Computer system and method for dark ship tracking

Abstract

Systems and methods for tracking a non-self-reporting vessel ("dark vessel") with an unknown identity are provided. A method includes: providing a system configured to receive and store a plurality of ship detections; comparing a timestamp T2 of a new ship detection having a geolocation of P2 to a timestamp T1 of a first ship detection having a geolocation of P1 and determining that the T2 is later than T1; using kinematical information of the first ship detection, projecting the first ship detection to a projected second geolocation at T2; determining that the projected second geolocation is within a threshold proximity of P2; forming a vessel track having a geolocation of P1 at T1 and a geolocation of P2 at T2; and storing the vessel track in the system as a dark vessel track.

Description

COMPUTER SYSTEM AND METHOD FOR DARK SHIP TRACKINGTechnical Field

[0001] The following relates generally to geointelligence and marine surveillance, and more particularly to systems and methods for dark ship tracking.Introduction

[0002] Constructing vessel tracks for targets that are self-reporting, (like though AIS or VMS, is quite straightforward. Vessels’ position reports are typically sorted according to vessel identities (typically MMISs or IMOs), and timestamps in a descending manner, and then are linked according to their timestamps.

[0003] With dark targets, this task is much harder to do because dark vessels’ identities are inherently not available.

[0004] Accordingly, there is a need for an improved system and method for dark ship tracking that overcomes at least some of the disadvantages of existing systems and methods.Summary

[0005] A method for tracking a non-self-reporting vessel (“dark vessel”) with an unknown identity is provided. The method includes providing a system configured to receive and store a plurality of ship detections of at least one sensor type, the plurality of ship detections including a first ship detection D1 having a sensor type, a first geolocation P1 , a timestamp T1 , and kinematical information associated therewith, the first ship detection D1 being non-self-reporting; receiving by the system a new ship detection D2 of the at least one sensor type, the new ship detection D2 having a sensor type, a geolocation P2, and a timestamp T2 associated therewith, the new ship detection D2 being non-self-reporting; comparing the timestamp T2 of the new detection D2 to the timestamp T1 of the first detection D1 and determining that the timestamp T2 of the new detection D2 is later than the timestamp T 1 of the first detection D1 ; using the kinematical information of the first ship detection D1 , projecting the first ship detection D1 to a projected second geolocation P2' at the timestamp T2 of the new ship detection D2;determining that the projected second geolocation P2' is within a threshold proximity of the geolocation P2 of the new detection, the threshold proximity stored by the system; forming a vessel track having a geolocation of P1 at timestamp T1 and a geolocation of P2 at timestamp T2; and storing the vessel track in the system as a dark vessel track.

[0006] The at least one sensor type may include any one or more of a radio frequency (RF) sensor, an infrared (IR) sensor, a synthetic aperture radar (SAR) sensor, and an optical sensor.

[0007] The kinematical information may include a vessel heading and a vessel speed.

[0008] Projecting the first ship detection D1 to projected second geolocation P2' may be based on a geodetic path.

[0009] The sensor type of the new detection D2 may be a sensor type from which speed and heading are not obtainable.

[0010] The method may be performed only if a predetermined time period between the timestamps T1 , T2 of the first and new ship detections is not exceeded.

[0011] The sensor type of the first ship detection may be a sensor type that produces a 180 degree orientation ambiguity in a heading of the first ship detection and projecting the first ship detection D1 to the projected second geolocation P2' is performed bidirectionally at 180 degrees apart.

[0012] Determining that the projected second geolocation P2' is within a threshold proximity of the geolocation P2 of the new detection may include determining a correlation confidence between the projected second geolocation P2' and the geolocation P2 of the new detection, comparing the correlation confidence to a threshold correlation confidence, and determining that the correlation confidence exceeds the threshold correlation confidence.

[0013] Projecting the first ship detection D1 to the projected second geolocation P2' may use a dead reckoning technique.

[0014] The method may further include displaying a visualization of the dark vessel track in a graphical user interface, the visualization including first and second nodesrepresenting the first ship detection D1 and new ship detection D2, respectively, and an edge connecting the first and second nodes representing the dark vessel track.

[0015] The first and second node may be selectable in the graphical user interface to display information about the first ship detection D1 and the new ship detection D2, respectively, including the sensor type, the geolocation, and the timestamp.

[0016] The method may further include receiving by the system a third ship detection D3 having a sensor type, a geolocation P3, a timestamp T3, and kinematical information associated therewith, the third ship detection D3 being non-self-reporting: comparing the timestamp T3 of the third detection D3 to the timestamp T2 of the new detection D2 and determining that the timestamp T2 of the new detection D2 is earlier than the timestamp T3 of the third detection D3; projecting the third ship detection D3 to a projected third geolocation P2" at the timestamp T2 of the new ship detection D2 using the kinematical information of the third ship detection D3; determining that the projected second geolocation P2" is within the threshold proximity of the geolocation P2 of the new detection; forming a second vessel track having a geolocation of P2 at timestamp T2 and a geolocation of P3 at timestamp T3; and storing the second vessel track along with the first vessel track as part of the dark vessel track.

[0017] Further provided is a system for dark ship tracking. The system includes a communication interface configured to receive a plurality of ship detections of at least one sensor type, including: a first ship detection D1 having a sensor type, a first geolocation P1 , a timestamp T1 , and kinematical information associated therewith, the first ship detection D1 being non-self-reporting; and a new detection D2 of the at least one sensor type, the new ship detection D2 having a sensor type, a geolocation P2, and a timestamp T2 associated therewith, the new ship detection D2 being non-self-reporting; a data storage device configured to store the plurality of ship detections; and a processor configured to: compare the timestamp T2 of the new detection D2 to the timestamp T 1 of the first detection D1 and determine that the timestamp T2 of the new detection D2 is later than the timestamp T 1 of the first detection D1 ; project the first ship detection D1 to a projected second geolocation P2' at the timestamp T2 of the new ship detection D2 using the kinematical information of the first ship detection D1 ; determine that theprojected second geolocation P2' is within a threshold proximity of the geolocation P2 of the new detection, the threshold proximity stored by the system in the data storage device; form a first vessel track [segment] having a geolocation of P1 at timestamp T1 and a geolocation of P2 at timestamp T2; and store the first vessel track in the data storage device as a dark vessel track.

[0018] The plurality of ship detections may further include a third ship detection D3 having a sensor type, a geolocation P3, a timestamp T3, and kinematical information associated therewith, the third ship detection D3 being non-self-reporting, and wherein the processor is further configured to: compare the timestamp T3 of the third detection D3 to the timestamp T1 of the new detection D2 and determine that the timestamp T2 of the new detection D2 is earlier than the timestamp T3 of the third detection D3; project the third ship detection D3 to a projected second geolocation P2" at the timestamp T2 of the new ship detection D2 using the kinematical information of the third ship detection D3; determine that the projected second geolocation P2" is within the threshold proximity of the geolocation P2 of the new detection; form a second vessel track having a geolocation of P2 at timestamp T2 and a geolocation of P3 at timestamp T3; and storing the second vessel track along with the first vessel track as part of the dark vessel track.

[0019] The at least one sensor type may include any one or more of a radio frequency (RF) sensor, an infrared (IR) sensor, a synthetic aperture radar (SAR) sensor, and an optical sensor.

[0020] The kinematical information may include a vessel heading and a vessel speed.

[0021] Projecting the first ship detection D1 to projected second geolocation P2' may be based on a geodetic path.

[0022] The sensor type of the new ship detection D2 may be a sensor type from which speed and heading are not obtainable.

[0023] Comparing, projecting, determining, forming, and storing may be performed only if a predetermined time period between the timestamps T1 , T2 of the first and new ship detections is not exceeded.

[0024] The sensor type of the first ship detection may be a sensor type that produces a 180 degree orientation ambiguity in a heading of the first ship detection and projecting the first ship detection D1 to the projected second geolocation P2' is performed bidirectionally at 180 degrees apart.

[0025] Determining that the projected second geolocation P2' is within a threshold proximity of the geolocation P2 of the new detection may include determining a correlation confidence between the projected second geolocation P2' and the geolocation P2 of the new detection, comparing the correlation confidence to a threshold correlation confidence, and determining that the correlation confidence exceeds the threshold correlation confidence.

[0026] Projecting the first ship detection D1 to the projected second geolocation P2' may use a dead reckoning technique.

[0027] The processor may be further configured to display a visualization of the dark vessel track in a graphical user interface, the visualization including first and second nodes representing the first ship detection D1 and new ship detection D2, respectively, and an edge connecting the first and second nodes representing the dark vessel track.

[0028] The first and second node may be selectable in the graphical user interface to display information about the first ship detection D1 and the new ship detection D2, respectively, including the sensor type, the geolocation, and the timestamp.

[0029] Other aspects and features will become apparent, to those ordinarily skilled in the art, upon review of the following description of some exemplary embodiments.Brief Description of the Drawings

[0030] The drawings included herewith are for illustrating various examples of articles, methods, and apparatuses of the present specification. In the drawings:

[0031] Figure 1 is a schematic diagram of a system for dark ship tracking, according to an embodiment;

[0032] Figure 2 is a block diagram of a computer system for dark ship tracking including the dark ship tracker application of Figure 1 , according to an embodiment;

[0033] Figure 3 is a block diagram of example dark ship detection data of Figure 2, according to an embodiment;

[0034] Figure 4 is an example graphical user interface screen generated by the computer system of Figure 2, according to an embodiment;

[0035] Figure 5 is a flow diagram of a method of dark ship tracking, according to an embodiment;

[0036] Figure 6 is an example dark ship track generated by a dark ship tracker application, according to an embodiment; and

[0037] Figure 7 is a schematic diagram illustrating a correlation confidence technique used in the systems and methods for dark ship tracking of the present disclosure, according to an embodiment.Detailed Description

[0038] Various apparatuses or processes will be described below to provide an example of each claimed embodiment. No embodiment described below limits any claimed embodiment and any claimed embodiment may cover processes or apparatuses that differ from those described below. The claimed embodiments are not limited to apparatuses or processes having all of the features of any one apparatus or process described below or to features common to multiple or all of the apparatuses described below.

[0039] One or more systems described herein may be implemented in computer programs executing on programmable computers, each comprising at least one processor, a data storage system (including volatile and non-volatile memory and / or storage elements), at least one input device, and at least one output device. For example, and without limitation, the programmable computer may be a programmable logic unit, a mainframe computer, server, and personal computer, cloud-based program or system, laptop, personal data assistance, cellular telephone, smartphone, or tablet device.

[0040] Each program is preferably implemented in a high-level procedural or object-oriented programming and / or scripting language to communicate with a computer system. However, the programs can be implemented in assembly or machine language,if desired. In any case, the language may be a compiled or interpreted language. Each such computer program is preferably stored on a storage media or a device readable by a general or special purpose programmable computer for configuring and operating the computer when the storage media or device is read by the computer to perform the procedures described herein.

[0041] A description of an embodiment with several components in communication with each other does not imply that all such components are required. On the contrary, a variety of optional components are described to illustrate the wide variety of possible embodiments of the present invention.

[0042] Further, although process steps, method steps, algorithms or the like may be described (in the disclosure and I or in the claims) in a sequential order, such processes, methods and algorithms may be configured to work in alternate orders. In other words, any sequence or order of steps that may be described does not necessarily indicate a requirement that the steps be performed in that order. The steps of processes described herein may be performed in any order that is practical. Further, some steps may be performed simultaneously.

[0043] When a single device or article is described herein, it will be readily apparent that more than one device I article (whether or not they cooperate) may be used in place of a single device I article. Similarly, where more than one device or article is described herein (whether or not they cooperate), it will be readily apparent that a single device I article may be used in place of the more than one device or article.

[0044] The following relates generally to geointelligence systems and marine surveillance, and more particularly to systems and methods for dark ship tracking. Generally, the systems and methods may make predictions based on geodetic paths (great circles) and thus may be used to track ships that follow such paths (e.g., ships in open waters).

[0045] The systems and methods of the present disclosure enable the tracking of ships where their identity is or may not be known, through the formation of tracks from a set of dark targets. Such dark tracks can provide valuable insight into where a ship has been.

[0046] Systems and methods of the present disclosure may be implemented at one or more computer devices. For example, the system may include a plurality of computer devices in communication via a network connection. Further, components of the system (e.g., software modules and components, etc.) may be implemented at a single computer device, or across a plurality of computer devices. In some embodiments, the system includes at least one user computing device and at least one server computing device in communication via a network connection. The user device may execute an application that can interact with server-side software components (“services”) hosted by the server computing device. For example, the computer system may execute a network-based software application that executes partially at the server computing device (via serverside software components) and partially at the user device (via client-side software components). In an embodiment, the client-side software components include a user interface (e.g., web-based user interface).

[0047] Referring now to Figure 1 , shown therein is a system for tracking dark ships, according to an embodiment.

[0048] The system 100 includes a plurality of earth observation (“EO”) data collectors 102 (a single EO data collector is shown in Figure 1 for simplicity). Each EO data collector 102 collects at least one type of sensor data via one or more sensors of that type on the EO data collector 102.

[0049] Types of sensor data collected by the EO data collectors 102 may include, for example, one or more of synthetic aperture radar (“SAR”) data, optical data, radio frequency (“RF”) data, and infrared (“IR”) data. Other types of sensor data may be collected.

[0050] Generally, any type of sensor data may be used if that type of sensor data includes the information required to perform the dark ship tracking techniques described herein (or that data can be processed to obtain such information). As will be understood, some types of sensor data collected by the EO data collectors 102 may include or be analyzed to determine kinematical information about the ships contained in the sensor data. Other types of sensor data may not be able to provide such information.

[0051] In particular, the EO data collector 102 collects sensor data of ships 104 in maritime or marine environments. The sensor data may be processed to obtain various information about the ships 104, such as a geolocation or position and kinematic information (e.g., heading, speed). The system 100 collects and processes at least one type of sensor data that includes kinematical information about the ships identified in the sensor data. Kinematical information may include, for example, a heading and a speed.

[0052] The system 100 further includes a ground terminal 106 for receiving the captured sensor data from the EO data collector 102. The ground terminal 106 may include a receiver and a computer system for processing received sensor data (e.g., raw sensor data). In some cases, the ground terminal 106 may process raw sensor data into a format suitable for ingestion or use by the computer system.

[0053] The system 100 includes a server system 108 and a user device 112. While a single server computer 108 and a single user device 112 are shown in Figure 1 , the number of servers 108 and user devices 112 may vary (e.g., multiple) and the number is not particularly limited.

[0054] The server system 108 communicates with the ground terminal 106 via a communication network 110. The network 110 may be a wide area network, such as the Internet. Communication in this context may include sending and receiving data. The server 108 receives sensor data from the ground terminal 106.

[0055] In other embodiments, the server 108 may receive the sensor data from a source or device other than the ground terminal 106. For example, in an embodiment, the server 108 may receive the sensor data from another computer or data storage device.

[0056] The server 108 communicates with the user device 1 12 via network 1 10.

[0057] The server 108 runs a dark ship tracker software application 120. The user device 112 communicates with the server system 108 over the network 110 and provides a user interface of the dark ship tracker application 120 for a user to request and review dark ship tracks automatically identified by the application 120.

[0058] According to various embodiments, the dark ship tracker application 120 is hosted by the server system 108, or is installed locally on the user device 112, or runs on both the server system 108 and the user device 112.

[0059] The user device 112 is configured to receive input from a user and display data generated by the server 108. The input data received from a user may be used to request certain data generated and stored by the server 108. The user device 112 is configured to display a graphical user interface that allows a user to interact with the server 108. The user interface may include a series of user interface screens for receiving user input and displaying output data generated by the server 108.

[0060] Referring now to Figure 2, shown therein is a computer system 200 for dark ship tracking, according to an embodiment. The computer system 200 may be implemented using the server system 108 and the user device 112 of Figure 1 .

[0061] The system 200 includes a memory 202 and a processor 204 in communication with the memory 202.

[0062] The system 200 includes a communication interface 206 for transmitting and receiving data. The communication interface 206 may include a network interface.

[0063] The system 200 includes a display 208 for displaying data generated by the system 100. The display 108 may be located at a user device of the system 200, such as user device 112.

[0064] The system 200 includes an input device for providing input data to the system 200 by a user, such as through a graphical user interface. The input device 200 may include a pointing device (e.g., a mouse), a keypad, or the like.

[0065] The system 200 further includes dark ship detection input data 212. The dark ship detection input data 212 may be received from one or more external sources or systems, such as through a network (e.g., network 110 of Figure 1 ). In such cases, the dark ship detection input data 212 is received via communication interface 206. In other embodiments, the dark ship detection input data 212 may be generated or processed by a separate application run by the computer system 200 that is in communication with the dark ship tracker application 120 or by the dark ship tracker application itself.

[0066] The memory 202 stores dark ship detections data 214 including a plurality of dark ship detections. A dark ship detection 214 is an instance of a dark ship that has been detected in sensor data.

[0067] Each dark ship detection 214 includes certain metadata associated with the detection including at least a sensor type (i.e. , the type of sensor used), a timestamp of the sensor data acquisition, and a geolocation of the detected dark ship (e.g., geocoordinates defining a position). Generally, such metadata is either already present in the dark ship detection input data 212 or may be obtained from the dark ship detection input data 212 by the dark ship tracker application 120.

[0068] Depending on the type of sensor data, the detection may include kinematical information about the detected dark ship. Kinematical information may include a heading and a speed. Some types of sensor data may provide no heading, speed, or other kinematical information.

[0069] In some cases, the dark ship tracker application 120 generates dark ship detection data 214 by processing or formatting processes the dark ship detection input data 212.

[0070] Kinematic information (e.g., speed) may be obtained from various sensors. Vessel speed may be inferred from the length of a vessel. Accordingly, in general, if the sensor type is one from which vessel length can be obtained (e.g., optical data, others), then a vessel speed can be inferred using the obtained vessel length.

[0071] Referring now to Figure 3, shown therein is example dark ship detection data 214, according to an embodiment.

[0072] Detection data 214 includes dark ship detections 302, 304, 306, 308, 310, 312.

[0073] Detection data 214 is represented as a plurality of attributes. The attributes may be represented or stored as key-value pairs. In other embodiments, any suitable manner of representing the data may be used.

[0074] Example attributes include a detection identifier 314, a sensor type 316, a timestamp 318, a geolocation 320, a heading 322, and a speed 324. In some cases,detection data 214 may also include the sensor data (raw or processed) from which the detection was made.

[0075] Detection identifier 314 may be an alphanumerical string or the like that is assigned to and uniquely identifies a given detection.

[0076] Sensor type 316 is data specifying a type of sensor that was used to make the detection (and corresponds to the type of sensor data used to make the detection). Examples include SAR, optical, RF, and IR. Other types of sensors may also be used.

[0077] Timestamp 318 is data specifying a time at which the sensor data in which the detection appears was collected. The timestamp 318 thus also represents the time of detection. The timestamp 318 is provided along with the sensor data (i.e., as metadata).

[0078] Geolocation 320 is data specifying a geographic location at which the vessel that has been detected is located. The geolocation 320 is determined and provided as metadata of the sensor data. The geolocation 320 may be represented as latitude and longitude coordinates.

[0079] Heading 322 is data specifying a heading of the detected vessel. The heading 322 is the direction in which a vehicle / vessel is pointing at the moment of the sensor data collection (i.e., the timestamp 318). The heading 322 may be determined from the sensor data. The heading 322 may be provided to the system from the sensor data source (e.g., along with the sensor data) or may be determined by the system by processing the sensor data. Heading may be obtained from SAR or optical images by measuring which way the vessel is oriented relative to North. It should be noted that in certain SAR or optical detections (for example SAR detection 604), the orientation of the ship can be determined exactly. For example, the superstructure of larger ships is usually located at the stem. If the superstructure is easily visible in the image data, then the bow and stem can be identified, and the heading inferred from that information. Additionally, if there is a wake visible in the image, this information can be used to infer the heading of the ship. In other types of SAR or optical detections, the superstructure may not be visible or readily identifiable, which produces a 180-degree ambiguity in the heading.

[0080] Speed 324 is data specifying a speed of the detected vessel at the time the sensor data in which the detection was made was collected (i.e., timestamp 318). The speed 324 may be determined from the sensor data. The speed 324 may be provided to the system from the sensor data source (e.g., along with the sensor data) or may be determined by the system by processing the sensor data.

[0081] For SAR, speed may be obtained in different ways.

[0082] In a first approach, a moving ship in SAR data will often appear as though it is displaced from its wake. A formula can be used to estimate the speed of the ship based on the displacement from the wake and other imaging parameters.

[0083] In a second approach, signal processing methods on a SLC (Single Look Complex) image can be used to obtain the speed.

[0084] The speed of a ship may also be inferred from the measured length. This may be done by performing a statistical analysis of all the ships in the open ocean, and seeing how fast ships of certain lengths typically travel. Such an approach may be crude and error prone but can be implemented.

[0085] In some embodiments, a speed may be obtained from RF data and thus provided with detection data 214 of an RF detection.

[0086] Heading 322 and speed 324 represent kinematical information of kinematical data of the detected vessel. In other embodiments, additional or other types of kinematical information may also be included in detection 214 data and, in some cases, used in the performance of dark ship tracking. In other embodiments, Course over Ground maybe be used instead of heading provided Course over Ground can be obtained from the imaging sensor data.

[0087] Kinematical information, such as heading 322 and speed 324, may not be available or determinable from some types of sensor data 316 supported by the system 200. For example, RF and IR sensor types may not provide heading and speed and thus detections of those sensor types (e.g., 304, 308) may not have such data available.

[0088] Referring again to Figure 2, the processor 204 is configured to execute dark ship tracker application 120 (e.g., dark ship tracker application 120 of Figure 1 ). Aspreviously noted, modules and components of dark ship tracker application 120 may be implemented or executed at or across multiple computing devices (e.g., networked computer devices).

[0089] Generally, the dark ship tracker application 120 operates on dark ship detections data 214 to determine which dark ship detections should be combined into a dark track. The dark track may indicate a plurality of geolocations occupied by the vessel (at various times) and a path followed by the vessel. In some cases, the dark track data may be further processed or analyzed to project a position or path of the vessel beyond what is found in the dark track.

[0090] The dark ship tracker application 120 includes a time comparator module 220, a location projector module 224, a match feasibility module 228, a dark vessel track generator module 230, a track visualization generator module 232, and a graphical user interface module 234.

[0091] The time comparator module 220 compares timestamps 318 from two detections 214 to determine whether the timestamps satisfy certain timing criteria (e.g., T 1 is later than T2, T2 is later than T 1 , etc.). In one example, the time comparator module 220 may determine whether a first timestamp is later than a second timestamp. In other embodiments, the comparison criteria may be more complex.

[0092] The time comparator module 220 may output a binary output indicating whether the timing criteria was satisfied by the timestamp comparison or not. In some cases, the output may determine whether further modules are executed.

[0093] The location projector module 224 projects the location of a vessel in a first detection to a second, projected location 216 at the timestamp 318 of another detection. In other words, the projected location 216 is a prediction of where the vessel would be at the timestamp 318 of another detection. The projected location 216 is stored in memory 202. The projected location of a detection 214 may be stored as part of or in association with the detection 214 from which the projection was made. The projected location 216 may be associated with a timestamp that is the same as the timestamp used to make the projection.

[0094] In some cases, the projection may be performed bidirectionally (e.g., 180 degrees from one another). In some cases, a detection 214 with a sensor type 316 that has a 180 degree ambiguity (e.g., SAR detection) is projected in two directions corresponding to the 180 degree ambiguity. The bidirectional projection may be performed automatically upon determining the sensor type 316 of the detection 214 satisfies the relevant criteria for bidirectional projection. Accordingly, in some cases, the vessel geolocation projection 216 may include two projected geolocations.

[0095] In an embodiment, the location projector module 224 projects the location of the detection 214 to a projected location 218 using the heading 322 and speed 324 data of that detection and the timestamp 318 of another detection 214 (that is being projected to).

[0096] In an embodiment, projection is performed via a dead reckoning technique (such as is used in marine navigation). Once a timestamp difference between two detections is determined, the projection module 224 uses the speed and heading information 324, 322 to infer the new (projected) location 218 of the ship. It assumes that ships travel along great circle routes (geodetics). The great circle route is computed based on the current location of the ship, and the direction in which the ship is headed. The projected location 218 is then computed by seeing how far the ship has traveled along the route with the speed 324 that was provided.

[0097] The location projector module 224 is configured to project only from detections 214 that have heading 322 and speed 324 data available. For sensor types that do not have heading 322 and speed 324 available, the location projector module 224 only projects to those detections and not from those detections.

[0098] In some cases, the location project module 224 is executed only upon the time comparator module 220 indicating the timing criteria has been satisfied.

[0099] The location projection module 224 outputs at least one projected geolocation 216 for the detection. The outputted projected geolocation 216 is provided as input to the match feasibility determinator module 228.

[0100] The match feasibility determinator module 228 determines whether the detection that was projected and the detection being projected to are the same dark vessel (i.e. , are to be treated as the same dark vessel for dark track formation).

[0101] In an embodiment, the match feasibility determinator module 228 receives the location of the projected-to detection and the projected location 216 of the projected- from detection as input and assigns a probabilistic value reflecting a match probability 218. The match probability may be a correlation confidence. The match probability 218 is stored in memory 202.

[0102] Generally, if the location of the predicted-to contact / detection and location of the contact / detection on which the prediction is being made to are spatially close together, the correlation confidence will be higher. If the locations are further apart, the correlation confidence will be lower. The correlation confidence may be represented as a decimal value ranging from 0 (indicating very low confidence) to 1 (indicating very high confidence) that two contacts correlate. The calculation of correlation confidence takes into account other contacts in the area. For example, if there is a very busy area with lots of ships in close proximity to each other, the correlation confidence for correlated contacts will typically be lower than for cases where we have just one detection. This is done to account for the fact that the correlator module 228 could have erroneously assigned a correlation in a busy area.

[0103] Referring now to Figure 7, shown therein is a diagram 700 illustrating variation in correlation confidence determined by the match feasibility determinator module 228 using three example scenarios 702, 704, 706, according to an embodiment.

[0104] In the first example 702, there is a location of an RF detection 708 and a predicted or projected location of a SAR detection 710. A location uncertainty cone 712 of the SAR detection is also identified. The predicted location 710 of the SAR detection is in close proximity to the location 708 of the RF detection. Given the close proximity of the contacts, the correlation confidence value is high (0.99).

[0105] In the second example 704, the projected location of the SAR detection 710 is in further proximity to the location of the RF detection 708. The correlation confidence is lower than in first example 702 but remain relatively high (0.73).

[0106] In the third example 706, there are competing correlations. Third example 706 includes first RF detection 708, first SAR detection 710, second RF detection 714, and second SAR detection 716. Second SAR detection 716 has a location uncertainty cone 718. In this example, there are competing correlations. First RF detection 708 and projected location of first SAR detection 710 have a determined correlation confidence of 0.57 and second RF detection 714 and projected location of second SAR detection 716 have a determined correlation confidence of 0.58. There is a potential for erroneous correlations as uncertainties 712, 718 overlap and there are multiple detections in close proximity. In other words, while locations 708, 710 are in close proximity and locations 714, 716 are in close proximity, their respective correlation confidences are lower compared to the first example 702 given the overlapping uncertainties 712, 718.

[0107] The match feasibility determinator module 228 then compares the determined match probability 218 to a predetermined match probability threshold 220. The match probability threshold 220 is stored in memory 202. In some embodiments, the match probability threshold 220 may be user configurable, such as by providing input through a graphical user interface of the system 200. In other embodiments, the match probability threshold 220 may be pre-set and non-user-configurable. In an example, the correlation confidence threshold may be set to N (e.g., 0.01 ), with any targets that have a correlation confidence lower than N (<0.01 ) being deemed to be not correlated.

[0108] In an embodiment, if the match probability 218 exceeds the match probability threshold 220, the match feasibility determinator module 228 generates an output indicating a match between the detections. If the match probability 218 does not exceed the threshold 220, the module 228 generates an output indicating a non-match. The output may be a binary output indicating match / no match. In other embodiments, the match probability 218 may be presented to a user through a graphical user interface and the user selects via the user interface whether the detections are a match. In yet other embodiments, multiple thresholds may be used. For example, a first, higher threshold, if met, may result in an automatic determination of a match, while a second, lower threshold, if met, may result in the detections being presented to a user via user interface as a potential or probable match for user review and confirmation. Failure to meet the secondthreshold may result in the match feasibility determinator module 228 automatically rejecting the match (i.e. , output of non-match).

[0109] The dark vessel track generator module 230 generates a dark vessel track 222. This may include the generation of a new dark vessel track 222 (e.g., when a first match between two detections 214 is identified) or the addition to or extension of an existing track 222 (e.g., when it is determined that there is a match between a new detection and a detection that is already part of a dark track 222).

[0110] Generally, the dark vessel track generator 230 may execute in response to a match output being output by the match feasibility determinator 228.

[0111] The dark vessel track 222 is stored in memory 202. The dark vessel track 222 may be a data object that includes or specifies a plurality of vessel positions, with each vessel position including a geolocation 320 and a corresponding timestamp 318. Each vessel position in the dark vessel track 222 may include or have associated therewith a detection identifier for identifying the detection 214 that was the origin of the vessel position. The detection identifier may be the detection identifier 314.

[0112] In the case of a match determined by the match feasibility determinator 228 between two detections 214, the geolocations 320 and timestamps 318 of the matching detections 214 are added as vessel positions to the vessel track 222. If one of the detections 214 in the match is already part of an existing dark vessel track 222, the dark vessel track generator 230 may add the other detection 214 in the match to the existing dark vessel track 222. In doing so, the dark vessel track generator 230 may reference the detection identifiers 314 of the matching detections 214 to determine if either is already part of an existing dark vessel track 222.

[0113] Over time and as more detections 214 are obtained, received, or processed, existing dark vessel tracks 222 may grow to include additional positions and new dark vessel tracks 222 may be formed.

[0114] Generally, the dark vessel track 222 may be configured and stored such that the information contained therein can be used by the processor 204 to generate a dark track visualization.

[0115] The dark vessel track 222 may include one or more identifiers (e.g., dark track identifier, vessel identifier) that uniquely identify the dark track 222 among other dark tracks stored in memory 202.

[0116] The track visualization generator module 232 generates a dark vessel track visualization 232 based on the dark vessel track 222. The dark vessel track visualization 232 is configured to display in a user interface of the system 200 as a human readable visualization.

[0117] The dark track visualization 232 includes at least visualizations of one or more dark vessel tracks 222. A dark track 222 may be visualized as a plurality of nodes each representing a position of the vessel and one or more edges connecting the plurality of nodes. Edges may be straight or curved.

[0118] In some embodiments, the visualization 232 may include geographical map data of an area or region with the visualized dark track overlaid a visualization (or map) of the area of region. This enables a viewer to view the dark track in the context of a geographical map. In some cases, the area or region in the visualization 232 may be determined by the geolocations in the dark track 222. In other embodiments, the area or region may be selected by a user, such as by providing a user input to the system 200 via a user interface. Generally, a dark track visualization 232 will include at least one node of the dark track 222.

[0119] In some cases, the dark track visualization 232 may include other detections 214 that are not part of a dark track 222. This may provide context for the area of the dark track. Such other detections 214 may be plotted or otherwise displayed based on their geolocation data 320.

[0120] The graphical user interface module 234 generates a graphical user interface for the system 200. The user interface enables a user to interact with the system 200, such as by displaying an output of the dark ship tracker application 120 (e.g., visualized dark tracks) and receiving input from the user which may affect the data displayed and the operation of the application 120.

[0121] The graphical user interface may include a plurality of user interface screens.

[0122] The graphical user interface displays the dark vessel track visualization 232. The graphical user interface may display additional information.

[0123] The graphical user interface module 234 may enable the user to interact with the dark track visualization 232 through providing an input (e.g., through input device 210). This may include, for example, selecting geographic locations or dark vessel tracks 222 to view. In some cases, visual components of the dark track visualization 232 may be selectable to display a subset of data stored by the application 120. This may include, for example, information about the dark track 222, dark ship detections 214, and the like. In an example, nodes in the dark track visualization 232 may be selectable to display information about the detection corresponding to the node. Such information may include detection data 214 associated with the node. It should be noted that selecting in this context may include selecting a visual component through clicking a pointer device, hovering a pointer over the visual component, or any other suitable mechanism of identifying a visual component.

[0124] In some embodiments, the graphical user interface may enable a user to interact with the dark track generation process, such as by verifying or confirming whether detections are matches. This may include, for example, displaying a match probability or other output of the match feasibility determinator 228, and possibly detection data 214 associated with the detections being analyzed.

[0125] Figure 4 illustrates and example user interface screen 400 generated by the graphical user interface module 234, according to an embodiment.

[0126] The user interface screen 400 may be displayed at the user device 112 of Figure 1 .

[0127] The user interface screen 400 includes a dark track visualization 232. The user interface screen also includes other detections 402, which are not part of the visualized dark track.

[0128] The dark track visualization 232 includes nodes 404, 406, 408, 410 and edges 412, 414, 416. Nodes represent dark ship detections 214 that have been identified through the dark ship tracker application 120 as part of the same dark track 222. Accordingly, each node includes a corresponding geolocation and timestamp.

[0129] As noted, each node represents a different dark ship detection 214 and thus has various metadata associated therewith. This may include, for example data elements 314-324 of Figure 3.

[0130] The user interface screen 400 also includes a pop up window 418. The pop up window 418 is displayed in response to a user selecting node 408. The pop up window 418 displays detection data 214 associated with node 408. Other nodes 404, 406, 410 in the dark track visualization 232 and other detections 402 are similarly selectable to display their respective detection data 214. The pop up window 418 may include a selectable “hide historical track” or “hide future track” feature (as in Figure 4). Selecting a III icon or element associated with such feature may cause any position reports that are part of the selected detection’s track, and that precede or superseded the timestamp of the selected detection 418, to not be displayed on the visualization 432.

[0131] In other embodiments, techniques of displaying data other than a popup window in response to a user action may be used.

[0132] Referring now to Figure 5, shown therein is a method of dark ship tracking 500, according to an embodiment. The method 500 may be encoded as computerexecutable instructions which, when executed by a processor, cause the processor to perform the method 500.

[0133] The method 500 may be implemented using the computer system 200 of Figure 2 and, in particular, by the dark ship tracker application 120.

[0134] At 502, the method 500 includes storing in a data storage device a plurality of dark ship detections including a first ship detection and a second ship detection. Each ship detection includes an associated sensor type, a geolocation, and a timestamp.

[0135] First ship detection also includes kinematical information associated with the detected ship. The kinematical information may include, for example, a heading anda speed. The sensor type of the first ship detection is thus a sensor type from which such kinematical information can be obtained. The sensor type of the second ship detection may be a sensor type from which kinematical information cannot be obtained.

[0136] For the purposes of this description, first ship detection includes geolocation P1 and timestamp T1 , indicating that the first ship was detected at geolocation P1 at timestamp T1 , and second ship detection includes geolocation P2 and timestamp T2, indicating that the second ship was detected at geolocation P2 at timestamp T2.

[0137] In a particular example, the first ship detection may be a detection that has previously been received and stored by the dark ship tracker application and the second detection is a new detection. In this case, a new detection is a detection that has more recently been received by the dark ship tracker application and does not necessarily imply anything about the relative time at which the detections were made (i.e. , the time at which the respective sensor data was collected).

[0138] At 504, the method 500 includes determining that the timestamp T2 of the second ship detection is later than the timestamp T1 of the first ship detection.

[0139] At 506, the method 500 includes projecting the first ship detection to a projected second geolocation P2' at the timestamp T2 of the second ship detection using the kinematical information of the first ship detection.

[0140] At 508, the method 500 includes determining that the projected second geolocation P2' of the first ship detection is within a threshold proximity of the geolocation P2 of the second ship detection. Such determination indicates that the dark ship in the first ship detection and the dark ship in the second ship detection are likely to be the same dark vessel. In some cases, probabilistic techniques are used to determine a likelihood of the first and second ship detections being a match. The likelihood may be represented as a probabilistic value or correlation confidence. In such cases, the probabilistic value is compared to a threshold and where the probabilistic value is greater than the threshold, the first and second detections are deemed a match.

[0141] At 510, the method 500 includes forming a vessel track that has a geolocation P1 at timestamp T 1 and a geolocation P2 at timestamp T2, which correspond to the first and second detections, respectively.

[0142] At 512, the method 500 includes storing the dark vessel track in the data storage device as a dark vessel track. The dark vessel track may subsequently be added on to with one or more further dark ship detections. The dark vessel track may include further information about each dark ship detection included therein. The dark vessel track may be stored such that the data contained therein can be processed by the dark ship tracker application to obtain a visualization of the dark track suitable display and review in a graphical user interface.

[0143] It will be understood that the method 500 may be carried out multiple times to obtain a dark vessel track that includes greater than two known positions (geolocation and timestamp). It will also be understood that, in practice, dark ship detections will be processed using method 500 and not all will be a match. In such cases, determinations at 504 or 508 may fail, resulting in a failure to form a dark vessel track between the detections.

[0144] Further, in practice, rules encoded as computer-readable instructions may be used to determine which dark ship detections are compared using method 500. For example, in an embodiment, each time a new detection is received by the dark ship application, that new detection may be compared to at least one other detection to determine if a dark track between the detections should be formed. The at least one other detection may or may not be part of an existing dark vessel track.

[0145] Referring now to Figure 6, shown therein is a visualization 600 of an example dark vessel track 602 generated by the systems and methods of the present disclosure, according to an embodiment.

[0146] The dark track 602 may be generated using the computer system 200 of Figure 2, and in particular by dark ship tracker application 120. The dark track 602 may be generated by performing method 500 multiple times.

[0147] The visualization 600 may be the dark track visualization 232 of Figure 2.

[0148] In this example, the dark track 602 includes a first dark SAR detection 604, a dark RF detection 606, a dark optical detection 608, a dark IR detection 610, and a second dark SAR detection 612. Each detection includes a geolocation and a corresponding timestamp.

[0149] The dark ship tracker application 120 determines that the timestamp of the RF detection 606 is later than the timestamp of the first SAR detection 604. The geolocation of the first SAR detection 602 is projected (predicted) to the timestamp of the RF detection 606 using the measured heading and speed of the detected vessel in the first SAR detection 604. The projected geolocation of the vessel in the dark SAR detection 604 is determined by the dark ship application 120 to be within close proximity to the RF detection 606. The first SAR detection 602 and RF detection are formed into a track (linked). The track is represented by edge 614 connecting detections 604 and 606.

[0150] The dark ship tracker application 120 compares the timestamp of the optical detection 608 to the timestamp of the RF detection 606 and determines that the timestamp of the optical detection 608 is later. The geolocation of the optical detection 608 is projected to the timestamp of the RF detection 606 using the measured heading and speed of the detected vessel in the optical detection 608. The predicted geolocation of the vessel in the optical detection 608 is determined to be within close proximity to the RF detection 606. The RF detection 606 and the optical detection 608 are formed into a track. The track is represented by edge 616 connecting detections 606 and 608.

[0151] Given both tracks 614, 616 include RF detection 606, detections 604, 606, 608 and edges 614, 616 form a dark track. Edges 614 and 616 may be considered and referred to as dark track segments.

[0152] The dark ship tracker application 120 compares the timestamp of the IR detection 610 to the timestamp of the optical detection 608 and determines that the timestamp of the IR detection 610 is later. The geolocation of the optical detection 608 is projected to the timestamp of the IR detection 610 using the measured heading and speed of the detected vessel in the optical detection 608. The predicted geolocation of the vessel in the optical detection 608 is determined to be within close proximity to the IR detection610. The optical detection 608 and the IR detection 610 are formed into a track. The track is represented by edge 618 connecting detections 608 and 610.

[0153] Given both dark track 614, 616 and dark track 618 include optical detection 608, detections 604, 606, 608, 610 and edges 614, 616, 618 form a dark track. Edge 618 may be considered and referred to as a dark track segment.

[0154] The dark ship tracker application 120 compares the timestamp of the IR detection 610 to the timestamp of the second SAR detection 612 and determines that the timestamp of the second SAR detection 612 is later. The geolocation of the second SAR detection 612 is projected to the timestamp of the IR detection 610 using the measured heading and speed of the detected vessel in the second SAR detection 612. The predicted geolocation of the vessel in the second SAR detection 612 is determined to be within close proximity to the IR detection 610. The second SAR detection 612 and the IR detection 610 are formed into a track. The track is represented by edge 620 connecting detections 610 and 612.

[0155] Given both dark track 614, 616, 618 and dark track 620 include IR detection 610, detections 604, 606, 608, 610, 612 and edges 614, 616, 618, 620 form a dark track. Edge 620 may be considered and referred to as a dark track segment.

[0156] It should be noted that in the visualization 600, detections with sensor types that provide no kinematical information are represented by a circle. These detections can only be projected to (i.e., their timestamps used to project another detection with kinematical information). Such detections include RF detection 606 and IR detection 610. Detections with sensor types that provide kinematical information and have a known orientation are represented by a single headed arrow. These detections are projected in one direction (that of the known orientation) and thus can be projected to another detection (using the timestamp of the other detection), particularly one with no kinematical information. An example of such a detection is first SAR detection 604. Detections with sensor types that provide kinematical information but have an orientation with a 180 degree ambiguity are represented by a double sided arrow. These detections are projected in opposite directions (representing each of the two possible orientations) and thus can be projected to another detection (using the timestamp of the other detection),particularly one with no kinematical information. Examples of such detections include optical detection 608 and second SAR detection 612.

[0157] The visualization 600 also includes polygons 622, 624, 626, 628 (shown as hashed lines) for each track segment 614, 616, 618, 620, respectively. The polygons represent statistical uncertainty for the prediction. This may be similar to statistical uncertainty that is shown of predicted paths of hurricanes.

[0158] Referring again to Figure 2, in an embodiment, processing by the dark ship tracker application 120 starts as soon as a dark contact (i.e., detection) is presented to the system 200.

[0159] A first check is done to determine if there are any other dark contacts that are within a set time period of the currently-presented contact / detection.

[0160] In an embodiment, the set time period is 12 hours or less (e.g., prediction to a maximum time of 12 hours). A set time period of longer than 12 hours may reduce the accuracy of the predicted locations to an unacceptable level. The set time period may be present as part of the application 120 or may be user-configurable, such as through a user interface. In either case, the set time period is considered “predetermined”.

[0161] If there are dark contacts that are within the set time period of each other, then a projection of one to the other is made by the application 120 to see if the two dark contacts may be part of the same track.

[0162] The foregoing process is then repeated by the application 120 on a rolling basis, as new dark contacts are presented to the system 200.

[0163] While the above description provides examples of one or more apparatus, methods, or systems, it will be appreciated that other apparatus, methods, or systems may be within the scope of the claims as interpreted by one of skill in the art.

Claims

Claims:1 . A method of tracking a non-self-reporting vessel (“dark vessel”) with an unknown identity, comprising: providing a system configured to receive and store a plurality of ship detections of at least one sensor type, the plurality of ship detections including a first ship detection D1 having a sensor type, a first geolocation P1 , a timestamp T1 , and kinematical information associated therewith, the first ship detection D1 being non- self-reporting; receiving by the system a new ship detection D2 of the at least one sensor type, the new ship detection D2 having a sensor type, a geolocation P2, and a timestamp T2 associated therewith, the new ship detection D2 being non-self-reporting; comparing the timestamp T2 of the new detection D2 to the timestamp T1 of the first detection D1 and determining that the timestamp T2 of the new detection D2 is later than the timestamp T 1 of the first detection D1 ; using the kinematical information of the first ship detection D1 , projecting the first ship detection D1 to a projected second geolocation P2' at the timestamp T2 of the new ship detection D2; determining that the projected second geolocation P2' is within a threshold proximity of the geolocation P2 of the new detection, the threshold proximity stored by the system; forming a vessel track having a geolocation of P1 at timestamp T1 and a geolocation of P2 at timestamp T2; and storing the vessel track in the system as a dark vessel track.

2. The method of claim 1 , wherein the at least one sensor type includes any one or more of a radio frequency (RF) sensor, an infrared (IR) sensor, a synthetic aperture radar (SAR) sensor, and an optical sensor.

3. The method of claim 2, wherein the kinematical information includes a vessel heading and a vessel speed.

4. The method of claim 1 , wherein projecting the first ship detection D1 to projected second geolocation P2' is based on a geodetic path.

5. The method of claim 1 , wherein the sensor type of the new detection D2 is a sensor type from which speed and heading are not obtainable.

6. The method of claim 1 , wherein the method is performed only if a predetermined time period between the timestamps T1 , T2 of the first and new ship detections is not exceeded.

7. The method of claim 1 , wherein the sensor type of the first ship detection is a sensor type that produces a 180 degree orientation ambiguity in a heading of the first ship detection and projecting the first ship detection D1 to the projected second geolocation P2' is performed bidirectionally at 180 degrees apart.

8. The method of claim 1 , wherein determining that the projected second geolocation P2' is within a threshold proximity of the geolocation P2 of the new detection includes determining a correlation confidence between the projected second geolocation P2' and the geolocation P2 of the new detection, comparing the correlation confidence to a threshold correlation confidence, and determining that the correlation confidence exceeds the threshold correlation confidence.

9. The method of claim 1 , wherein projecting the first ship detection D1 to the projected second geolocation P2' uses a dead reckoning technique.

10. The method of claim 1 , further comprising displaying a visualization of the dark vessel track in a graphical user interface, the visualization including first and second nodes representing the first ship detection D1 and new ship detection D2, respectively, and an edge connecting the first and second nodes representing the dark vessel track.

11. The method of claim 10, wherein the first and second node are selectable in the graphical user interface to display information about the first ship detection D1 and the new ship detection D2, respectively, including the sensor type, the geolocation, and the timestamp.

12. The method of claim 1 , further comprising: receiving by the system a third ship detection D3 having a sensor type, a geolocation P3, a timestamp T3, and kinematical information associated therewith, the third ship detection D3 being non-self-reporting: comparing the timestamp T3 of the third detection D3 to the timestamp T2 of the new detection D2 and determining that the timestamp T2 of the new detection D2 is earlier than the timestamp T3 of the third detection D3; projecting the third ship detection D3 to a projected third geolocation P2" at the timestamp T2 of the new ship detection D2 using the kinematical information of the third ship detection D3; determining that the projected second geolocation P2" is within the threshold proximity of the geolocation P2 of the new detection; forming a second vessel track having a geolocation of P2 at timestamp T2 and a geolocation of P3 at timestamp T3; andstoring the second vessel track along with the first vessel track as part of the dark vessel track.

13. A system for dark ship tracking, comprising: a communication interface configured to receive a plurality of ship detections of at least one sensor type, including: a first ship detection D1 having a sensor type, a first geolocation P1 , a timestamp T1 , and kinematical information associated therewith, the first ship detection D1 being non-self-reporting; and a new detection D2 of the at least one sensor type, the new ship detection D2 having a sensor type, a geolocation P2, and a timestamp T2 associated therewith, the new ship detection D2 being non-self-reporting; a data storage device configured to store the plurality of ship detections; and a processor configured to: compare the timestamp T2 of the new detection D2 to the timestamp T1 of the first detection D1 and determine that the timestamp T2 of the new detection D2 is later than the timestamp T1 of the first detection D1 ; project the first ship detection D1 to a projected second geolocation P2' at the timestamp T2 of the new ship detection D2 using the kinematical information of the first ship detection D1 ; determine that the projected second geolocation P2' is within a threshold proximity of the geolocation P2 of the new detection, the threshold proximity stored by the system in the data storage device;form a first vessel track [segment] having a geolocation of P1 at timestamp T1 and a geolocation of P2 at timestamp T2; and store the first vessel track in the data storage device as a dark vessel track.

14. The system of claim 13, wherein the plurality of ship detections further includes a third ship detection D3 having a sensor type, a geolocation P3, a timestamp T3, and kinematical information associated therewith, the third ship detection D3 being non-self-reporting, and wherein the processor is further configured to: compare the timestamp T3 of the third detection D3 to the timestamp T1 of the new detection D2 and determine that the timestamp T2 of the new detection D2 is earlier than the timestamp T3 of the third detection D3; project the third ship detection D3 to a projected second geolocation P2" at the timestamp T2 of the new ship detection D2 using the kinematical information of the third ship detection D3; determine that the projected second geolocation P2" is within the threshold proximity of the geolocation P2 of the new detection; form a second vessel track having a geolocation of P2 at timestamp T2 and a geolocation of P3 at timestamp T3; and storing the second vessel track along with the first vessel track as part of the dark vessel track.

15. The system of claim 13, wherein the at least one sensor type includes any one or more of a radio frequency (RF) sensor, an infrared (IR) sensor, a synthetic aperture radar (SAR) sensor, and an optical sensor.

16. The system of claim 13, wherein the kinematical information includes a vessel heading and a vessel speed.

17. The system of claim 13, wherein projecting the first ship detection D1 to projected second geolocation P2' is based on a geodetic path.

18. The system of claim 13, wherein the sensor type of the new ship detection D2 is a sensor type from which speed and heading are not obtainable.

19. The system of claim 13, wherein the comparing, projecting, determining, forming, and storing are performed only if a predetermined time period between the timestamps T1 , T2 of the first and new ship detections is not exceeded.

20. The system of claim 13, wherein the sensor type of the first ship detection is a sensor type that produces a 180 degree orientation ambiguity in a heading of the first ship detection and projecting the first ship detection D1 to the projected second geolocation P2' is performed bidirectionally at 180 degrees apart.21 . The system of claim 13, wherein determining that the projected second geolocation P2' is within a threshold proximity of the geolocation P2 of the new detection includes determining a correlation confidence between the projected second geolocation P2' and the geolocation P2 of the new detection, comparing the correlation confidence to a threshold correlation confidence, and determining that the correlation confidence exceeds the threshold correlation confidence.

22. The system of claim 13, wherein projecting the first ship detection D1 to the projected second geolocation P2' uses a dead reckoning technique.

23. The system of claim 13, wherein the processor is further configured to display a visualization of the dark vessel track in a graphical user interface, the visualization including first and second nodes representing the first ship detection D1 and newship detection D2, respectively, and an edge connecting the first and second nodes representing the dark vessel track.

24. The system of claim 23, wherein the first and second node are selectable in the graphical user interface to display information about the first ship detection D1 and the new ship detection D2, respectively, including the sensor type, the geolocation, and the timestamp.

Images