A fusion method for border and coastal defense big data

Abstract

The present invention provides a method for fusing border and coastal defense big data, belonging to the technical field of data fusion. The method comprises collecting basic data from various data sources, preprocessing the basic data, classifying the preprocessing results into types, and extracting data features from each type of data; performing multi-dimensional analysis on the data features, and creating a multi-dimensional data model by combining the analysis results of multiple dimensions; fusing the preprocessing results based on the multi-dimensional data model to obtain a fusion result, and making a trust judgment on the fusion result; obtaining a comprehensive view of border and coastal defense based on the trust judgment result, effectively solving the problems of different formats, precisions and time spans of multi-source data, improving the compatibility and integration efficiency of data, and providing more comprehensive situation awareness and strategy support for border and coastal defense.

Description

Technical Field

[0001] The present invention relates to the technical field of data fusion, and in particular to a method for fusing border and coastal defense big data. Background Art

[0002] Border and coastal defense monitoring involves data from multiple sources, such as radar data, satellite images, drone monitoring, sensor data, geographic information system data, etc. These data may have different formats, precisions, and time spans. How to efficiently integrate and analyze these heterogeneous data is a huge challenge. Radar data may provide dynamic information but lacks geographic context, while GIS data provides static geographic information but cannot reflect real-time changes.

[0003] Therefore, the present invention provides a method for integrating border and coastal defense big data. Summary of the invention

[0004] The present invention provides a method for integrating border and coastal defense big data. The method collects multi-source heterogeneous data, performs preprocessing and feature extraction, creates a data model in combination with multi-dimensional analysis, and finally performs data fusion and trust evaluation to generate a comprehensive border and coastal defense view. It effectively solves the problems of different formats, precisions and time spans of multi-source data, improves data compatibility and integration efficiency, and provides more comprehensive situational awareness and strategy support for border and coastal defense.

[0005] The present invention provides a method for fusing border and coastal defense big data, comprising:

[0006] Step 1: Collect basic data from various data sources, preprocess the basic data, classify the preprocessing results into types, and extract data features from each type of data;

[0007] Step 2: Perform multi-dimensional analysis on data features and create a multi-dimensional data model by combining the analysis results of multiple dimensions;

[0008] Step 3: Fuse the preprocessing results based on the multidimensional data model to obtain the fusion results, and make a trustworthiness judgment on the fusion results;

[0009] Step 4: Draw a comprehensive picture of border and coastal defense based on the trust judgment results.

[0010] The present invention provides a method for integrating border and coastal defense big data, which collects basic data from various data sources, preprocesses the basic data, divides the preprocessing results into types, and extracts data features from each type of data, including:

[0011] Identify data sources related to border and coastal defense, extract basic data from the data sources, and perform preprocessing;

[0012] The preprocessing results are divided into the first type according to the nature of the data, and the second type according to the application field of the data in border and coastal defense;

[0013] Data features of the preprocessing results are extracted based on the first type and the second type.

[0014] The present invention provides a method for integrating border and coastal defense big data, performs multi-dimensional analysis on data feature transformation, and creates a multi-dimensional data model by combining the analysis results of multiple dimensions, including:

[0015] According to the actual needs of border and coastal defense, the analysis target is clarified, the dimension of the data feature is identified based on the analysis target, the target gain of each analysis target is determined according to the dimension identification result, and then the first feature is determined;

[0016] The first feature is converted into a first vector, and a multi-dimensional analysis is performed on the first vector. The multi-dimensional analysis result is combined with the first feature to obtain a second vector, and a multi-dimensional data model is created based on the second vector.

[0017] The present invention provides a fusion method for border and coastal defense big data, which determines the target gain of each analysis target according to the dimension recognition result, including:

[0018] ; Where F represents the data feature; T represents the variable corresponding to the analysis target; Represents the weighted information gain rate between the data features and the corresponding variables of the analysis target; Indicates the specific value of the i-th dimension corresponding to the variable T corresponding to the analysis target; represents the probability that the variable T corresponding to the analysis target takes the specific value of the i-th dimension; n represents the total number of specific values ​​of the dimension of the variable T corresponding to the analysis target; Indicates the weight factor of the specific value of the i-th dimension of the variable T corresponding to the analysis target; Indicates the weight factor of the data feature F taking the specific value of the jth dimension; represents the j-th dimension specific value corresponding to the data feature F; m represents the total number of dimension specific values ​​of the data feature F; Indicates that the specific value of the dimension of data feature F is Under the condition of The conditional probability of Indicates the probability that the data feature F takes a specific value of the jth dimension;

[0019] A target gain situation is determined according to the weighted information gain rate.

[0020] The present invention provides a fusion method for border and coastal defense big data, which fuses preprocessing results based on a multidimensional data model to obtain a fusion result, and makes a trustworthiness judgment on the fusion result, including:

[0021] Determine the dimensional data viewing requirements based on the border and coastal defense-historical event table, perform data fusion on the preprocessing results according to the dimensional data viewing requirements, merge similar and related data points, and use a multidimensional data model to perform fusion calculation on the merged results and output the fusion results;

[0022] Verifying the fusion result based on the border and coastal defense-historical event table, and at the same time, evaluating the accuracy of the fusion result to obtain a first evaluation result;

[0023] Check the consistency between the information from different data sources in the fusion result to obtain a second evaluation result;

[0024] Based on the border and coastal defense-historical event table, necessary point analysis is performed on the border and coastal defense data, and the completeness of the fusion results is evaluated based on the necessary point analysis results to obtain the third evaluation result;

[0025] The first evaluation result, the second evaluation result and the third evaluation result are combined to obtain a confidence judgment result of the fusion result.

[0026] The present invention provides a fusion method for border and coastal defense big data, which determines the dimensional data viewing requirements according to the border and coastal defense-historical event table, performs data fusion on the preprocessing results according to the dimensional data viewing requirements, and merges similar and related data points, including:

[0027] Analyze the Border and Coast Defense-Historical Events table to determine the key dimensions, and determine the dimensional data viewing requirements based on the key dimensions;

[0028] Performing a first similarity judgment on the preprocessing results based on a preset numerical attribute threshold;

[0029] The preprocessing results are grouped by clustering algorithm, the differences between different groups are analyzed, the classification attributes are obtained according to the difference analysis results, and the second similarity judgment of the classification attributes is performed;

[0030] The first judgment result and the second judgment result are combined to determine similar and related data points in the preprocessing result, and merge them to obtain a merged result.

[0031] The present invention provides a fusion method for border and coastal defense big data, which uses a multidimensional data model to perform fusion calculation on the fusion result and outputs the fusion result, including:

[0032] Use a multi-dimensional data model to perform dimension fusion calculation on the merged results;

[0033] Analyze the correlation between different dimension attributes, determine the dependency in the merged result based on the dimension correlation, and perform dimension attribute fusion calculation based on the dependency;

[0034] A fusion result is determined based on the dimension fusion calculation and the dimension attribute fusion calculation.

[0035] The present invention provides a method for integrating border and coastal defense big data, and obtains a comprehensive view of border and coastal defense based on the trust judgment result, including:

[0036] Filter the reliable data in the fusion results based on the trust judgment results, map the reliable data to the border and coastal defense view framework, and form an overview of the border and coastal defense situation;

[0037] According to the trust judgment results, the weights of the key dimensions are adjusted, and a comprehensive view of border and coastal defense is generated based on an overview of the border and coastal defense situation.

[0038] Compared with the existing technology, the beneficial effects of the present application are as follows: by collecting multi-source heterogeneous data, performing preprocessing and feature extraction, and then combining multi-dimensional analysis to create a data model, and finally performing data fusion and trust evaluation, a comprehensive border and coastal defense view is generated, which effectively solves the problems of different formats, accuracy and time spans of multi-source data, improves data compatibility and integration efficiency, and provides more comprehensive situational awareness and strategic support for border and coastal defense.

[0039] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0040] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0042] Figure 1 It is a flow chart of a method for integrating border and coastal defense big data provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0044] Embodiment 1:

[0045] The embodiment of the present invention provides a method for integrating border and coastal defense big data, such as Figure 1 As shown, including:

[0046] Step 1: Collect basic data from various data sources, preprocess the basic data, classify the preprocessing results into types, and extract data features from each type of data;

[0047] Step 2: Perform multi-dimensional analysis on data features and create a multi-dimensional data model by combining the analysis results of multiple dimensions;

[0048] Step 3: Based on the multidimensional data model, the preprocessing results are integrated to obtain the integrated results, and the trustworthiness of the integrated results is judged;

[0049] Step 4: Draw a comprehensive picture of border and coastal defense based on the trust judgment results.

[0050] In this embodiment, basic data refers to raw or slightly processed data extracted from the above data sources, such as: image and video data: satellite and aerial images, surveillance camera recordings, location data: location coordinates, heading and speed from radar, AIS and ADS-B.

[0051] In this embodiment, the process of type classification is to identify border and coastal defense related data sources, extract and preprocess basic data, then perform dual type classification according to data properties and application fields, and extract data features.

[0052] In this embodiment, data features refer to key information extracted from preprocessed data for subsequent analysis and strategy, such as: features of image data: edges, textures, color distribution, etc., features of location data: speed, direction, historical trajectory, etc., features of text data: keywords, emotional tendencies, themes, etc.

[0053] In this embodiment, the multi-dimensional analysis refers to a further analysis of the first feature vector, taking into account the impact of other relevant dimensions (such as time, weather conditions, etc.) on the first feature.

[0054] In this embodiment, the multidimensional data model is created based on the second vector, which can comprehensively consider the influence of multiple dimensions and provide an in-depth understanding of the analysis target. Input: The input of the multidimensional data model is the second vector, that is, the feature vector that integrates the first feature and the multidimensional analysis result. Output: The output is the prediction or identification result of the analysis target, such as the predicted location and estimated arrival time of the illegal crossing vessel.

[0055] In this embodiment, the trust judgment result is a combination of the first, second, and third evaluation results to obtain the overall trust of the fusion result. For example, trust judgment: after comprehensive evaluation, the trust of the fusion result is 75%, indicating that under the current conditions, the data fusion result has a certain degree of reliability, but still needs to be improved.

[0056] In this embodiment, the fusion result is a comprehensive information output obtained by fusing, merging and calculating the preprocessed data. It integrates data from different data sources and reflects key events and dynamic situations in border and coastal defense monitoring. For example, if the fusion result shows that multiple illegal intrusion incidents have occurred in a specific sea area in the past year, and the time, location and participant information of these incidents are integrated together, strategists can better understand the patterns and trends of these incidents and take corresponding preventive measures.

[0057] In this embodiment, the comprehensive view of border and coastal defense is a more detailed and comprehensive monitoring view generated based on an overview of the border and coastal defense situation and the adjusted weights of key dimensions, including multi-dimensional analysis and forecasting information. For example, in the comprehensive view of border and coastal defense, in addition to displaying current monitoring data, it may also include historical data trends, potential threat analysis, and future forecasts to help strategists formulate response strategies.

[0058] The working principle and beneficial effects of the above technical solution are: by collecting multi-source heterogeneous data, preprocessing and feature extraction, and then combining multi-dimensional analysis to create a data model, and finally performing data fusion and trust evaluation, a comprehensive view of border and coastal defense is generated, which effectively solves the problems of different formats, accuracy and time spans of multi-source data, improves data compatibility and integration efficiency, and provides more comprehensive situational awareness and strategic support for border and coastal defense.

[0059] Embodiment 2:

[0060] The embodiment of the present invention provides a method for integrating border and coastal defense big data, which collects basic data from various data sources, preprocesses the basic data, divides the preprocessing results into types, and extracts data features from each type of data, including:

[0061] Identify data sources related to border and coastal defense, extract basic data from the data sources, and perform preprocessing;

[0062] The preprocessing results are divided into the first type according to the nature of the data, and the second type according to the application field of the data in border and coastal defense;

[0063] Data features of the preprocessing results are extracted based on the first type and the second type.

[0064] In this embodiment, relevant data sources include satellite and aerial images, radar system data, sensor network data, open source intelligence, ship and aircraft automatic identification systems, etc.

[0065] In this embodiment, the first type of classification is based on the nature of the data, which generally includes structured data: such as tabular data in the database, location data, time series data, etc., and unstructured data: such as text, image and video data; the second type of classification is based on the application areas of the data in border and coastal defense, including: reconnaissance and surveillance: using satellite images, radar data and surveillance videos to conduct real-time monitoring of borders and sea areas, threat identification: using AI to analyze images, videos and other sensor data to identify potential illegal border crossings, smuggling and other threats, environmental monitoring: using sensor data to monitor the environmental conditions of borders and sea areas, such as the impact of climate change on border security, information collection: understanding political, economic and social events related to borders and sea areas by analyzing OSINT data.

[0066] The working principle and beneficial effects of the above technical solution are: by identifying border and coastal defense related data sources, extracting and preprocessing basic data, and then dividing the data into dual types according to the nature of the data and the application field, and extracting data features to support comprehensive analysis, improve the consistency and efficiency of data processing, and enhance the applicability of data in different border and coastal defense application fields.

[0067] Embodiment 3:

[0068] The embodiment of the present invention provides a method for integrating border and coastal defense big data, performs multi-dimensional analysis on data feature transformation, and creates a multi-dimensional data model by combining the analysis results of multiple dimensions, including:

[0069] According to the actual needs of border and coastal defense, the analysis target is clarified, the dimension of the data feature is identified based on the analysis target, the target gain of each analysis target is determined according to the dimension identification result, and then the first feature is determined;

[0070] The first feature is converted into a first vector, and a multi-dimensional analysis is performed on the first vector. The multi-dimensional analysis result is combined with the first feature to obtain a second vector, and a multi-dimensional data model is created based on the second vector.

[0071] In this embodiment, the actual needs of border and coastal defense monitoring include: improving the security of borders and sea areas, protecting key infrastructure, monitoring environmental changes, etc.

[0072] In this embodiment, the analysis goal is determined based on actual needs, and the specific goals may be: identifying and tracking illegal crossing vessels or aircraft, monitoring illegal activities in a specific area, and predicting potential security threats or environmental disasters.

[0073] In this embodiment, dimension identification is based on analysis goals, identifying key information dimensions from the data that help achieve these goals, such as location, time, speed, direction, etc.

[0074] In this embodiment, the first feature is determined based on the dimension recognition result and is a feature that has a key impact on the analysis target. For example, if the analysis target is to track illegal crossing ships, then the first feature may be an abnormal pattern of the speed and course of the ship.

[0075] The working principle and beneficial effects of the above technical solution are: through comprehensive analysis of data from multiple sources, such as radar, satellite images, etc., the border and coastal defense monitoring system can identify and integrate data of different formats and precisions, extract key features and convert them into vectors to create a multidimensional data model that can reflect real-time changes and static geographic information, providing a comprehensive perspective, thereby improving the efficiency and accuracy of response to border and maritime security threats.

[0076] Embodiment 4:

[0077] The embodiment of the present invention provides a method for fusing border and coastal defense big data, which determines the target gain of each analysis target according to the dimension recognition result, including:

[0078] ; Where F represents the data feature; T represents the variable corresponding to the analysis target; Represents the weighted information gain rate between the data features and the corresponding variables of the analysis target; Indicates the specific value of the i-th dimension corresponding to the variable T corresponding to the analysis target; represents the probability that the variable T corresponding to the analysis target takes the specific value of the i-th dimension; n represents the total number of specific values ​​of the dimension of the variable T corresponding to the analysis target; Indicates the weight factor of the specific value of the i-th dimension of the variable T corresponding to the analysis target; Indicates the weight factor of the data feature F taking the specific value of the jth dimension; represents the j-th dimension specific value corresponding to the data feature F; m represents the total number of dimension specific values ​​of the data feature F; Indicates that the specific value of the dimension of data feature F is Under the condition of The conditional probability of Indicates the probability that the data feature F takes a specific value of the jth dimension;

[0079] A target gain situation is determined according to the weighted information gain rate.

[0080] In this embodiment, the dimension specific values ​​of the data features refer to specific information extracted from various data sources, which are used to describe the characteristics of the monitored objects. The dimension specific values ​​include: location information: such as longitude and latitude coordinates, used to identify the geographical location of ships, aircraft or other monitored objects, time information: such as timestamp, identifying the specific time when an observation or event occurs, speed and direction: for moving objects, their speed and direction of movement are key dynamic information, type information: such as ship type, aircraft model, etc., used to describe the category of monitored objects, behavior patterns: specific behaviors or activity patterns, such as ships lingering in a specific area, which may indicate illegal fishing or smuggling activities, sensor data: data from sensors such as radar, sonar, thermal imagers, etc., providing additional physical characteristics about the monitored objects.

[0081] In this embodiment, the specific value of the dimension of the variable corresponding to the analysis target refers to the specific indicator or state related to the monitoring target, which is the result that needs to be predicted or detected during the analysis process. For example: Threat level: security threat level determined based on analysis, such as high, medium, and low; Activity type: specific activity type identified in the monitoring area, such as illegal crossing of the border, etc.; Abnormal state: whether the monitored object is in an abnormal state, such as normal or abnormal; Potential target: whether the monitored object is identified as a potential target or threat of concern to border and coastal defense; Event prediction: prediction of specific events that may occur in the future, such as the time and location of crossing the border.

[0082] The working principle and beneficial effects of the above technical solution are: using the weighted information gain rate to evaluate the contribution of different data features to the analysis target, giving priority to data dimensions that are highly relevant to the target, more accurately identifying and responding to security threats in border and coastal defense, reducing false alarms and missed alarms, adapting to monitoring needs in various complex environments, and enhancing the flexibility and robustness of the system.

[0083] Embodiment 5:

[0084] The embodiment of the present invention provides a method for fusing border and coastal defense big data, which fuses preprocessing results based on a multidimensional data model to obtain a fusion result, and makes a trustworthiness judgment on the fusion result, including:

[0085] Determine the dimensional data viewing requirements based on the border and coastal defense-historical event table, perform data fusion on the preprocessing results according to the dimensional data viewing requirements, merge similar and related data points, and use a multidimensional data model to perform fusion calculation on the merged results and output the fusion results;

[0086] Verifying the fusion result based on the border and coastal defense-historical event table, and at the same time, evaluating the accuracy of the fusion result to obtain a first evaluation result;

[0087] Check the consistency between the information from different data sources in the fusion result to obtain a second evaluation result;

[0088] Based on the border and coastal defense-historical event table, necessary point analysis is performed on the border and coastal defense data, and the completeness of the fusion results is evaluated based on the necessary point analysis results to obtain the third evaluation result;

[0089] The first evaluation result, the second evaluation result and the third evaluation result are combined to obtain a confidence judgment result of the fusion result.

[0090] In this embodiment, the dimensional data viewing requirement refers to the key data dimensions required for specific targets or events in border and coastal defense monitoring, including time, location, event type, participants, etc. For example, the time dimension: the need to view event data within the past year, the location dimension: events in specific sea areas or border areas, event types: such as illegal intrusion, maritime rescue, environmental monitoring, etc.

[0091] In this embodiment, the border and coastal defense-historical event table is a data set that records historical events and contains information in multiple dimensions.

[0092] In this embodiment, the first evaluation result is based on the accuracy evaluation of the fusion result to determine whether the fused data reflects the actual situation. For example, the evaluation result: the accuracy of the fusion result is 85%, indicating that 85% of the event predictions are consistent with the actual events.

[0093] In this embodiment, the second evaluation result is to check the consistency between different data sources and evaluate the reliability of the data. For example, the evaluation result: the consistency between different data sources (such as radar data and satellite images) is 90%, indicating that most of the data are consistent.

[0094] In this embodiment, the third evaluation result is to evaluate the completeness of the fusion result through necessary point analysis to ensure that all key data points have been considered. For example, the evaluation result: the fusion result contains 80% of the necessary data points, indicating that 20% of the key data are not covered.

[0095] In this embodiment, the necessary point analysis is to analyze the border and coastal defense data and identify key data points to ensure the comprehensiveness and effectiveness of the fusion results, such as key data points: monitoring records in a specific area, the frequency of historical events, the identity information of participants, etc.

[0096] In this embodiment, the necessary point analysis result is to evaluate the integrity and effectiveness of data fusion based on the results of the necessary point analysis. For example, the analysis result shows that monitoring data of certain high-risk areas are missing, and it is recommended to supplement relevant data to improve the integrity of the fusion result.

[0097] In this embodiment, the merged results are obtained by analyzing the border and coastal defense-historical event table, determining the key dimensions and proposing the dimensional data viewing requirements, making a first similarity judgment on the preprocessing results based on a preset numerical attribute threshold, applying a clustering algorithm to group the data, analyzing the differences between the groups to obtain the classification attributes, and making a second similarity judgment. The results of the two are combined to identify and merge similar and related data points to form a merged result.

[0098] The working principle and beneficial effects of the above technical solution are: by analyzing the dimensional data viewing requirements of the border and coastal defense-historical event table, pre-processing results for data fusion, merging similar data points, and using a multidimensional data model for fusion calculations, by verifying the accuracy, consistency and completeness of the fusion results, a comprehensive evaluation is made to draw a trust judgment on the fusion results, reducing redundant data, and improving data processing efficiency. It can flexibly adjust data processing and analysis methods according to the characteristics and needs of different data sources to enhance the adaptability of the system.

[0099] Embodiment 6:

[0100] The embodiment of the present invention provides a method for fusing border and coastal defense big data, which determines the dimensional data viewing requirement according to the border and coastal defense-historical event table, performs data fusion on the preprocessing results according to the dimensional data viewing requirement, and merges similar and related data points, including:

[0101] Analyze the Border and Coast Defense-Historical Events table to determine the key dimensions, and determine the dimensional data viewing requirements based on the key dimensions;

[0102] Performing a first similarity judgment on the preprocessing results based on a preset numerical attribute threshold;

[0103] The preprocessing results are grouped by clustering algorithm, the differences between different groups are analyzed, the classification attributes are obtained according to the difference analysis results, and the second similarity judgment of the classification attributes is performed;

[0104] The first judgment result and the second judgment result are combined to determine similar and related data points in the preprocessing result, and merge them to obtain a merged result.

[0105] In this embodiment, the key dimensions are the main attributes or characteristics used to analyze border and coastal defense incidents, which usually include time, location, event type, participants, impact level, etc. For example, time (date of the incident), location (geographic coordinates of the incident), event type (such as invasion, smuggling), and impact level (high, medium, and low).

[0106] In this embodiment, the preset numerical attribute threshold is a numerical standard used to determine whether data points have similarities, and is usually set based on historical data or expert experience. For example, a monitoring indicator (such as target speed) exceeding 10 kilometers per hour is considered suspicious, or the frequency of abnormal activities in a certain area exceeds 5 times per hour.

[0107] In this embodiment, the first judgment of similarity is to preliminarily screen the data points based on a preset threshold to determine whether the similarity conditions are met, for example, the time difference between the two events does not exceed 1 hour, and the locations are within the same geographical area.

[0108] In this embodiment, the classification attribute is an attribute obtained by grouping data points based on the cluster analysis results, reflecting their common characteristics. For example, events are divided into three categories: "high-risk events", "medium-risk events" and "low-risk events".

[0109] In this embodiment, the second similarity judgment is to further analyze the classification attributes to judge the similarity between data points in the same category. For example, in the "high-risk event" category, the specific characteristics of the event are further compared, such as the type of participants, the means used, etc., to determine whether there is similarity.

[0110] In this embodiment, the merged result is a comprehensive analysis based on the first judgment and the second judgment, identifying similar related data points and merging them into one data set. For example, multiple similar intrusion events are merged into one event record, marked as "intrusion event A", and attached with comprehensive information such as time, location and degree of impact.

[0111] The working principle and beneficial effects of the above technical solution are: by analyzing the border and coastal defense-historical event table, determining the key dimensions and proposing the dimensional data viewing requirements, making a first similarity judgment on the preprocessing results based on the preset numerical attribute threshold, applying the clustering algorithm to group the data, analyzing the differences between the groups to obtain the classification attributes, and making a second similarity judgment, combining the results of the two, identifying and merging similar and related data points to form a merged result, thereby improving the availability and consistency of the data.

[0112] Embodiment 7:

[0113] The embodiment of the present invention provides a fusion method for border and coastal defense big data, which uses a multidimensional data model to perform fusion calculation on the fusion result and outputs the fusion result, including:

[0114] Use a multi-dimensional data model to perform dimension fusion calculation on the merged results;

[0115] Analyze the correlation between different dimension attributes, determine the dependency in the merged result based on the dimension correlation, and perform dimension attribute fusion calculation based on the dependency;

[0116] A fusion result is determined based on the dimension fusion calculation and the dimension attribute fusion calculation.

[0117] In this embodiment, dimensional fusion calculation is to integrate data of multiple dimensions for comprehensive analysis, which usually involves weighting, merging or transforming data of different dimensions. For example, time (such as the time when an event occurs) and place (such as the geographic coordinates where the event occurs) are combined to form a new composite dimension, representing "the time and place when the event occurs."

[0118] In this embodiment, dimensional correlation refers to the relationship and degree of influence between different dimensional attributes, which is usually analyzed through statistical methods (such as correlation coefficient). For example, it is found that there is a correlation between "event type" and "degree of influence". For example, intrusion incidents usually have a high degree of influence, while smuggling incidents may have a smaller impact.

[0119] In this embodiment, dependency refers to the dependence of the value of a certain dimension attribute on the value of another dimension attribute, which is usually used to identify the causal relationship or logical relationship between data. For example, if the "event type" is "intrusion", the "participant type" may depend on a specific group, which indicates the dependency between the event type and the participant type.

[0120] In this embodiment, the dimensional attribute fusion calculation is based on the identified dependencies, and the related dimensional attributes are merged or adjusted to generate new data attributes. For example, based on the dependency relationship between "event type" and "impact level", the participant type of "intrusion event" and its impact level are fused to form a new attribute "high-risk intrusion participant" for subsequent analysis.

[0121] The working principle and beneficial effects of the above technical solution are: use a multi-dimensional data model to perform dimensional fusion calculation on the merged results, analyze the correlation between different dimensional attributes, determine the dependencies in the merged results, perform dimensional attribute fusion calculation based on these dependencies, and finally obtain the fusion results, so as to have a more comprehensive understanding of the data, effectively integrate heterogeneous data, overcome the challenges of format and time span, and realize collaborative analysis of data.

[0122] Embodiment 8:

[0123] The embodiment of the present invention provides a method for integrating border and coastal defense big data, which obtains a comprehensive view of border and coastal defense based on the trust judgment result, including:

[0124] Filter the reliable data in the fusion results based on the trust judgment results, map the reliable data into the border and coastal defense view framework, and form an overview of the border and coastal defense situation;

[0125] According to the trust judgment results, the weights of the key dimensions are adjusted, and a comprehensive view of border and coastal defense is generated based on an overview of the border and coastal defense situation.

[0126] In this embodiment, reliable data is data that meets certain accuracy and credibility standards after trust evaluation. For example, radar data has been verified multiple times and shows that there are suspicious ships in a certain area, and it is consistent with satellite images. These data are considered reliable data.

[0127] In this embodiment, the border and coastal defense situation overview is to map reliable data into a visualization framework to form an overall view of the border and coastal defense status, which helps to quickly understand the current situation. For example, all monitoring activities in the current sea area are displayed on the map, including suspicious ships, the location of monitoring equipment and its status and other information.

[0128] In this embodiment, the influence weights of various key dimensions (such as time, location, event type, etc.) are dynamically adjusted according to the trust judgment results to better reflect the actual situation. For example, if the time dimension of an event is judged to be highly trusted, while the location dimension is less trusted, then during the comprehensive analysis, the weight of the time dimension is increased and the weight of the location dimension is reduced.

[0129] The working principle and beneficial effects of the above technical solution are: after reliable data is screened out through trust judgment, it is mapped to the border and coastal defense view framework to form an overview of the border and coastal defense situation, the weights of key dimensions are adjusted according to the trust, and a comprehensive view is generated based on the overview to achieve dynamic monitoring and comprehensive analysis of the border and coastal defense situation, help quickly identify potential threats and risks, optimize resource allocation, and improve monitoring efficiency.

[0130] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for integrating border and coastal defense big data, characterized in that: include: Step 1: Collect basic data from various data sources, preprocess the basic data, classify the preprocessing results into types, and extract data features from each type of data; Step 2: Perform multi-dimensional analysis on data features and create a multi-dimensional data model by combining the analysis results of multiple dimensions; Step 3: Based on the multidimensional data model, the preprocessing results are integrated to obtain the integrated results, and the trustworthiness of the integrated results is judged; Step 4: Draw a comprehensive picture of border and coastal defense based on the trust judgment results; Among them, a multi-dimensional analysis is performed on the data feature transformation, and a multi-dimensional data model is created by combining the analysis results of multiple dimensions, including: According to the actual needs of border and coastal defense, the analysis target is clarified, the dimension of the data feature is identified based on the analysis target, the target gain of each analysis target is determined according to the dimension identification result, and then the first feature is determined; Convert the first feature into a first vector, perform a multi-dimensional analysis on the first vector, combine the multi-dimensional analysis result with the first feature to obtain a second vector, and create a multi-dimensional data model based on the second vector; Among them, the target gain of each analysis target is determined according to the dimension recognition result, including: ; Where F represents the data feature; T represents the variable corresponding to the analysis target; Represents the weighted information gain rate between the data features and the corresponding variables of the analysis target; Indicates the specific value of the i-th dimension corresponding to the variable T corresponding to the analysis target; represents the probability that the variable T corresponding to the analysis target takes the specific value of the i-th dimension; n represents the total number of specific values ​​of the dimension of the variable T corresponding to the analysis target; Indicates the weight factor of the specific value of the i-th dimension of the variable T corresponding to the analysis target; Indicates the weight factor of the data feature F taking the specific value of the jth dimension; represents the j-th dimension specific value corresponding to the data feature F; m represents the total number of dimension specific values ​​of the data feature F; Indicates that the specific value of the dimension of data feature F is Under the condition of The conditional probability of Indicates the probability that the data feature F takes a specific value of the jth dimension; Determining a target gain situation according to the weighted information gain rate; The dimension specific values ​​of data features refer to the specific information extracted from various data sources, which are used to describe the characteristics of the monitored objects. The dimension specific values ​​include: location information: used to identify the geographical location of ships, aircraft or other monitored objects, time information: identifies the specific time when an observation or event occurs, speed and direction: for moving objects, their speed and direction of movement are key dynamic information, type information: used to describe the category of the monitored object, behavior pattern: specific behavior or activity pattern, sensor data: provides additional physical characteristics about the monitored object; The specific values ​​of the dimensions of the variables corresponding to the analysis targets refer to the specific indicators or states related to the monitoring targets, which are the results that need to be predicted during the analysis process, including: threat level, activity type, abnormal state, potential targets and event prediction.

2. The method for fusing border and coastal defense big data according to claim 1 is characterized in that: Collect basic data from various data sources, preprocess the basic data, classify the preprocessing results into types, and extract data features from each type of data, including: Identify data sources related to border and coastal defense, extract basic data from the data sources, and perform preprocessing; The preprocessing results are divided into the first type according to the nature of the data, and the second type according to the application field of the data in border and coastal defense; Data features of the preprocessing results are extracted based on the first type and the second type.

3. The method for fusing border and coastal defense big data according to claim 1 is characterized in that: Based on the multidimensional data model, the preprocessing results are integrated to obtain the integrated results, and the trustworthiness of the integrated results is judged, including: Determine the dimensional data viewing requirements based on the border and coastal defense-historical event table, perform data fusion on the preprocessing results according to the dimensional data viewing requirements, merge similar and related data points, and use a multidimensional data model to perform fusion calculation on the merged results and output the fusion results; Verifying the fusion result based on the border and coastal defense-historical event table, and at the same time, evaluating the accuracy of the fusion result to obtain a first evaluation result; Check the consistency between the information from different data sources in the fusion result to obtain a second evaluation result; Based on the border and coastal defense-historical event table, necessary point analysis is performed on the border and coastal defense data, and the completeness of the fusion results is evaluated based on the necessary point analysis results to obtain the third evaluation result; The first evaluation result, the second evaluation result and the third evaluation result are combined to obtain a confidence judgment result of the fusion result.

4. The method for fusing border and coastal defense big data according to claim 3 is characterized in that: Determine the dimensional data viewing requirements based on the Border and Coast Defense-Historical Events Table, perform data fusion on the preprocessing results according to the dimensional data viewing requirements, and merge similar and related data points, including: Analyze the Border and Coast Defense-Historical Events table to determine the key dimensions, and determine the dimensional data viewing requirements based on the key dimensions; Performing a first similarity judgment on the preprocessing results based on a preset numerical attribute threshold; The preprocessing results are grouped by clustering algorithm, the differences between different groups are analyzed, the classification attributes are obtained according to the difference analysis results, and the second similarity judgment of the classification attributes is performed; The first judgment result and the second judgment result are combined to determine similar and related data points in the preprocessing result, and merge them to obtain a merged result.

5. The method for fusing border and coastal defense big data according to claim 3 is characterized in that: Use the multidimensional data model to perform fusion calculation on the merged results and output the fusion results, including: Use a multi-dimensional data model to perform dimension fusion calculation on the merged results; Analyze the correlation between different dimension attributes, determine the dependency in the merged result based on the dimension correlation, and perform dimension attribute fusion calculation based on the dependency; A fusion result is determined based on the dimension fusion calculation and the dimension attribute fusion calculation.

6. The method for fusing border and coastal defense big data according to claim 4 is characterized in that: Based on the trustworthiness judgment results, a comprehensive view of border and coastal defense is obtained, including: Filter the reliable data in the fusion results based on the trust judgment results, map the reliable data to the border and coastal defense view framework, and form an overview of the border and coastal defense situation; According to the trust judgment results, the weights of the key dimensions are adjusted, and a comprehensive view of border and coastal defense is generated based on an overview of the border and coastal defense situation.

Images