to obtain information on the geological structure of the bottom layer of the ocean; Gravity data collection: Through ship-borne and submarine-borne gravimeter sensors, the gravity field changes in the ocean area are monitored in real time to collect accurate ocean gravity data; Electromagnetic data acquisition: Using ship-borne or submarine-borne electromagnetic sensors to measure changes in the electromagnetic field at the bottom of the ocean and obtain conductivity and resistivity data of the seabed; Sonar data collection: emit sound waves through sonar equipment, receive their echoes, and collect the time and intensity changes of the echoes; Through the above content, various marine geophysical data including seismic waves, gravity, electromagnetic and sonar are collected in real time with extremely high flexibility and accuracy. Multiple sensors work together to not only provide comprehensive and three-dimensional marine geological information, but also ensure the time synchronization and spatial consistency of the data, significantly improving the reliability of data collection. By integrating different types of data, it can more comprehensively reflect the underlying structure and environmental changes of the ocean, has significant multi-dimensional perception capabilities, and supports more accurate marine exploration and environmental monitoring.

[0024] The data preprocessing module includes: Denoising: The collected marine geophysical data are denoised using the wavelet threshold denoising algorithm, which is expressed as: ; in, For signal The coefficient after wavelet transformation, T is based on the noise standard deviation and the threshold calculated from the signal length N ( ), is the sign function of the wavelet coefficients, It is a soft threshold function, which means that after subtracting the threshold T from the amplitude of the coefficient, if the result is negative, it is reset to zero, and the part greater than zero is retained. is the wavelet mother function, is the signal after denoising; Filtering: Use a bandpass filter algorithm to remove high-frequency noise and low-frequency drift in marine geophysical data to ensure that the frequency components of the signal are within the effective range, expressed as: ; in, is the impulse response function of the filter according to A time shift was performed. is the original signal, is the signal after filtering; Standardization: The standard deviation standardization method is used to normalize the marine geophysical data to eliminate the impact of different dimensions or magnitudes, which is expressed as: ; in, is the original ocean geophysical data, For signal The mean of is the standard deviation of the signal, is the standardized data; Through the above content, the quality and accuracy of marine geophysical data are effectively improved. De-noising removes random noise in the signal through algorithms such as wavelet transform to ensure the purity of the signal. The filtering operation effectively removes high-frequency noise and low-frequency drift, making the data smoother and more stable. Standardization processing eliminates the dimensional differences between different data types and unifies the scale of the data. On the whole, the data preprocessing module can provide high-quality and clear data input for subsequent analysis, greatly improving the accuracy and efficiency of data analysis, while reducing errors caused by noise or external factors, ensuring the accuracy and availability of marine geophysical data in complex marine environments.

[0025] The marine geological feature identification module includes: Geological feature extraction: Based on the preprocessed marine geophysical data, a convolutional neural network (CNN) model is used to extract seabed geological features, including sediment type, rock layer thickness, wave velocity, and density; Geological feature classification: The extracted seabed geological features are classified through the seabed geological classification model, and the seabed geological features are automatically divided into different geological types, including soft sediments, hard rock layers, sand layers, and mud layers; Geological distribution and spatial analysis: Conduct spatial distribution analysis on classified submarine geological features, evaluate the distribution patterns and spatial relationships of submarine geological structures, use the K-means clustering algorithm to automatically identify the distribution patterns of submarine geology, and generate geological maps; Through the above content, the seabed geological characteristics are effectively extracted and classified from the pre-processed marine geophysical data, which significantly improves the automation and accuracy of marine geological exploration. It can not only automatically identify and classify different geological types (such as soft sediments, hard rock layers, sand layers, mud layers, etc.), but also use spatial analysis technology to deeply analyze the distribution patterns and spatial relationships of seabed geological structures, thereby generating high-precision geological maps, which can greatly improve work efficiency, reduce human errors, provide faster and more accurate marine geological information, support decision-making in multiple fields such as resource exploration, environmental monitoring and disaster warning, and provide strong technical support for marine scientific research and seabed resource development.

[0026] The Convolutional Neural Network (CNN) model consists of: Multi-channel input processing: Multi-channel input is used to process marine geophysical data of different characteristics and scales, expressed as: ; ; ; in, , , are the weights of the convolution kernels for seismic waves, gravity, and electromagnetic data, respectively. , , They are seismic waves, gravity, and electromagnetic data, , , are the bias terms for seismic wave, gravity and electromagnetic data respectively, , , These are the convolution outputs of seismic waves, gravity, and electromagnetic data respectively; Multi-scale convolution: For the marine geophysical data with multi-channel input and output, a variety of convolution kernels of different sizes are used for processing, and the results of multi-scale convolution are fused to generate a feature map, which is expressed as: ; ; ; in, , , are the convolution kernel weights at different scales (3×3, 5×5, 7×7), X is the input data, including , , , , , They are the convolution kernel bias items at different scales (3×3, 5×5, 7×7), , , The output of the multi-scale convolution layer corresponds to the convolution results of different scales; ; in, is the fused feature map, , , are the learning weights for each scale respectively; Spatial self-attention mechanism: The spatial self-attention mechanism is introduced to enhance the fused feature map by calculating the weight of each pixel position, which is expressed as: ; Among them, A is the spatial attention weight, and the Softmax function calculates the relative weight of each position. is the temperature coefficient; ; in, It is the feature map enhanced by the attention mechanism; Output layer and classification: Through the fully connected layer, the enhanced feature map is mapped to different geological categories, including sediment type, rock thickness, wave velocity, and density, expressed as: ; ; Among them, Y is the output of the fully connected layer, is the weight matrix, is the bias, is the class probability; Through the above content, it is possible to effectively integrate and extract features from different physical data sources (such as seismic waves, gravity, electromagnetic, etc.), so as to comprehensively capture the multi-level and complex characteristics of marine geology. Multi-scale convolution improves the model's ability to recognize rock layers and sediments of different thicknesses. The spatial self-attention mechanism enables the model to focus on key geological features and improve recognition accuracy. Through multi-label classification, the model can simultaneously predict multiple geological features and provide accurate geological information for marine resource exploration and environmental monitoring.

[0027] The seabed geological classification model uses a deep neural network (DNN) model, which includes: Receive extracted seabed geological features: The input layer receives extracted seabed geological features, including sediment type, rock layer thickness, wave velocity, and density, expressed as: ; in, is the seafloor geological characteristics of the input layer, For sediment type, is the rock layer thickness, is the wave speed, is the density; Feature mapping and nonlinear transformation: The hidden layer maps the input seabed geological features to a high-dimensional feature space, and gradually extracts the seabed geological features through multiple hidden layers, which can be expressed as: ; in, is the weighted sum of the lth layer, is the weight matrix of the lth layer, is the activation value of the previous layer (for the first layer, it is the input feature ), is the bias term of the lth layer; ; ; in, is the activation value of the lth layer; Generate category predictions: The output layer uses the Softmax activation function to map the seabed geological features extracted by the hidden layer to the probability distribution of each geological category, expressed as: ; in, is the linear output of the output layer, is the weight matrix of the output layer, is the activation value of the last hidden layer, is the bias term of the output layer; ; in, is the predicted probability for each category, is the linear result of the output layer; ; Where C is the number of categories, is the output of the i-th class, is the predicted probability of the ith class, is the output of the jth class; Determine the final geological type: According to the output probability of the Softmax activation function, select the geological category with the highest probability as the final prediction result , expressed as: ; in, The final geological category predicted (soft sediment, hard rock, sand, mud); Through the above content, it is possible to automatically learn complex nonlinear relationships from the extracted seabed geological features without the need for manual feature design, which greatly improves processing efficiency. Secondly, the multi-layer structure of DNN can gradually extract and optimize features, from low-level local information to high-level global patterns, capturing the complex characteristics of seabed geology and making the classification process more accurate. Furthermore, DNN can generate a probability distribution for each geological category through Softmax classification output, provide clear classification results, and adapt to the automatic identification of different geological types (such as soft sediments, hard rock layers, sand layers, and mud layers). Finally, DNN's high-dimensional data processing capabilities and powerful learning capabilities enable it to effectively handle the multidimensionality and complexity of marine geological data, providing accurate and stable technical support for marine resource exploration and environmental monitoring.

[0028] Geological distribution and spatial analysis include: Input data preparation: Collect the result data of geological feature classification. Each data point of the result data includes the category label of the seabed geology (soft sediment, hard rock layer, sand layer, mud layer) and its corresponding spatial position (3D coordinates), expressed as: ; Where D is the seabed geological feature dataset, is the spatial position of the i-th seafloor geological data point, is the classification label of the i-th seafloor geological data point, and N is the total number of data points; Spatial distribution analysis: Use the K-means clustering algorithm to perform cluster analysis on the data points, automatically dividing the data points into K different clusters, each cluster represents a geological type with similar spatial distribution; The K-means clustering algorithm specifically includes: (1) Initialize cluster centers: Randomly select K initial cluster centers as the starting point of clustering and initialize K cluster centers , , , ; (2) Assign each data point to the nearest cluster: For each data point , calculate its distance to each cluster center using Euclidean distance and assign it to the nearest cluster, expressed as: ; in, It is a data point With cluster center The Euclidean distance between ; in, Represents data points The cluster to which it belongs; (3) Update cluster centers: Calculate the mean of all data points in each cluster and use the mean as the new cluster center, expressed as: ; in, Cluster The number of data points in is the cluster center, and the updated cluster center is the average value of the positions of all data points in the cluster; (4) Iterate until convergence: Repeat steps (2) and (3) until the change in cluster center is less than a preset threshold. (i.e. convergence), indicating that the clustering process is completed, expressed as: ; in, and Respectively represent the cluster centers before and after the update, is the set convergence threshold (set according to the scale and characteristics of the data. For cases where data points are dense and feature changes are small, a smaller value is set. , set to For cases where data points are sparsely distributed and vary greatly, choose a larger , set to ); Generate geological distribution map: Based on the results of K-means clustering algorithm, generate the spatial distribution map of seabed geology, showing the distribution patterns of different geological types (soft sediments, hard rock layers, sand layers, mud layers) in space, and divide each data point into Classification results Visualization, generating geological maps; Through the above content, the spatial distribution patterns of different geological types (soft sediments, hard rock layers, sand layers, mud layers) can be automatically identified, providing accurate geological information. The aggregation and distribution patterns of geological features can be automatically discovered from spatial data, avoiding human intervention. By effectively dividing geological clusters and generating geological maps, the spatial relationship of seabed geological structures can be clearly displayed, providing strong support for marine resource exploration, environmental monitoring, etc. The generated geological maps provide intuitive visualization results for subsequent analysis, which is helpful for a comprehensive understanding of complex geological environments.

[0029] Data analysis modules include: Data fusion: Fusion processing of pre-processed marine geophysical data to generate comprehensive data; Trend analysis: Based on the generated comprehensive data, combined with the different geological types identified by the seabed geological features, the changing trend of the marine geological environment is identified; Through the above content, it is possible to fully integrate marine geophysical data from different data sources, reveal the dynamic change trend of seabed geology, help predict future changes in the geological environment, reduce human intervention, improve accuracy and efficiency, and provide clear visualization results to support more accurate decision-making and planning.

[0030] Data fusion includes: Data normalization: The minimum-maximum normalization method is used to normalize the marine geophysical data from different sources to eliminate the dimensional differences between different data sources, which is expressed as: ; in, is the normalized data, x is the original data, and are the minimum and maximum values ​​of the original data respectively; Weighted fusion: Perform weighted fusion processing on the normalized marine geophysical data to generate comprehensive data , expressed as: ; in, is the integrated data after fusion. is the location of the data point, , , , is the corresponding data value, , , , is the weight coefficient for different ocean geophysical data; Through the above content, multi-source data can be effectively integrated, dimensional differences can be eliminated, and various types of data can be processed at a unified scale. This not only improves the consistency and accuracy of the data, but also assigns weights according to the importance of different data sources, thereby optimizing the quality of the integrated data.

[0031] Trend analysis includes: Input data: The integrated data Integration with geological types as input data , expressed as: ; in, is the geological type of the i-th data point (soft sediment, hard rock layer, sand layer, mud layer); Trend analysis: Based on the integrated input data, the relationship between geological types and marine geophysical data is analyzed through regression models, which can be expressed as: ; in, is the predicted value of the ocean geophysical data for the ith data point, is the weighted ocean geophysical data, is the geological type of the ith data point, , is the regression coefficient, Is related to geological type The correlation coefficient, is the error term, n is the number of marine geophysical data; Through the above content, it is possible to quantitatively evaluate the impact of different geological types on the physical characteristics of the ocean, thereby helping to identify and predict changes in the seabed geological structure, and providing accurate trend forecasts to help scientists and engineers better understand the dynamic changes in marine geology, and support applications such as marine resource exploration, environmental protection, and disaster prevention. In addition, the regression model of trend analysis can reveal the underlying laws and correlations through in-depth analysis of the physical and geological characteristics of the ocean, providing reliable data support for decision-making, and improving prediction accuracy and the scientific nature of decision-making.

[0032] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0033] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A shipborne and submarine-borne marine geophysical data real-time processing and analysis system, characterized in that: It includes data acquisition module, data preprocessing module, marine geological feature recognition module and data analysis module, among which; The data acquisition module collects marine geophysical data, including seismic wave data, gravity data, electromagnetic data, and sonar data, through a variety of sensors on ship-borne and submarine-borne equipment; The data preprocessing module preprocesses the collected marine geophysical data, including denoising, filtering and standardization; The marine geological feature recognition module recognizes and classifies the seabed geology based on the preprocessed marine geophysical data, automatically recognizes the geological features of the ocean bottom, and divides different geological types according to the seabed geological features; The data analysis module integrates the pre-processed marine geophysical data into comprehensive data, and analyzes the changing trend of the marine geological environment in combination with the identification results of the marine bottom geological characteristics.

2. The shipborne and submarine-borne marine geophysical data real-time processing and analysis system according to claim 1 is characterized in that: The data acquisition module comprises: Seismic wave data acquisition: The seismic wave signals are transmitted and received by ship-borne or submarine-borne seismograph sensors, and the time delay between reflected and transmitted waves is recorded to obtain information on the geological structure of the bottom layer of the ocean; Gravity data collection: Through ship-borne and submarine-borne gravimeter sensors, the gravity field changes in the ocean area are monitored in real time to collect accurate ocean gravity data; Electromagnetic data acquisition: Using ship-borne or submarine-borne electromagnetic sensors to measure changes in the electromagnetic field at the bottom of the ocean and obtain conductivity and resistivity data of the seabed; Sonar data collection: Sonar equipment emits sound waves, receives their echoes, and collects the time and intensity changes of the echoes.

3. The shipborne and submarine-borne marine geophysical data real-time processing and analysis system according to claim 1 is characterized in that: The data preprocessing module comprises: De-noising: The collected marine geophysical data are de-noised using the wavelet threshold denoising algorithm; Filtering: Use bandpass filtering algorithm to remove high-frequency noise and low-frequency drift in marine geophysical data; Standardization processing: The standard deviation normalization method is used to normalize the marine geophysical data.

4. The shipborne and submarine-borne marine geophysical data real-time processing and analysis system according to claim 1 is characterized in that: The marine geological feature recognition module includes: Geological feature extraction: Based on the preprocessed marine geophysical data, a convolutional neural network model is used to extract seabed geological features, including sediment type, rock layer thickness, wave velocity, and density; Geological feature classification: The extracted seabed geological features are classified through the seabed geological classification model, and the seabed geological features are automatically divided into different geological types, including soft sediments, hard rock layers, sand layers, and mud layers; Geological distribution and spatial analysis: Conduct spatial distribution analysis on classified seabed geological features, evaluate the distribution patterns and spatial relationships of seabed geological structures, use the K-means clustering algorithm to automatically identify the distribution patterns of seabed geology, and generate geological maps.

5. The shipborne and submarine-borne marine geophysical data real-time processing and analysis system according to claim 4 is characterized in that: The convolutional neural network model includes: Multi-channel input processing: Use multi-channel input to process marine geophysical data of different characteristics and scales; Multi-scale convolution: For the marine geophysical data with multi-channel input and output, convolution kernels of various sizes are used for processing, and the results of multi-scale convolution are fused to generate feature maps; Spatial self-attention mechanism: Introduce the spatial self-attention mechanism to enhance the fused feature map by calculating the weight of each pixel position; Output layer and classification: Through the fully connected layer, the enhanced feature map is mapped to different geological categories, including sediment type, rock thickness, wave velocity, and density.

6. The shipborne and submarine-borne marine geophysical data real-time processing and analysis system according to claim 5 is characterized in that: The seabed geological classification model adopts a deep neural network model, and the deep neural network model includes: Receiving extracted seabed geological features: The input layer receives extracted seabed geological features, including sediment type, rock layer thickness, wave velocity, and density; Feature mapping and nonlinear transformation: The hidden layer maps the input seabed geological features to a high-dimensional feature space, and gradually extracts the seabed geological features through multiple hidden layers; Generate category predictions: The output layer uses the Softmax activation function to map the seabed geological features extracted by the hidden layer to the probability distribution of each geological category; Determine the final geological type: According to the output probability of the Softmax activation function, select the geological category with the highest probability as the final prediction result .

7. The shipborne and submarine-borne marine geophysical data real-time processing and analysis system according to claim 6 is characterized in that: The geological distribution and spatial analysis include: Input data preparation: Collect the result data of geological feature classification. Each data point of the result data includes the category label of seabed geology and its corresponding spatial location. Spatial distribution analysis: Use the K-means clustering algorithm to perform cluster analysis on the data points, automatically dividing the data points into K different clusters, each cluster represents a geological type with similar spatial distribution; Generate geological distribution map: Based on the results of K-means clustering algorithm, generate the spatial distribution map of seabed geology, showing the distribution pattern of different geological types in space, and divide each data point into Classification results Visualization, generation of geological maps.

8. The shipborne and submarine-borne marine geophysical data real-time processing and analysis system according to claim 7 is characterized in that: The data analysis module includes: Data fusion: Fusion processing of pre-processed marine geophysical data to generate comprehensive data; Trend analysis: Based on the generated comprehensive data, combined with the different geological types identified from the seabed geological features, the changing trend of the marine geological environment is identified.

9. The shipborne and submarine-borne marine geophysical data real-time processing and analysis system according to claim 8 is characterized in that: The data fusion includes: Data normalization: Use the minimum-maximum normalization method to normalize the marine geophysical data from different sources; Weighted fusion: Perform weighted fusion processing on the normalized marine geophysical data to generate comprehensive data .

10. The shipborne and submarine-borne marine geophysical data real-time processing and analysis system according to claim 9 is characterized in that: The trend analysis includes: Input data: The integrated data Integration with geological types as input data ; Trend analysis: Based on the integrated input data, the relationship between geological types and marine geophysical data is analyzed through regression models.

Images