A marine digital fusion system and method based on GIS

Abstract

The present invention discloses a GIS-based marine digital integration system and method, which relates to the marine digital technology field. The system includes: a business integration module, which is used to realize that the originally independent business modules can provide data support for multiple applications at the same time through data integration and application integration; a technology integration module, which is used to realize technical intercommunication between different applications through framework integration and technology stack integration; an interaction integration module, which is used to realize visual interaction design based on business logic through UI integration and Web component integration. According to the technical solution of the present application, the business integration, technology integration and interaction integration of the marine digital business system and application scenarios can be realized, which has high application value.

Description

Technical Field

[0001] The present invention relates to the field of marine digital technology, and in particular to a marine digital fusion system and method based on GIS. Background Art

[0002] Application scenarios are the main carriers of digital reform, and are the pillars and links for realizing digital empowerment and institutional reconstruction. In recent years, under the wave of digital reform, batches of excellent digital application results have been born, providing efficient, stable and reliable technical support and decision-making assistance for urban smart governance. However, some inevitable problems have also followed. For example, some research and development results have duplicated construction in data governance and application development; in some areas, due to the limited skills of system operation and maintenance personnel, the workload at the grassroots level has increased instead of decreased. The duplicate construction of the system and the increase in the burden of grassroots operation and maintenance have further increased the complexity of digital application management and the operation and maintenance costs, especially in the field of marine business, involving multiple independent systems. Therefore, based on this problem, this application proposes a marine digital fusion system and method based on GIS. Summary of the invention

[0003] Technical Purpose

[0004] In order to solve the above problems, the purpose of the present invention is to provide a GIS-based marine digital integration system and method, which targets four types of marine business systems and application scenarios, namely marine space resource supervision, marine ecological early warning and evaluation, marine disaster perception and prevention, and marine economic monitoring and evaluation, and realizes the business integration, technical integration and interactive integration of business systems and application scenarios from the three dimensions of business, technology and interaction. The system and method have a wide range of application scenarios, which solves the problem of duplicate construction of some marine digital research and development results in data governance and application development, and reduces the burden of grassroots operation and maintenance.

[0005] Technical Solution

[0006] In order to achieve the above-mentioned objectives, the present invention provides a GIS-based marine digital integration system and method, which realizes that the originally independent business modules can provide data support for multiple applications at the same time through data integration and application integration, realizes technical intercommunication between different applications through framework integration and technology stack integration, and realizes visual interactive design based on business logic through UI integration and Web component integration, thereby realizing the deep integration of marine digital business systems and application scenarios in the three dimensions of business, technology and interaction.

[0007] In a first aspect, the present invention provides a marine digital fusion system based on GIS, comprising:

[0008] Business integration module, which is used to realize that originally independent business modules can provide data support for multiple applications at the same time through data integration and application integration;

[0009] Technology integration module, which is used to achieve technical interoperability between different applications through framework integration and technology stack integration;

[0010] The interactive fusion module is used to implement visual interactive design based on business logic through UI fusion and Web component fusion.

[0011] Furthermore, the data fusion specifically includes: adding space and time as indexes on the basis of existing business operations, and according to data visualization and storage scheduling requirements, for data types such as IoT perception, GIS vectors, image maps, etc., based on the CGCS2000 coordinate system and based on the WebGIS standard data format GeoJSON, two-dimensional spatial visualization and three-dimensional multi-source data fusion management are performed on GDB, MDB, SHP, ASCⅡ, NetCDF and other format data.

[0012] Furthermore, the database is divided into a business database and a spatial database according to application requirements. The business database is stored in MySQL, and the spatial database is stored in PostgreSQL, thereby realizing the separate storage of patch data and attribute data. Object storage technologies such as Minio are used to improve the storage and transmission efficiency of GeoJSON spatial data, thereby reducing the response load of the server when spatial data is concurrently overloaded.

[0013] Furthermore, based on the digital platform for spatial governance and the integrated intelligent public data platform, a data warehouse management submodule is built according to the integrated results of data governance and storage, and through data conversion and mapping technology. The data warehouse management submodule includes: service publishing, application monitoring, call trends, data sharing and other functions, which realizes the integrated management of multi-source heterogeneous marine data and can uniformly dispatch two- and three-dimensional spatial GIS and business data.

[0014] Furthermore, the application integration specifically includes: based on marine space resource data, combining data to develop intelligent site selection applications; combining marine economic operation monitoring and evaluation data to develop marine economic heat maps, spatial distribution of marine-related enterprises and other applications; combining front-end IoT perception data to develop intelligent identification of illegal sea use behaviors and marine disaster early warning and observation applications; combining marine comprehensive management data to develop comprehensive decision-making cockpit applications.

[0015] Furthermore, the technical integration module realizes the integration of the framework based on the front-end and back-end separation method and the Nginx proxy forwarding method, wherein the front-end adopts the Vue method, the back-end adopts the Springboot method, and the Server components in different applications are designed according to the microservice concept, and the independent deployment and load distribution of new and old back-end application services are realized through the Nginx proxy forwarding technology. It improves the efficiency of integration and subsequent application research and development work, and improves the compatibility and scalability of marine-related applications.

[0016] Furthermore, the technology stack integration specifically includes: encapsulating codes with large differences through secondary development, using WebGL three-dimensional graphics rendering method and OpenLayers two-dimensional GIS rendering method on the front end to meet the front-end rendering applications of various graphics engine platforms; using service registration method on the back end to uniformly schedule various services provided by the platform by type to maximize service compatibility.

[0017] Furthermore, according to the business needs and interaction logic of the marine system, UI integration is achieved based on the CSS3 method. The homepage of the application cockpit includes five interactive areas: GIS base, top title bar and business management entrance, analysis overview, toolbox and integrated application quick entrance. The element content in the interactive area is determined according to different application scenarios.

[0018] Furthermore, based on the MVVM method, the top title bar and four interactive areas, including business management entrance, analysis overview, toolbox and fusion application quick entrance, are covered on the GIS base area without overloading the two- and three-dimensional visualization components in the GIS base area, and the same component elements in each area are reused to reduce front-end overload and back-end request load, thereby achieving seamless switching between application scenarios.

[0019] Furthermore, the system also includes an application development module, which integrates GDAL and PostGIS technologies based on the PostgreSQL method to serve as a customized spatial analysis engine outside the supporting platform, and meets various application needs through spatial analysis functions; based on the REST and GeoJSON methods, it integrates WebGL and OpenLayers front-end rendering methods to achieve efficient two- and three-dimensional data query and spatial analysis functions; based on the FFmpeg method, it integrates Video.js and Flv.js methods to meet the transmission and loading of various video stream data, thereby ensuring the development needs of business applications and integrated applications.

[0020] Furthermore, the system also includes a marine geological data 3D display module, which generates a digital elevation model by performing spatial interpolation operations on GIS data, and vertically stretches the irregular triangulated network converted from raster data to establish a 3D model of the seabed strata. It can provide a scientific basis for the development and management of marine resources, improve the efficiency of data utilization, and reduce the phenomenon of information islands.

[0021] Furthermore, the system also includes a machine learning module, which extracts features from GIS data, trains the data representation with more semantic information through a random forest model, and predicts the future direction of the marine environment monitoring project through an adaptive cross-validation method. It can realize the automation and intelligence of marine environmental monitoring, reduce manual intervention, improve the efficiency and response speed of monitoring, provide a scientific basis for marine environmental management, and help decision makers formulate more effective environmental management strategies.

[0022] In a second aspect, the present invention further provides a GIS-based marine digital fusion method, the method is based on the system described in the first aspect, and comprises:

[0023] Through data fusion and application fusion, originally independent business modules can provide data support for multiple applications at the same time;

[0024] Achieve technical interoperability between different applications through framework integration and technology stack integration;

[0025] Visual interaction design based on business logic is achieved through UI fusion and Web component fusion.

[0026] In a third aspect, the present invention further provides a computer device, comprising a processor and a memory, wherein the processor is connected to the memory, the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the computer device executes the aforementioned GIS-based marine digital fusion method.

[0027] In a fourth aspect, the present invention further provides a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the aforementioned marine digital fusion method based on GIS is implemented.

[0028] The present invention improves the overall efficiency of marine business systems through the deep integration of business, technology and interaction, specifically including the integration of application scenarios of four types of business systems, namely, marine space resource supervision, marine ecological early warning and evaluation, marine disaster perception and prevention, and marine economic monitoring and evaluation. In terms of data fusion, a unified CGCS2000 coordinate system and WebGIS standard are adopted to realize the comprehensive management and scheduling of two-dimensional and three-dimensional spatial data. In terms of technical integration, different development frameworks and technology stacks are integrated through front-end and back-end separation technology, microservice architecture, Nginx proxy and other means to improve the compatibility and scalability of the system. In terms of interactive integration, the operating experience of various marine applications is optimized through a uniformly designed user interface, realizing seamless switching and collaborative work of business systems. The system and method effectively avoid the problem of duplicate construction in data governance and application development of marine digital research and development results, reduce the burden of grassroots operation and maintenance, improve the effectiveness of digital reform, and provide stable and reliable technical support for the expansion of marine business in the future.

[0029] Beneficial Effects

[0030] By implementing the above-mentioned GIS-based marine digital fusion system and method provided by the present invention, the following technical effects are achieved:

[0031] (1) Through data fusion and application fusion, originally independent business modules can provide data support for multiple applications at the same time, enrich the basis for analysis and decision-making, strengthen business collaboration, reduce the response load of the server when spatial data is concurrently overloaded, improve the storage and transmission efficiency of GeoJSON spatial data, and realize the fusion management of multi-source heterogeneous marine data with physical unification as the main and logical unification as the auxiliary.

[0032] (2) Framework fusion and technology stack fusion have achieved technical interoperability between different applications, which has improved the efficiency of fusion and subsequent application development, improved the compatibility and scalability of marine-related applications, and can meet the front-end rendering applications of various graphics engine platforms, thereby maximizing the compatibility of services.

[0033] (3) UI fusion and Web component fusion can realize visual interaction design based on business logic. Based on the principle of "simple and easy to use", it satisfies both regulatory coordination and scientific research user profiles. While reducing the operation and rendering pressure between the front-end visualization engine and the browser, it also reduces the computing load of the back-end data scheduling and statistical analysis services, thus realizing seamless switching between various application scenarios.

[0034] (4) Perform spatial interpolation on GIS data to generate a digital elevation model, and vertically stretch the irregular triangulated network converted from raster data to establish a three-dimensional model of the seabed strata. This can provide a scientific basis for the development and management of marine resources, improve data utilization efficiency, and reduce the phenomenon of information islands.

[0035] (5) The solution of the present invention also extracts features from GIS data, trains the data representation with more semantic information through a random forest model, and predicts the changing trend of the marine environment through an adaptive cross-validation method. It can realize the automation and intelligence of marine environmental monitoring, reduce human intervention, improve the efficiency and response speed of monitoring, provide a scientific basis for marine environmental management, and help decision makers formulate more effective environmental management strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to make the above-mentioned GIS-based marine digital fusion system and method of the present invention more obvious and easy to understand, the following is a brief introduction to the drawings required for use in the specific implementation of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0037] Figure 1 A schematic diagram showing the marine digital fusion method based on GIS;

[0038] Figure 2 It represents the schematic diagram of the framework fusion deployment architecture;

[0039] Figure 3 A schematic diagram showing an interactive fusion layout;

[0040] Figure 4 Schematic diagram of the fusion framework. DETAILED DESCRIPTION

[0041] Embodiment 1:

[0042] A GIS-based marine digital fusion system and method is provided. The GIS-based marine digital fusion method is as follows: Figure 1 As shown, the system includes: a business integration module, a technical integration module and an interactive integration module, which are as follows:

[0043] The business integration module is used to enable originally independent business modules to provide data support for multiple applications at the same time through data integration and application integration.

[0044] The data fusion specifically includes: adding space and time as indexes on the basis of existing business operations, and according to data visualization and storage scheduling requirements, for data types such as IoT perception, GIS vectors, image maps, etc., based on the CGCS2000 coordinate system and based on the WebGIS standard data format GeoJSON, two-dimensional spatial visualization and three-dimensional multi-source data fusion management are performed on GDB, MDB, SHP, ASCⅡ, NetCDF and other format data.

[0045] According to application requirements, the database is divided into a business database and a spatial database. The business database is stored in MySQL, and the spatial database is stored in PostgreSQL, thereby realizing the separate storage of patch data and attribute data. Object storage technologies such as Minio are used to improve the storage and transmission efficiency of GeoJSON spatial data, thereby reducing the response load of the server when spatial data is concurrently overloaded.

[0046] Based on the digital platform for spatial governance and the integrated intelligent public data platform, a data warehouse management submodule is built according to the integrated results of data governance and storage, and through data conversion and mapping technology. The data warehouse management submodule includes: service publishing, application monitoring, call trends, data sharing and other functions, which realizes the integrated management of multi-source heterogeneous marine data and can uniformly dispatch two- and three-dimensional spatial GIS and business data.

[0047] The application integration specifically includes: based on marine space resource data, combining data to develop intelligent site selection applications; combining marine economic operation monitoring and evaluation data to develop marine economic heat maps, spatial distribution of marine-related enterprises and other applications; combining front-end IoT perception data to develop intelligent identification of illegal sea use behaviors and marine disaster early warning and observation applications; combining marine comprehensive management data to develop comprehensive decision-making cockpit applications.

[0048] The technology integration module is used to achieve technical interoperability between different applications through framework integration and technology stack integration.

[0049] The technical integration module realizes the integration of frameworks based on the front-end and back-end separation method and the Nginx proxy forwarding method. The framework integration deployment architecture is as follows: Figure 2 As shown, the Vue method is used in the front end, the Springboot method is used in the back end, and the Server components in different applications are designed according to the microservice concept. The independent deployment and load distribution of new and old back-end application services are realized through Nginx proxy forwarding technology.

[0050] The technology stack fusion specifically includes: encapsulating codes with large differences through secondary development, using WebGL 3D graphics rendering method and OpenLayers 2D GIS rendering method in the front end to meet the front-end rendering applications of various graphics engine platforms; using service registration method in the back end to uniformly schedule various services provided by the platform by type to maximize service compatibility. The technology stack fusion is shown in Table 1:

[0051] Table 1: Technology stack integration

[0052]

[0053] The interactive fusion module is used to implement visual interactive design based on business logic through UI fusion and Web component fusion.

[0054] According to the business requirements and interaction logic of the marine system, the UI integration is realized based on the CSS3 method. The interactive integration layout is as follows: Figure 3 As shown, the homepage of the application cockpit includes five interactive areas: GIS base, top title bar and business management entrance, analysis overview, toolbox and fusion application quick entrance. The element content in the interactive area is determined according to different application scenarios.

[0055] Based on the MVVM method, the top title bar and four interactive areas, including business management entrance, analysis overview, toolbox and fusion application quick entrance, are covered on the GIS base area without overloading the two- and three-dimensional visualization components in the GIS base area. The same component elements in each area are reused to reduce front-end overload and back-end request load, thereby achieving seamless switching between application scenarios.

[0056] The system also includes an application development module, which integrates GDAL and PostGIS technologies based on the PostgreSQL method to serve as a customized spatial analysis engine outside the support platform, and meets various application needs through spatial analysis functions; based on the REST and GeoJSON methods, it integrates WebGL and OpenLayers front-end rendering methods to achieve efficient two- and three-dimensional data query and spatial analysis functions; based on the FFmpeg method, it integrates Video.js and Flv.js methods to meet the transmission and loading of various video stream data, thereby ensuring the development needs of business applications and integrated applications.

[0057] The system also includes a marine geological data three-dimensional display module, which generates a digital elevation model by performing spatial interpolation operations on GIS data, and vertically stretches an irregular triangulated network converted from raster data to establish a three-dimensional model of the seabed strata, specifically including:

[0058] Generate point feature vector files through preprocessed data, and perform spatial interpolation operations to generate digital elevation models. The interpolation method formula is as follows:

[0059]

[0060] in, Indicates the value of the forecast point; Represents the value of a known point; Indicates the distance between the predicted point and the known point; Represents the weight assigned to each known point based on data quality or other relevant factors; Represents the weight index.

[0061] The irregular triangulated network converted from raster data is vertically stretched to establish a three-dimensional model of the seabed strata. The preset tolerance parameters are used to ensure that the converted model is close to the actual terrain undulation. The stratum unit stretching formula is as follows:

[0062]

[0063] in, Indicates the new elevation value after stretching; Indicates the original elevation value; Indicates a certain attribute value of the stratum; Represents the stretch scale factor.

[0064] Embodiment 2:

[0065] On the basis of the above-mentioned embodiments, from the business, technology and interaction dimensions, the marine development and utilization supervision system, the marine space resources one-picture comprehensive supervision platform, the marine economic visualization system and the marine enterprise direct reporting system are integrated and iterated, and the intelligent control ocean system is designed based on the space governance digital platform, and the sub-applications and platforms are connected and unified in terms of government user system, data management services and UI interaction design to form a comprehensive and unified marine thematic application. The integration framework is as follows: Figure 4 shown.

[0066] For marine data such as marine space resources, marine ecological monitoring and assessment, marine disaster monitoring and early warning, and marine economic operation monitoring, we carry out intelligent marine data fusion work in terms of data governance, storage, and scheduling, establish spatial database and business database index association, and provide support for the subsequent development of marine military-civilian integration applications. Taking the spatial vector data of islands, coastlines, key defense areas, typical ecosystems, and marine-related enterprise directories as an example, the average loading speed has been significantly improved compared to before fusion. The data record test situation is shown in Table 2:

[0067] Table 2. Data recording test situation

[0068]

[0069] Based on the integrated data base, with the integrated cockpit as the interactive carrier, four major application scenarios of marine resources intelligent management, ecological intelligent judgment, marine disaster intelligent prevention, and marine economic research are formed, as well as integrated applications such as intelligent site selection, intelligent identification, blue carbon map, and marine economic heat map. Marine resources intelligent management integrates the marine development and utilization supervision system and the marine space resources one-picture comprehensive supervision platform, and is built around the management of marine space resources and the service needs of various entities; ecological intelligent judgment integrates offshore buoy observation and monitoring data, and uses carbon sink estimation methods to build around the related needs of ecological monitoring, ecological evaluation, ecological early warning, and blue carbon investigation; marine disaster intelligent prevention integrates coastal disaster-bearing body data and IoT perception monitoring data as well as the overtopping amount calculation method and overtopping risk level classification proposed by seawalls, and is built around risk perception, risk assessment, risk judgment and other related needs; marine economic research integrates the marine economic visualization system and the direct reporting system of marine-related enterprises, and is built around the related needs of marine economic intelligence and marine economic development trend according to the marine economic historical data extrapolation method and the marine economic development trend research method.

[0070] Embodiment 3:

[0071] On the basis of the aforementioned embodiment, the system adds a machine learning module, which extracts features from GIS data, trains the data representation with more semantic information through a random forest model, and predicts the changing trend of the marine environment through an adaptive cross-validation method.

[0072] First, the relevant GIS data needs to be cleaned and preprocessed to ensure data quality. The data normalization formula is as follows:

[0073]

[0074] in, Represents the original data; Indicates the minimum value in the data; Indicates the maximum value in the data; Represents normalized data.

[0075] Extract features from GIS data, perform feature dimensionality reduction through principal component analysis, and consider feature weights:

[0076]

[0077] in, represents the original feature matrix; W represents the principal component loading matrix; represents the feature weight diagonal matrix; Represents the feature matrix after dimensionality reduction.

[0078] The model is trained and validated through the adaptive cross-validation method, and the ratio of the training set and the test set is dynamically adjusted. The formula of the adaptive cross-validation method is as follows:

[0079]

[0080] in, Represents statistical methods used in model selection and hyperparameter tuning; represents the number of folds, which indicates the number of parts the data set is divided into; Represents a model evaluation and selection technique.

[0081] Image classification and target detection are performed through convolutional neural networks, where the attention mechanism formula is as follows:

[0082]

[0083] in, Represents the output of each layer in the network; Represents the feature map extracted by the convolution operation; represents the activation function; Represents the attention mechanism.

[0084] For example, suppose you need to work on a marine environmental monitoring project where the goal is to predict red tide phenomena in a specific sea area.

[0085] Assume that the data is collected as follows:

[0086] Remote sensing image data: including water temperature, chlorophyll concentration, etc. in the sea area; meteorological data: including wind speed, precipitation, etc.; historical red tide occurrence records: including occurrence time, location and duration, etc.

[0087] Extract features:

[0088] Extract features such as water temperature and chlorophyll concentration from remote sensing image data; extract features such as wind speed and precipitation from meteorological data; extract features such as time and location from historical red tide records.

[0089] Feature dimensionality reduction is performed through principal component analysis, assuming that the original feature matrix is a 100×10 matrix containing 100 samples and 10 features; is a 10×5 matrix representing the loadings of the first five principal components; is a diagonal matrix containing the feature weights.

[0090] Assuming that the basic model is a support vector machine, the model is selected and the hyperparameters are adjusted by the adaptive cross-validation method. Assuming that there are three candidate values ​​for hyperparameters, they are and ; The input remote sensing image data is a 32×32 color image, and the convolution layer has 32 filters with a size of 3×3.

[0091] The effect of the machine learning module is shown in Table 3:

[0092] Table 3. Summary of machine learning module effects

[0093]

[0094] As shown in Table 3, the prediction accuracy of the method after optimization by the machine learning module increased by 10%, which shows that the machine learning module can effectively improve the prediction accuracy of the marine digital system for the changing trend of the marine environment.

[0095] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable non-transient storage media containing computer-usable program code.

[0096] The present invention can provide computer program instructions to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce means for implementing the system.

[0097] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a product including an instruction device that implements the functions of the system.

[0098] These computer program instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide steps for implementing the functions of the system.

Claims

1. A GIS-based marine digital fusion system, characterized by: include: Business integration module, which is used to realize that originally independent business modules can provide data support for multiple applications at the same time through data integration and application integration; The application integration specifically includes: based on marine space resource data, combining data to develop intelligent site selection applications; combining marine economic operation monitoring and evaluation data to develop marine economic heat maps and marine-related enterprise spatial distribution applications; combining front-end IoT perception data to develop intelligent identification of illegal sea use behaviors and marine disaster early warning observation applications; combining marine comprehensive management data to develop comprehensive decision-making cockpit applications; The technology integration module is used to achieve technical interoperability between different applications through framework integration and technology stack integration; the technology stack integration specifically includes: encapsulating codes with large differences through secondary development, using WebGL 3D graphics rendering method and OpenLayers 2D GIS rendering method in the front end to meet the front-end rendering applications of various graphics engine platforms; the back end adopts the service registration method to uniformly schedule various services provided by the platform by type to maximize service compatibility; Interaction fusion module, which is used to realize visual interaction design based on business logic through UI fusion and Web component fusion; The marine geological data three-dimensional display module is used to generate a digital elevation model by performing spatial interpolation operations on GIS data, and to establish a three-dimensional model of the seabed strata by vertically stretching the irregular triangulated network converted from raster data; specifically, it includes: Generate point feature vector files through preprocessed data, and perform spatial interpolation operations to generate digital elevation models. The interpolation method formula is as follows: ,in, Represents the value of the forecast point; Represents the value of a known point; Indicates the distance between the predicted point and the known point; Represents the weight assigned to each known point based on data quality or other relevant factors; represents the weight index; The three-dimensional model of the seabed strata is established by vertically stretching the irregular triangulated network converted from raster data. The preset tolerance parameters are used to ensure that the converted model is close to the actual terrain undulation. The stratum unit stretching formula is as follows: ,in, Indicates the new elevation value after stretching; Indicates the original elevation value; Indicates a certain attribute value of the stratum; Represents the stretching scale factor; The system also includes a machine learning module, which extracts features from GIS data, trains the data representation with more semantic information through a random forest model, and predicts the changing trend of the marine environment through an adaptive cross-validation method, specifically including: First, the relevant GIS data needs to be cleaned and preprocessed to ensure data quality. The data normalization formula is as follows: ,in, Represents the original data; Indicates the minimum value in the data; Indicates the maximum value in the data; represents the normalized data; Extract features from GIS data, perform feature dimensionality reduction through principal component analysis, and consider feature weights: ,in, represents the original feature matrix; W represents the principal component loading matrix; represents the feature weight diagonal matrix; Represents the feature matrix after dimensionality reduction; The model is trained and validated through the adaptive cross-validation method, and the ratio of the training set and the test set is dynamically adjusted. The formula of the adaptive cross-validation method is as follows: ,in, Represents statistical methods used in model selection and hyperparameter tuning; represents the number of folds, which indicates the number of parts the data set is divided into; represents a model evaluation and selection technique; Image classification and target detection are performed through convolutional neural networks, where the attention mechanism formula is as follows: ,in, Represents the output of each layer in the network; Represents the feature map extracted by the convolution operation; represents the activation function; Represents the attention mechanism.

2. The GIS-based marine digital fusion system according to claim 1 is characterized by: The data fusion specifically includes: adding space and time as indexes without affecting the original business operation, performing two-dimensional spatial visualization and three-dimensional multi-source data fusion management on the data based on the standard data format according to the requirements of data visualization and storage scheduling, and realizing the separate storage of map data and attribute data by distinguishing databases.

3. The GIS-based marine digital fusion system according to claim 1 is characterized by: The technology fusion module realizes the fusion of the framework based on the front-end and back-end separation method and the proxy forwarding method, and designs the Server components in different applications according to the microservice concept; in the fusion process of the technology stack, the front-end adopts the WebGL three-dimensional graphics rendering method and the OpenLayers two-dimensional GIS rendering method, and the back-end adopts the service registration method.

4. The GIS-based marine digital fusion system according to claim 1 is characterized by: The UI is integrated according to the business needs and interaction logic of the marine system, and the content of the elements in the interactive area is determined according to the application scenario. On the premise that the two- and three-dimensional visualization components in the GIS base area are not overloaded, the integration of Web components is achieved by covering the interactive area on the GIS base area and reusing the same component elements in each area.

5. The GIS-based marine digital fusion system according to claim 1 is characterized by: The system also includes an application development module, which meets the development needs of business applications and integrated applications through the spatial analysis functions in the spatial analysis customization engine, realizes the functions of two-dimensional and three-dimensional data query and spatial analysis by integrating WebGL and OpenLayers front-end rendering methods, and realizes the transmission and loading of various video stream data by integrating Video.js and Flv.js methods.

Images