Methods, systems, and readable storage media for identifying ship anomalies

Abstract

This invention provides a method, system, and readable storage medium for identifying abnormal vessels, relating to the field of vessel identification technology. The method includes: determining an abnormal behavior cluster; performing anomaly analysis on the behavior of vessels within the cluster to generate abnormal behavior events for each vessel; confirming whether a vessel is abnormal based on the abnormal behavior events; generating an abnormal vessel alarm record and / or issuing an alarm for the confirmed abnormal vessel. The method for identifying abnormal vessels in this application can pre-determine abnormal behavior clusters and then analyze the behavior of vessels within these clusters, thereby intelligently identifying abnormal vessels. This avoids the problem of strong subjectivity in existing methods of determining abnormal vessels and improves the accuracy of abnormal vessel identification.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ship identification, in particular to a ship abnormality identification method, system and readable storage medium. BACKGROUND

[0002] A large number of ships are driven by interests to carry out transportation operations in violation of relevant regulations, which not only has great safety hazards, but also relates to the safety of people's lives and property, so the management task of the ships carrying out transportation operations in violation of relevant regulations is imminent.

[0003] In order to avoid supervision and management, the ships carrying out transportation operations in violation of relevant regulations often choose to avoid law enforcement time of relevant departments and often choose to "hide by day and go out by night".

[0004] The ships participating in transportation operations in violation of relevant regulations basically sail away from the shore and do not berth at regular wharfs and berths, pass through the jurisdictional sea area without staying, so it is difficult to find their behavior of carrying out transportation operations in violation of relevant regulations. In order to escape the supervision of the relevant departments on the sea, many ships even disguise as other ships to carry out operations, and it is difficult to detect through simply observing the signal information played by the ships. For the situation of the ships carrying out transportation operations in violation of relevant regulations on the sea, the traditional on-site correction method of relevant behavior by the law enforcement units on the sea has been far from meeting the needs of the development of modern maritime traffic management, so the non-site processing method based on modern scientific and technological means has emerged as the times require, therefore, how to fully use the modern scientific and technological means, constantly improve the dynamic sensing facilities on the sea and timely grasp the ship dynamics has become a problem to be solved at present.

[0005] However, the existing identification method of the ships carrying out transportation operations in violation of relevant regulations has the following problems: 1. strong subjectivity, excessive dependence on expert experience, the weight and threshold of each field need to be set artificially, and the ship behavior is mined by relying on the experience to be set subjectively, rather than by objective data such as AIS trajectory data, due to the inertia or thinking set of human thinking, some associated relationships or indirect associated relationships that are difficult to find are often ignored. 2. poor universality, the model is not flexible, for each scene and case type, independent modeling analysis is needed, the model has no good reusability and scalability, once the environment or condition changes, the cost of new model parameter adjustment or retraining is high.

[0006] Therefore, how to propose a ship abnormality identification method to solve the problems of strong subjectivity, poor universality and inflexible model in the prior art has become a problem to be solved at present. SUMMARY

[0007] To solve the above technical problems, the purpose of the present application is to provide a ship abnormality identification method with strong universality and flexible model.

[0008] Therefore, the first aspect of the present application provides an abnormality identification method of a ship.

[0009] The second aspect of the present application also provides an abnormality identification system of a ship.

[0010] The third aspect of the present application also provides an abnormality identification system of a ship.

[0011] The fourth aspect of the present application also provides a readable storage medium.

[0012] Therefore, the first aspect of the present application provides an abnormality identification method of a ship.

[0013] According to the abnormality identification method of the ship provided by the present application, the abnormality behavior aggregation area includes the sea area prone to abnormal behavior and the sea area with dense routes obtained by analyzing the ships with abnormal behavior in previous years as the abnormal behavior aggregation area, and the abnormal behavior aggregation area has a higher possibility of abnormal behavior than other sea areas. After determining the abnormal behavior aggregation area, the ships passing through the abnormal behavior aggregation area are analyzed for abnormal behavior, including but not limited to signal abnormality and trajectory abnormality, etc. In this way, the abnormal behavior events of each ship can be determined. Since signal abnormality may be caused by equipment failure or channel congestion, after determining the abnormal behavior events of each ship, it is necessary to further confirm whether the ship is an abnormal ship according to the abnormal behavior events, that is, whether abnormal behavior occurs on the ship. Specifically, if the trajectories of two ships are too close in the same time range or the first ship information becomes the second ship information when entering the jurisdiction and the second ship information becomes the first ship information after leaving the jurisdiction, and the first ship information and the second ship information have continuity in the spatial trajectory, it is considered that the ship with abnormal behavior is an abnormal ship. After determining the abnormal ship, the abnormal ship alarm record is generated and / or the confirmed abnormal ship is alarmed, so as to realize early warning of the abnormal ship or evidence collection of the behavior of the abnormal ship and generation of the alarm record. Since all ships passing through the abnormal behavior aggregation area are detected in the present application, and the big data analysis method is used, the problem that the determination method of the abnormal ship in the prior art is subjective is avoided, and the identification accuracy of the abnormal ship is improved. At the same time, this kind of way model is simple, and can update the data in real time by changing the abnormal behavior aggregation area, and then investigate the new abnormal behavior aggregation area, which is very flexible.

[0014] In addition, the ship anomaly identification method provided by the above technical solution can further have the following additional technical features:

[0015] In the above technical solution, the step of determining the abnormal behavior aggregation area includes: determining historical trajectory information of ships meeting the preset characteristics within a preset time period; and analyzing the historical trajectory information of the ships meeting the preset characteristics to determine the abnormal behavior aggregation area.

[0016] In this technical solution, abnormal behavior on the sea often has a certain timeliness, such as periodic replacement of ships, so that the designated ships only participate in abnormal behavior within a period of time. However, in the initial period of use, there is less data to refer to, so it is often overlooked by law enforcement units. In addition, the transaction method and transaction location between abnormal ships will also change, so data that is too long is of no benefit to analyzing the abnormal behavior aggregation area, but rather it will expand the detection range and lead to ships that violate relevant regulations having an opportunity to take advantage of it. Therefore, it is necessary to limit the historical trajectory information of the preset time period, and abnormal ships often have certain commonalities, such as the same starting point or the same organization or country. Although ships with these characteristics may not all have abnormal behavior, the likelihood of abnormal behavior will increase, and the historical trajectory information of these ships becomes the key to determining whether a ship has abnormal behavior. Without knowing which ships will have abnormal behavior, in order to reduce the search range, the ships prone to abnormal behavior, such as ships meeting the preset characteristics, can be focused on, and the historical trajectory information of the ships prone to abnormal behavior, such as ships meeting the preset characteristics, can be analyzed. In this way, the abnormal behavior aggregation area can be determined. Then, the ships passing through the abnormal behavior aggregation area are further investigated, thereby improving the investigation of abnormal ships.

[0017] In the above technical solution, the ships meeting the preset characteristics within the preset time period include ships with abnormal characteristics within the preset time period, or ships from a preset nationality within the preset time period and ships from a preset departure location within the preset time period; and / or the preset time period is one year.

[0018] In the technical solution, since the judgment of the abnormal behavior gathering area is mainly the analysis of the historical navigation data, the selection of the historical navigation data becomes the standard of the actual range of the abnormal behavior gathering area. An excessively large abnormal behavior gathering area will lead to an increased workload, thus providing an opportunity for abnormal ships, and an excessively small abnormal behavior gathering area will lead to a large number of escaped fish, and a large number of abnormal ships cannot be found. Therefore, the range of the abnormal behavior gathering area is the key to the implementation of the application. In the application, the historical route information of the ships with abnormal characteristics in the preset time period, or the ships from the preset nationality or the preset departure place, or the ships of the preset organization in the preset time period is integrated, so as to early warn the actions of these abnormal ships. Since the predetermined nationality or departure place is the object of key control, the abnormal behavior of the ships with the predetermined characteristics is more likely to occur. The investigation of the ships with the predetermined characteristics can effectively improve the efficiency of the early warning. At the same time, new abnormal behavior routes and ships can be detected, and therefore all ships in the preset time period can be analyzed, so as to realize more comprehensive detection. Since the psychology of the personnel who violate the relevant regulations is considered, the routes and ships that have been investigated as the key are not used, and therefore the ships and routes with a long time may mislead the existing situation, thus providing an opportunity for the personnel who violate the relevant regulations. Therefore, the optimal solution is to set the preset time to one year. One year is not too short for relatively concealed abnormal behavior, and it is not too long for abnormal behavior of frequently changing ships or routes.

[0019] At the same time, the original data of the ship should be the global navigation trajectory data in one year. If only the route data of the ship near the sea area of China is considered, the determination of the abnormal behavior gathering area has certain limitations, and the abnormal behavior is global. No matter where the abnormal behavior gathering area is located, the abnormal ship passing through the territorial sea of China needs to be checked by the law enforcement departments of China. In order to prevent foreign abnormal ships from entering the territorial sea of China to perform abnormal behavior, the historical trajectory information used should be global navigation trajectory data.

[0020] Further, the step of analyzing the historical trajectory information of the ship satisfying the preset characteristics to determine the abnormal behavior gathering area includes: analyzing the historical trajectory information based on the improved DBSCAN algorithm (Density-Based Spatial Clustering of Applications with Noise, a representative density-based clustering algorithm) to determine the abnormal behavior gathering area.

[0021] In the above technical solution, the steps for determining the historical trajectory information of ships that meet the preset characteristics within a preset time period include: acquiring the original data of ships that meet the preset characteristics within the preset time period; processing the original data to remove abnormal information; connecting the original data after removing abnormal information according to time to generate the navigation trajectory of ships that meet the preset characteristics; correcting the navigation trajectory to obtain the historical trajectory information of ships that meet the preset characteristics within the preset time period; the original data includes the maritime mobile communication service identification code, latitude and longitude, ship speed, and heading.

[0022] In this technical solution, the steps for determining the historical trajectory information of ships that meet the preset characteristics within a preset time period include: acquiring the original data of ships that meet the preset characteristics within the preset time period; processing the original data to remove abnormal information; connecting the original data after removing abnormal information according to time to generate the navigation trajectory line of the ships that meet the preset characteristics; correcting the navigation trajectory to obtain the historical trajectory information of ships that meet the preset characteristics within the preset time period; the original data includes the maritime mobile communication service identification code, latitude and longitude, ship speed, and heading. In actual use, the raw data received by this application may be tampered with, or contain abnormal information due to equipment failure and channel congestion, such as latitude and longitude significantly exceeding 180°, excessive ship speed, or missing key records. Raw data with erroneous information cannot be used as a standard for determining abnormal behavior clusters. Therefore, it is necessary to clean the raw data to restore the true and reasonable raw data as much as possible. After cleaning, the raw data needs to be connected according to time to generate the preset ship navigation trajectory line. Since it is not possible to remove all unreasonable data during data cleaning, non-smooth and unreasonable points will be processed when connecting the route trajectory line to obtain the preset ship historical trajectory information, thereby improving the accuracy of historical trajectory information and improving the reliability of abnormal behavior clusters.

[0023] Furthermore, the original data for ships is global navigation trajectory data within one year. If only the route data of ships near my country's territorial waters is considered, the determination of abnormal behavior clusters will have certain limitations. However, abnormal behavior is global. Regardless of where the abnormal behavior cluster is located, abnormal ships must be inspected by my country's law enforcement agencies when passing through my country's territorial waters. In order to prevent foreign abnormal ships from entering my country's territorial waters and engaging in abnormal behavior, the historical trajectory information used should be global navigation trajectory data.

[0024] The raw data includes one or more of the following: maritime mobile communication service identification code, latitude and longitude, ship speed, and heading. The raw data can provide evidence for each other to prove the rationality of the data. For example, if the ship speed and heading are determined, under non-extreme weather conditions, the latitude and longitude, ship speed, and heading in the next raw data will have a certain range, thus making the raw data objective and improving the accuracy of historical trajectory information.

[0025] In the above technical solution, the step of performing anomaly analysis on the behavior of at least one vessel within the abnormal behavior cluster area to generate anomaly behavior events for each vessel includes: performing anomaly analysis based on the pre-arrival behavior, dwelling behavior, and departure behavior of each vessel to generate anomaly behavior events for each vessel; the step of performing anomaly analysis on the behavior of vessels within the abnormal behavior cluster area includes: determining whether the signals during the pre-arrival, dwelling, and departure processes of the vessel are abnormal, wherein signal abnormality includes at least signal disappearance and recurrence; determining whether the trajectory during the pre-arrival, dwelling, and departure processes of the vessel is abnormal, wherein trajectory abnormality includes at least one of the following: outlier points appearing in the vessel trajectory formed by connecting position points according to a time sequence, sparse trajectory, and segmented trajectory.

[0026] In this technical solution, conventional vessels regularly feed back raw data and navigate along designated tracks. Track changes are also reported. The processes of arrival, stopover, and departure follow certain rules. However, abnormal vessels attempt to circumvent these rules, so their raw data feedback and tracks may not conform to these rules. Therefore, abnormal behavior requires close monitoring. Abnormal behavior includes signal anomalies, such as signal disappearance and reappearance; and abnormal navigation tracks, such as outliers, sparse tracks, or even segmented tracks formed by connecting time-series position points. These indicate abnormal vessel behavior and warrant close attention.

[0027] In the above technical solution, the step of confirming whether a vessel is an abnormal vessel based on abnormal behavior events includes: combining abnormal behavior events into at least one abnormal behavior event group; when a vessel in the abnormal behavior cluster area meets one or more abnormal behavior event groups, the corresponding vessel is confirmed as an abnormal vessel, specifically including: after the first vessel enters the abnormal behavior cluster area, its signal disappears, and a second vessel's signal reappears in the abnormal behavior cluster area; subsequently, the first vessel's signal reappears in the abnormal behavior cluster area, while the second vessel's signal disappears, and the first and second vessels are correlated on their trajectories, the first vessel is identified as an abnormal vessel; when it is determined that within the same time range, the straight-line distance between multiple trajectory points of two vessels is less than a preset distance, both vessels are identified as abnormal vessels.

[0028] Furthermore, if the ratio of the number of daily trajectory points to the number of normal navigation trajectory points of a vessel passing through the abnormal behavior cluster area is less than the first preset ratio and this occurs for a preset number of consecutive days, and the online rate of the AIS device (AIS, short for Automatic Identification System, the shipborne equipment of the Automatic Identification System) is less than the second preset ratio, then the vessel is identified as an abnormal vessel.

[0029] In this technical solution, when abnormal behavior events are combined, correlations may be found. For example, the signal of the first vessel disappears after entering the abnormal behavior cluster area, while the signal of the second vessel reappears within the abnormal behavior cluster area. After a period of time, the signal of the first vessel reappears, and the signal of the second vessel disappears. At the same time, when performing trajectory analysis on the first and second vessels, it is found that the course trajectories of the first and second vessels are correlated. For example, if the subsequent course analysis based on the last original data of the first vessel overlaps with or is similar to the course trajectory of the second vessel, it indicates that the first and second vessels are likely the same vessel, with the first vessel impersonating the second vessel to navigate. Therefore, the first vessel is judged to be an abnormal vessel.

[0030] Furthermore, this also includes: during the tracking of vessels passing through areas of abnormal behavior, if it is found that the straight-line distance between multiple trajectory points of another vessel within the same time frame is less than a preset distance, then it is considered that the two vessels have docked. Therefore, it is necessary to focus on investigating vessels that have docked, and then determine that both vessels are abnormal.

[0031] Furthermore, if the ratio of the number of daily trajectory points to the number of normal navigation trajectory points of a vessel passing through the abnormal behavior cluster area is less than the first preset ratio and this occurs for a preset number of consecutive days, and the online rate of the AIS equipment (the shipborne equipment of the Automatic Identification System) is less than the second preset ratio, it indicates that the operator of the vessel does not want to send the original data, and it can be considered that the vessel is deliberately concealing its navigation trajectory. Therefore, the vessel is identified as an abnormal vessel.

[0032] In the above technical solution, the information recorded in the abnormal vessel alarm record includes one or more of the following: maritime mobile communication service identification code, vessel name, abnormal behavior type, alarm description, trigger location, and alarm time; and / or the distance between the abnormal behavior cluster area and the port is greater than 30 nautical miles; and / or the abnormal vessel identification method also includes: sending the abnormal vessel alarm record to the user terminal; and / or recording relevant information of the abnormal vessel, including maritime mobile communication service identification code, vessel name, abnormal behavior type, abnormal behavior occurrence area, and abnormal behavior occurrence time; receiving query instructions to query relevant information of the abnormal vessel, or outputting relevant information of the abnormal vessel after receiving an information export instruction.

[0033] In the above technical solution, the information recorded in the abnormal vessel alarm record includes one or more of the following: maritime mobile communication service identification code, vessel name, abnormal behavior type, alarm description, trigger location, and alarm time.

[0034] Furthermore, the distance between the abnormal behavior cluster area and the port is greater than 30 nautical miles. Ships must enter the port, therefore their navigation tracks at the port are the most concentrated, offering no reference value for identifying abnormal behavior. Therefore, by limiting the distance between the abnormal behavior cluster area and the port to greater than 30 nautical miles, the accuracy of identifying abnormal behavior through the abnormal behavior cluster area is improved.

[0035] Furthermore, the method for identifying abnormal vessels also includes: sending abnormal vessel alarm records to the user terminal; and / or recording relevant information about the abnormal vessel, including at least one of the following: maritime mobile communication service identifier, vessel name, type of abnormal behavior, area where the abnormal behavior occurred, and time of occurrence of the abnormal behavior; receiving a query instruction to query relevant information about the abnormal vessel, or, upon receiving an information export instruction, outputting relevant information about the abnormal vessel. After identifying an abnormal vessel, it is necessary to record information such as the vessel name, type of abnormal behavior, alarm description and triggering location, and alarm time to provide corresponding evidence for subsequent judgment.

[0036] A second aspect of the present invention provides a ship anomaly identification system, comprising: an abnormal behavior cluster area determination module for determining abnormal behavior cluster areas; an abnormal behavior detection module for performing anomaly analysis on the behavior of ships within the abnormal behavior cluster areas and generating abnormal behavior events for each ship; an abnormal ship confirmation module for confirming whether a ship is an abnormal ship based on the abnormal behavior events; and an alarm module for generating an abnormal ship alarm record and / or issuing an alarm to the confirmed abnormal ship based on the confirmed abnormal ship.

[0037] The ship anomaly identification system provided by the technical solution of the present invention includes an abnormal behavior cluster area determination module for determining abnormal behavior cluster areas; an abnormal behavior detection module for performing abnormal analysis on the behavior of ships within the abnormal behavior cluster areas and generating abnormal behavior events for each ship; an abnormal ship confirmation module for confirming whether a ship is an abnormal ship based on the abnormal behavior events; and an alarm module for generating an abnormal ship alarm record and / or issuing an alarm to the confirmed abnormal ship based on the confirmed abnormal ship. Furthermore, since the ship anomaly identification system provided by the technical solution of the present invention implements the steps of the ship anomaly identification method provided in the first aspect of the present invention, the ship anomaly identification system possesses all the technical effects of the ship anomaly identification method, which will not be elaborated further here.

[0038] A third aspect of the present invention provides a ship anomaly identification system, including a memory and a processor, wherein the memory stores programs and / or instructions that can run on the processor, and the programs and / or instructions, when executed by the processor, implement the steps of the ship anomaly identification method of any of the above-described technical solutions.

[0039] The ship anomaly identification system provided by the technical solution of the present invention includes a memory, a processor, and a program stored in the memory and executable on the processor. When the program is executed by the processor, it implements the steps defined in any of the aforementioned ship anomaly identification methods. Furthermore, since the ship anomaly identification system of this application can implement the steps defined in any of the aforementioned ship anomaly identification methods, the ship anomaly identification system provided by this technical solution possesses all the beneficial effects of the ship anomaly identification methods provided in any of the aforementioned technical solutions.

[0040] A fourth aspect of the present invention provides a readable storage medium having a program and / or instructions stored thereon, wherein the program and / or instructions, when executed by a processor, implement the steps of the ship anomaly identification method in any of the above-described technical solutions.

[0041] The readable storage medium provided by the technical solution of the present invention has all the beneficial technical effects of the above-described ship anomaly identification method when the program and / or instructions stored thereon are executed by a processor. Therefore, it has all the beneficial technical effects of the above-described ship anomaly identification method.

[0042] Additional aspects and advantages of the invention will become apparent in the following description or may be learned by practice of the invention. Attached Figure Description

[0043] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:

[0044] Figure 1This is a flowchart illustrating a method for identifying ship anomalies according to an embodiment of the present invention.

[0045] Figure 2 This is a flowchart illustrating a method for identifying ship anomalies according to another embodiment of the present invention;

[0046] Figure 3 This is a block diagram of a ship anomaly identification system according to an embodiment of the present invention;

[0047] Figure 4 This is another block diagram of a ship anomaly identification system according to an embodiment of the present invention;

[0048] Figure 5 This is a schematic diagram of flight paths and abnormal behavior clusters according to an embodiment of the present invention;

[0049] Figure 6 This is a schematic diagram of abnormal behavior events according to an embodiment of the present invention;

[0050] Figure 7 This is a schematic diagram of an abnormal ship alarm record according to an embodiment of the present invention.

[0051] in, Figures 3 to 6 The correspondence between the reference numerals and component names in the attached drawings is as follows:

[0052] 300 Anomaly identification system for vessels; 302 Abnormal behavior cluster area determination module; 304 Abnormal behavior detection module; 306 Abnormal vessel confirmation module; 308 Alarm module; 400 Anomaly identification system for vessels; 402 Memory; 404 Processor; 502 Historical trajectory information; 504 Abnormal behavior cluster area; 602 Route information of vessel A; 604 Signal disappearance point of vessel A; 606 Route information of vessel B; 608 Signal reproduction point of vessel B. Detailed Implementation

[0053] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0054] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0055] The following reference Figures 1 to 7 This invention describes methods and systems for identifying ship anomalies in some embodiments, as well as readable storage media.

[0056] In one embodiment according to this application, such as Figure 1 As shown, a method for anomaly identification of ships is proposed, including:

[0057] S102, Identify areas where abnormal behavior is concentrated.

[0058] S104, perform anomaly analysis on the behavior of at least one vessel within the abnormal behavior cluster area, and generate abnormal behavior events for each vessel.

[0059] S106, confirm whether a vessel is an abnormal vessel based on abnormal behavior events.

[0060] S108, Generate an abnormal vessel alarm record and / or issue an alarm to the confirmed abnormal vessel based on the confirmed abnormal vessel.

[0061] According to the ship anomaly identification method provided by this invention, the confirmation of abnormal behavior clusters includes existing sea areas prone to abnormal behavior, as well as sea areas with dense shipping routes obtained by analyzing ships with abnormal behavior in previous years. Abnormal behavior clusters are more likely to exhibit abnormal behavior than other sea areas. After identifying the abnormal behavior clusters, anomaly analysis is performed on ships passing through them, including but not limited to signal anomalies and trajectory anomalies, thereby identifying the abnormal behavior events of each ship. Since signal anomalies may be caused by equipment failure or channel congestion, after identifying the abnormal behavior events of each ship, it is necessary to further confirm whether the ship is an abnormal ship, i.e., whether abnormal behavior has occurred on the ship. Specifically, if the trajectories of two ships are within the same time frame, the spatial distance is too close, or the ship information changes from the first ship information to the second ship information upon entering the jurisdiction and then back to the first ship information upon leaving the jurisdiction, and the first and second ship information have continuity in their spatial trajectories, then the ship exhibiting abnormal behavior is considered an abnormal ship. After identifying abnormal vessels, an abnormal vessel alarm record is generated and / or an alarm is triggered on the confirmed abnormal vessel, thereby enabling early warning of abnormal vessels or evidence collection and alarm record generation for abnormal vessel behavior. Since this application detects all vessels passing through abnormal behavior clusters and uses big data analysis, it avoids the subjectivity issues of existing abnormal vessel identification methods, thus improving the accuracy of abnormal vessel identification. Furthermore, this method has a simple model and can continuously update data in real time by changing abnormal behavior clusters, thereby investigating new abnormal behavior clusters, making it highly flexible.

[0062] In the above embodiments, the step of determining the abnormal behavior cluster area includes: determining the historical trajectory information of ships that meet the preset characteristics within a preset time period; and analyzing the historical trajectory information of ships that meet the preset characteristics to determine the abnormal behavior cluster area.

[0063] In this embodiment, abnormal maritime behavior often has a certain time frame, such as periodic vessel changes, meaning that a designated vessel only participates in abnormal behavior for a limited period. However, in the initial period after activation, due to the limited available data, it is often overlooked by law enforcement agencies. Furthermore, the transaction methods and locations between abnormal vessels can change. Therefore, data from excessively long periods is not beneficial for analyzing abnormal behavior clusters; instead, it expands the detection scope, giving vessels violating relevant regulations an opportunity to exploit loopholes. Thus, it is necessary to limit historical trajectory information to a preset time period. Abnormal vessels often share certain commonalities, such as the same origin or belonging to the same organization or country. While not all vessels with these characteristics will engage in abnormal behavior, the probability of such behavior increases. Therefore, the historical trajectory information of these vessels becomes crucial for determining whether a vessel has engaged in abnormal behavior. Without knowing which vessels will engage in abnormal behavior, to reduce the search scope, the focus can be on vessels prone to abnormal behavior, such as those meeting preset characteristics. Simultaneously, the historical trajectory information of vessels prone to abnormal behavior (e.g., those meeting preset characteristics) is analyzed to identify abnormal behavior clusters. Then, by conducting further investigations on vessels passing through areas with a high concentration of abnormal behavior, the investigation of abnormal vessels was strengthened.

[0064] In the above embodiments, such as Figure 5 As shown, vessels meeting the preset characteristics within the preset time period include vessels with abnormal characteristics within the preset time period, vessels departing from a preset nationality within the preset time period, and vessels departing from a preset departure point within the preset time period.

[0065] In this embodiment, due to the abnormal behavior cluster area 504 (e.g. Figure 5The judgment (shown) mainly relies on the analysis of historical navigation data. Therefore, the selection of which historical navigation data to use becomes the standard for the actual scope of the abnormal behavior cluster zone 504. An excessively large abnormal behavior cluster zone 504 will increase the workload, thus providing opportunities for abnormal vessels. Conversely, an excessively small abnormal behavior cluster zone 504 will result in a large number of vessels slipping through the net, and many abnormal vessels will go undetected. Therefore, the scope of the abnormal behavior cluster zone 504 is crucial for the implementation of this application. This application integrates the historical route information of vessels with abnormal characteristics within a preset time period, or vessels from preset nationalities, preset departure points, or preset organizations within a preset time period, thereby providing early warnings of the actions of these abnormal vessels. Since the predetermined nationalities or departure points are already key targets for control, vessels with predetermined characteristics are more likely to exhibit abnormal behavior. Focusing on investigating vessels with predetermined characteristics can effectively improve the efficiency of early warning. At the same time, to prevent relevant personnel from changing the routes and vessels with abnormal behavior using entirely new methods, all vessels within the preset time period can also be analyzed, thereby achieving more comprehensive detection. Considering the psychology of suspicious individuals, routes and vessels that have already been heavily investigated will not be used. Therefore, vessels and routes with excessively long durations may mislead the existing situation, thus giving relevant personnel an opportunity to exploit the situation. Therefore, the optimal solution is to set the preset time to one year. One year is not too short for relatively concealed abnormal behavior, but not too long for abnormal behavior that frequently changes vessels or routes.

[0066] Further, the step of analyzing the historical trajectory information 502 of ships that meet the preset characteristics to determine the abnormal behavior cluster area 504 includes: analyzing the historical trajectory information 502 based on the spatial density algorithm to determine the abnormal behavior cluster area 504. By using the spatial density algorithm, the area with the most abnormal ships passing through can be clearly identified, and this area is defined as the abnormal behavior cluster area 504.

[0067] In the above embodiments, the step of determining the historical trajectory information of ships that meet the preset characteristics within a preset time period includes: acquiring the original data of ships that meet the preset characteristics within the preset time period; processing the original data to remove abnormal information from the original data; connecting the original data after removing abnormal information according to time to generate the navigation trajectory line of the ships that meet the preset characteristics; correcting the navigation trajectory to obtain the historical trajectory information of ships that meet the preset characteristics within the preset time period; the original data includes the maritime mobile communication service identification code, latitude and longitude, ship speed, and heading.

[0068] In this embodiment, the step of determining the historical trajectory information of ships that meet the preset characteristics within a preset time period includes: acquiring the original data of ships that meet the preset characteristics within the preset time period; processing the original data to remove abnormal information; connecting the original data after removing abnormal information according to time to generate the navigation trajectory line of the ships that meet the preset characteristics; correcting the navigation trajectory to obtain the historical trajectory information of ships that meet the preset characteristics within the preset time period; the original data includes the maritime mobile communication service identification code, latitude and longitude, ship speed, and heading. In actual use, the raw data received by this application may be tampered with, or due to equipment failure and channel congestion, the raw data may contain abnormal information, such as latitude and longitude significantly exceeding 180°, excessive ship speed, or missing key records. Raw data with erroneous information cannot be used as a standard to determine the abnormal behavior cluster area 504. Therefore, it is necessary to clean the raw data to restore the true and reasonable raw data as much as possible. After cleaning, the raw data needs to be connected according to time to generate the preset ship navigation trajectory line. Since it is not possible to remove all unreasonable data during data cleaning, non-smooth and unreasonable points will be processed when connecting the route trajectory line to obtain the preset ship historical trajectory information 502, thereby improving the accuracy of the historical trajectory information and improving the reliability of the abnormal behavior cluster area 504.

[0069] Furthermore, the original data for the vessel is global navigation trajectory data within one year. If only the route data of the vessel near my country's territorial waters is considered, the determination of the abnormal behavior cluster area 504 will have certain limitations. However, abnormal behavior is global. Regardless of where the abnormal behavior cluster area 504 is located, when an abnormal vessel passes through my country's territorial waters, it must be inspected by my country's law enforcement agencies. In order to prevent foreign abnormal vessels from entering my country's territorial waters and engaging in abnormal behavior, the historical trajectory information 502 used should be global navigation trajectory data.

[0070] The raw data includes one or more of the following: maritime mobile communication service identification code, latitude and longitude, ship speed, and heading. The raw data can mutually corroborate each other, proving the reasonableness of the data. For example, given that the ship's speed and heading are fixed, under non-extreme weather conditions, the latitude and longitude, ship speed, and heading in the next set of raw data will fall within a certain range, thus ensuring the objectivity of the raw data and improving the accuracy of historical trajectory information 502.

[0071] In the above embodiments, the step of performing anomaly analysis on the behavior of ships within the abnormal behavior cluster area and generating anomaly behavior events for each ship includes: performing anomaly analysis based on the pre-arrival behavior, dwelling behavior, and departure behavior of each ship to generate anomaly behavior events for each ship; the step of performing anomaly analysis on the behavior of ships within the abnormal behavior cluster area includes: determining whether the signals during the pre-arrival, dwelling, and departure processes of the ship are abnormal, wherein signal abnormality includes at least signal disappearance and recurrence; determining whether the trajectory during the pre-arrival, dwelling, and departure processes of the ship is abnormal, wherein trajectory abnormality includes at least one of the following: outlier points appearing in the ship trajectory formed by connecting position points according to the time sequence, trajectory sparseness, and segmented trajectory.

[0072] In this embodiment, trajectory anomalies include at least one of the following: outliers, sparse trajectories, and segmented trajectories in a ship's trajectory formed by connecting position points in a time sequence. Conventional ships periodically report raw data and navigate according to designated trajectories. Changes in trajectory are also reported. The processes of arrival, stopover, and departure follow certain rules. Abnormal ships, however, attempt to circumvent these rules, so both their reported raw data and trajectories may deviate from these rules. Therefore, abnormal behavior requires close monitoring. Abnormal behavior includes signal anomalies, such as signal disappearance and reappearance, and navigation trajectory anomalies, such as outliers, sparse trajectories, or even segmented trajectories in a ship's trajectory formed by connecting position points in a time sequence. This indicates abnormal ship behavior and warrants close monitoring.

[0073] In the above embodiments, the step of confirming whether a vessel is an abnormal vessel based on abnormal behavior events includes: grouping abnormal behavior events into at least one abnormal behavior event group; when a vessel in the abnormal behavior cluster area meets one or more abnormal behavior event groups, the corresponding vessel is confirmed as an abnormal vessel, specifically including: when the signal of the first vessel disappears after entering the abnormal behavior cluster area, and the signal of the second vessel reappears in the abnormal behavior cluster area, and then the signal of the first vessel reappears in the abnormal behavior cluster area while the signal of the second vessel disappears, and the first vessel and the second vessel are correlated on the trajectory, the first vessel is identified as an abnormal vessel; when it is determined that the straight-line distance between multiple trajectory points of two vessels within the same time range is less than a preset distance, the two vessels are identified as abnormal vessels; when the ratio of the number of trajectory points per day to the number of normal navigation trajectory points of a vessel passing through the abnormal behavior cluster area is less than a first preset ratio and occurs continuously for a preset number of days, and the online rate of the AIS equipment is less than a second preset ratio, the vessel is identified as an abnormal vessel.

[0074] In this embodiment, when abnormal behavior events are combined, a correlation may be found. For example, the signal of the first vessel disappears after entering the abnormal behavior aggregation zone 504, while the signal of the second vessel appears in the abnormal behavior aggregation zone 504, i.e., the signal of the second vessel reappears. After a period of time, the signal of the first vessel reappears, and the signal of the second vessel disappears. At the same time, when performing trajectory analysis on the first and second vessels, it is found that the course trajectories of the first and second vessels have a certain correlation. For example, if the subsequent course analysis based on the last original data of the first vessel overlaps with or is similar to the course trajectory of the second vessel, it indicates that the first and second vessels are likely the same vessel, and the first vessel is impersonating the second vessel to navigate, thus judging the first vessel as an abnormal vessel.

[0075] Furthermore, this also includes: during the tracking of vessels passing through areas of abnormal behavior, if it is found that the straight-line distance between multiple trajectory points of another vessel within the same time frame is less than a preset distance, then it is considered that the two vessels have docked. Therefore, it is necessary to focus on investigating vessels that have docked, thus determining that both vessels are abnormal.

[0076] Furthermore, if the ratio of the number of daily trajectory points to the number of normal navigation trajectory points of a vessel passing through the abnormal behavior cluster 504 is less than the first preset ratio and this occurs for a preset number of consecutive days, and the AIS equipment online rate is less than the second preset ratio, it indicates that the operator of the vessel does not want to send the original data. It can be considered that the vessel is deliberately concealing its navigation trajectory. Therefore, the vessel is identified as an abnormal vessel.

[0077] In the above embodiments, the information recorded in the abnormal vessel alarm record includes one or more of the following: maritime mobile communication service identification code, vessel name, abnormal behavior type, alarm description, trigger location, and alarm time.

[0078] In this embodiment, the information recorded in the abnormal vessel alarm record includes one or more of the following: maritime mobile communication service identification code, vessel name, abnormal behavior type, alarm description, trigger location, and alarm time. After an abnormal vessel is identified, the vessel name, abnormal behavior type, alarm description, trigger location, alarm time, and other information need to be recorded to provide corresponding evidence for subsequent judgment.

[0079] Furthermore, the distance between the abnormal behavior cluster area 504 and the port is greater than 30 nautical miles. Since ships must enter the port, their navigation tracks at the port are the most concentrated, thus offering no reference value for identifying abnormal behavior. Therefore, by limiting the distance between the abnormal behavior cluster area 504 and the port to greater than 30 nautical miles, the accuracy of the abnormal behavior cluster area 504 in identifying abnormal behavior is improved.

[0080] Furthermore, the method for identifying abnormal vessels also includes: sending abnormal vessel alarm records to the user terminal; and / or recording relevant information about the abnormal vessel, including at least one of the following: maritime mobile communication service identifier, vessel name, abnormal behavior type, abnormal behavior occurrence area, and abnormal behavior occurrence time; receiving a query instruction to query relevant information about the abnormal vessel, or, upon receiving an information export instruction, outputting relevant information about the abnormal vessel. Specifically, after identifying an abnormal vessel, it is necessary to record information such as the vessel name, abnormal behavior type, alarm description and triggering location, and alarm time to provide corresponding evidence for subsequent judgment.

[0081] In one embodiment according to this application, such as Figure 2 As shown, the proposed method for anomaly identification of ships also includes:

[0082] S202, determine the historical trajectory information of ships that meet the preset characteristics within a preset time period.

[0083] S204, analyze the historical trajectory information of ships that meet the preset characteristics to determine the cluster areas of abnormal behavior.

[0084] S206 performs anomaly analysis on the behavior of ships within the abnormal behavior cluster area, generating abnormal behavior events for each ship.

[0085] S208, confirming whether a vessel is an abnormal vessel based on abnormal behavior events.

[0086] S210, Generate an abnormal vessel alarm record and / or issue an alarm to the confirmed abnormal vessel based on the confirmed abnormal vessel.

[0087] Specifically: The first step is data acquisition. Raw AIS data (data acquired by the Automatic Identification System of Ships) is obtained through AIS base stations (base stations of the Automatic Identification System of Ships) and satellite equipment. Based on the special identifier translation, the raw AIS data is parsed and outputs data fields that can be used intuitively, such as the maritime mobile communication service identification code, latitude and longitude, speed, and heading.

[0088] The second step is data cleaning. The AIS data parsed in the previous step is cleaned. Due to human tampering, equipment failure, channel congestion, and other reasons, the parsed AIS data contains a large number of errors. For example, in terms of accuracy, the latitude and longitude are significantly higher than 180°, the speed is too high, and the maritime mobile communication service identification code does not meet the norm. In terms of physical integrity, there are records with missing key fields, such as latitude and longitude, timestamp, and maritime mobile communication service identification code. These points are cleaned to obtain the cleaned data.

[0089] The third step is trajectory correction. Based on the AIS ship position data cleaned in the second step, these data points are connected according to their timestamps to generate the ship's navigation trajectory. The trajectory of each ship is then corrected. Further correction primarily addresses out-of-range points on the ship's trajectory, such as points where speed is unreachable, points where the distance between adjacent points is too large, or points where adjacent points form extremely sharp angles. These noise points on the trajectory are processed to obtain a smooth trajectory curve.

[0090] Step 4: Identification of abnormal behavior clusters, based on the vessel locations and trajectories processed in Step 3, such as... Figure 5 As shown, historical trajectory information 502 of a certain nationality (based on information about the vessel involved in the incident in news reports, the nationality is analyzed to be Sierra Leone) was generated over the past year. Furthermore, by comparing different clustering recognition algorithms, the DBSCAN algorithm based on spatial density was found to meet the calculation scenario of this patent. This algorithm can identify samples of arbitrary shapes, divide areas with sufficient density into clusters, and discover clusters of arbitrary shapes in a noisy spatial database. In this patent, the minimum number of points used in the DBSCAN algorithm is specified as 1000, identifying areas with high trajectory clustering. It should be noted that the initial identification results were further processed. Since a large number of points are generated during normal ship operations and port calls, these clusters were discarded. The distance of the cluster center point to the nearest port was determined; when the distance was greater than 30 nautical miles, the cluster was retained. The final identification result is that areas at sea or within a certain range near a port are designated as areas prone to abnormal maritime behavior, i.e., abnormal behavior cluster areas 504.

[0091] Step 5: Anomaly detection. Based on the abnormal behavior cluster area 504 identified in Step 4, the behavior of ships (including those arriving, staying, and departing) within it is identified, and abnormal behavior events are generated, including signal anomalies such as signal disappearance and recurrence; trajectory anomalies, that is, the ship trajectory formed by connecting the position points according to the time sequence has outliers, sparse trajectories, segmented trajectories, etc.

[0092] Furthermore, signal disappearance and reappearance are determined by analyzing the historical trajectory information of regional vessels and vessels expected to arrive. A record is made of instances where AIS equipment has not updated for 30 minutes. When a vessel that previously updated AIS data normally suddenly stops updating for 30 minutes, it is recorded as signal disappearance. Similarly, signal reappearance refers to an event where a vessel that has not updated for a long time suddenly updates. This mainly includes two situations: one is that a vessel had a signal in the past (e.g., within 7 days), but the signal disappeared for more than 30 minutes in the middle and then returned; the other is that a vessel had no signal in the past (e.g., within 7 days) and suddenly sent a signal to update. Both of these situations are identified as signal reappearance.

[0093] The detection of trajectory anomalies includes outliers, also known as fly-outs, which are points that suddenly appear far away from the normal navigation points in close proximity (this is usually due to excessive time intervals, and the trajectory correction in the third step has no effect on it), and the angle between them and the two points before and after is extremely acute. This type of trajectory anomaly is called an outlier trajectory anomaly. Sparse trajectory refers to the fact that the trajectory of a normally navigating ship should show points with small intervals and continuous points. If there are continuous trajectories but the distance between the two points before and after is too large (and the time difference is also large), this situation is identified as sparse trajectory. Segmented trajectory is usually caused by ship clone (i.e., two AIS devices in far-away locations simultaneously sending the same ship information). This situation is manifested on the trajectory as the ship's segmented trajectory appearing at different locations at the same time.

[0094] Step 6: Abnormal Vessel Identification. Based on the detection of abnormal vessel behavior events in Step 4, the abnormal behavior events are combined to identify vessels within the abnormal behavior cluster area.

[0095] Abnormal behavior type one: such as Figure 6 As shown, this type of behavior manifests as follows: within the area of ​​abnormal behavior aggregation, ship A's information is used to generate ship A's route information 602. After entering the jurisdiction, ship A's signal disappears, generating ship A's signal disappearance point 604. It then disguises itself as ship B's information, and ship B's signal reappears, generating ship B's signal reappearance point 608. Operating at night, ship A uses ship B's information to conduct transactions within the jurisdiction, generating ship B's route information 606. When leaving the jurisdiction, it switches back to ship A's information, and ship A's signal reappears. Simultaneously, ship A and ship B have spatial correlation in their trajectories, such as... Figure 6 As shown, the identification of abnormal vessels includes the disappearance and reappearance of signals associated with the vessels and the spatial correlation on their trajectories.

[0096] Type 2 of Abnormal Behavior: The trajectory of two ships is abnormal. The trajectories of two ships are too close in space within the same time range (there are multiple trajectory points with a straight-line distance of less than 50 meters). This is considered to be a docking phenomenon, and suspected abnormal ships are identified.

[0097] Type 3 of Abnormal Behavior: Abnormal ship signals, sparse tracks, and low equipment online rate. By analyzing the historical 7-day tracks of ships and the online rate of AIS equipment in the abnormal behavior cluster area, if it is found that the number of track points per day is less than 20% of the number of normal navigation track points (the number of normal navigation track points is set to 8500 points / day), and the online rate of AIS equipment is less than 20%, this type of behavior occurs in the abnormal behavior cluster area with good signal coverage, and the ship is suspected of having abnormal behavior, and an abnormal ship identification record is generated.

[0098] Furthermore, based on this method, different abnormal ship events can be combined as needed without inputting multiple parameters and model debugging. By analyzing the AIS data of the abnormal behavior cluster 504, abnormal behavior types can be generated and the corresponding abnormal ships can be identified.

[0099] Step 7: Generate and push alarm records. Based on the vessels identified in Step 6, generate alarm records for abnormal vessels, such as... Figure 7 As shown, the records include the Maritime Mobile Communication Service Identifier (MMSI), vessel name, abnormal behavior type, alarm description, trigger location, alarm time, etc. The corresponding abnormal records are pushed to users via email and WeChat official account so that they can be informed in a timely manner and take further action. At the same time, it supports querying and exporting records for specific vessels, specific areas, and specific types.

[0100] like Figure 3 As shown, a second aspect of the present invention provides a ship anomaly identification system 300, comprising: an abnormal behavior cluster area determination module 302, used to determine an abnormal behavior cluster area 504; an abnormal behavior detection module 304, used to perform anomaly analysis on the behavior of ships within the abnormal behavior cluster area and generate an abnormal behavior event for each ship; an abnormal ship confirmation module 306, used to confirm whether a ship is an abnormal ship based on the abnormal behavior event; and an alarm module 308, used to generate an abnormal ship alarm record and / or issue an alarm to the confirmed abnormal ship based on the confirmed abnormal ship.

[0101] The ship anomaly identification system 300 provided according to the embodiments of the present invention, since it is used to implement the steps of the ship anomaly identification method provided in the first aspect of the present invention, has all the technical effects of the ship anomaly identification method, which will not be repeated here.

[0102] A third aspect of the present invention provides a ship anomaly identification system 400, such as... Figure 4 As shown, it includes a memory 402 and a processor 404. The memory 402 stores programs and / or instructions that can be executed on the processor 404. When the programs and / or instructions are executed by the processor 404, they implement the steps of the ship anomaly identification method of the first aspect described above.

[0103] The ship anomaly identification system 400 provided according to an embodiment of the present invention includes a memory 402, a processor 404, and a program stored in the memory 402 and executable on the processor 404. When the program is executed by the processor 404, it implements the steps defined in any of the aforementioned ship anomaly identification methods. Furthermore, since the ship anomaly identification system of this application can implement the steps defined in the aforementioned first aspect of the ship anomaly identification method, the ship anomaly identification system provided in this embodiment has all the beneficial effects of the ship anomaly identification method provided in the first aspect of the embodiment.

[0104] A fourth aspect of the present invention provides a readable storage medium having a program and / or instructions stored thereon, which, when executed by a processor 404, implement the steps of the ship anomaly identification method described in the first aspect embodiment above.

[0105] The readable storage medium provided according to the embodiments of the present invention has all the beneficial technical effects of the above-described ship anomaly identification method when the program and / or instructions stored thereon are executed by the processor 404.

[0106] In this specification, the term "multiple" refers to two or more unless otherwise expressly defined. The terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; "linking" can be a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0107] In the description of this specification, the terms "one embodiment," "some embodiments," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0108] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method of anomaly recognition of a ship, characterized by, The method comprises the following steps: determining an abnormal behavior aggregation area; performing abnormal analysis on the behavior of at least one ship in the abnormal behavior aggregation area to generate abnormal behavior events of each ship; confirming whether the ship is an abnormal ship according to the abnormal behavior events; generating an abnormal ship alarm record and / or alarming the confirmed abnormal ship according to the confirmed abnormal ship; The step of determining the abnormal behavior aggregation area comprises: determining historical trajectory information of ships meeting preset characteristics within a preset time period; analyzing the historical trajectory information of the ships meeting the preset characteristics to determine the abnormal behavior aggregation area; The step of performing abnormal analysis on the behavior of at least one ship in the abnormal behavior aggregation area to generate abnormal behavior events of each ship comprises: performing abnormal analysis on the pre-arrival behavior, stay behavior and departure behavior of each ship to generate abnormal behavior events of each ship; The step of performing abnormal analysis on the behavior of the ship in the abnormal behavior aggregation area comprises: judging whether the signals in the pre-arrival, stay and departure processes of the ship are abnormal, wherein the signal abnormality at least includes signal disappearance and reappearance; judging whether the trajectory in the pre-arrival, stay and departure processes of the ship is abnormal, wherein the trajectory abnormality includes at least one of outlier, sparse trajectory and segmented trajectory of the ship trajectory formed by connecting position points in time sequence.

2. The abnormality identification method of a ship according to claim 1, characterized by, The ships meeting the preset characteristics within the preset time period include ships with abnormal characteristics within the preset time period, or ships from a preset nationality within the preset time period and ships from a preset departure location within the preset time period; and / or The preset time period is one year; and / or The step of analyzing the historical trajectory information of the ships meeting the preset characteristics to determine the abnormal behavior aggregation area comprises: analyzing the historical trajectory information based on an improved DBSCAN algorithm of spatial density to determine the abnormal behavior aggregation area. The step of determining the historical trajectory information of the ships meeting the preset characteristics within the preset time period comprises:

3. The abnormality identification method of a ship according to claim 1, characterized by, obtaining raw data of the ships meeting the preset characteristics within the preset time period; processing the raw data to remove abnormal information in the raw data; connecting the raw data after removing the abnormal information according to time to generate a sailing trajectory of the ships meeting the preset characteristics; correcting the sailing trajectory to obtain the historical trajectory information of the ships meeting the preset characteristics within the preset time period; The raw data includes one or more of the following data: water mobile communication service identification code, latitude and longitude, ship speed and ship heading. The step of confirming whether the ship is an abnormal ship according to the abnormal behavior events comprises:

4. The abnormality identification method of a ship according to claim 1, characterized by, combining the abnormal behavior events into at least one abnormal behavior event group; when the ships in the abnormal behavior aggregation area meet one or more abnormal behavior event groups, confirming that the corresponding ship is an abnormal ship, specifically comprising: ​ The first ship is determined as an abnormal ship when the signal disappears after the first ship enters the abnormal behavior gathering area, the second ship signal reappears in the abnormal behavior gathering area, then the first ship reappears in the abnormal behavior gathering area, the second ship signal disappears, and the first ship and the second ship have relevance in the trajectory. The ship is determined as an abnormal ship when the ratio of the number of daily trajectory points of the ship passing through the abnormal behavior gathering area to the number of normal navigation trajectory points is less than a first preset ratio and appears continuously for a preset number of days, and the online rate of the AIS device is less than a second preset ratio.

5. The abnormality identification method of a ship according to any one of claims 1 to 4, characterized in that, The information recorded in the abnormal ship alarm record includes one or more of the following: a water mobile communication service identification code, a ship name, an abnormal behavior type, an alarm description, a triggering location, and an alarm time. The distance between the abnormal behavior gathering area and the port is greater than 30 nautical miles. The abnormal ship identification method further includes: The abnormal ship alarm record is sent to a user terminal; and / or The related information of the abnormal ship is recorded, and the related information includes at least one of the following: a water mobile communication service identification code, a ship name, an abnormal behavior type, an abnormal behavior occurrence area, and an abnormal behavior occurrence time. The related information of the abnormal ship is outputted after receiving a query instruction to query the related information of the abnormal ship or receiving an information export instruction.

6. An abnormality recognition system for a ship, characterized by comprising: The abnormal behavior gathering area determination module is configured to determine an abnormal behavior gathering area. The abnormal behavior detection module is configured to perform abnormal analysis on the behavior of the ship in the abnormal behavior gathering area to generate an abnormal behavior event of each ship. The abnormal ship confirmation module is configured to confirm whether the ship is an abnormal ship according to the abnormal behavior event. The alarm module is configured to generate an abnormal ship alarm record according to the confirmed abnormal ship and / or alarm the confirmed abnormal ship. The step of determining the abnormal behavior gathering area includes: Determining historical trajectory information of ships satisfying a preset feature in a preset time period. Analyzing the historical trajectory information of the ships satisfying the preset feature to determine the abnormal behavior gathering area. The step of performing abnormal analysis on the behavior of the ship in the abnormal behavior gathering area to generate an abnormal behavior event of each ship includes: Performing abnormal analysis on the pre-arrival behavior, stay behavior, and departure behavior of each ship to generate an abnormal behavior event of each ship. The step of performing abnormal analysis on the behavior of the ship in the abnormal behavior gathering area includes: Judging whether the signal in the pre-arrival, stay, and departure process of the ship is abnormal, wherein the signal abnormality at least includes signal disappearance and reappearance. Judging whether the trajectory in the pre-arrival, stay, and departure process of the ship is abnormal, wherein the trajectory abnormality includes at least one of the following: an outlier point in the ship trajectory formed by connecting position points in time sequence, trajectory sparseness, and segmented trajectory. ​ 7. An abnormality recognition system for a ship, characterized by comprising: A computer program product including a memory storing a program and / or instructions executable on a processor, the program and / or instructions, when executed by the processor, implementing the steps of the anomaly identification method of a ship according to any one of claims 1 to 5.

8. A readable storage medium, characterized by, A computer program product including a memory storing a program and / or instructions executable on a processor, the program and / or instructions, when executed by the processor, implementing the steps of the anomaly identification method of a ship according to any one of claims 1 to 5.

Images