Ship target detection system, ship target detection method, reliability estimation device, and program

Abstract

To provide a target detection system for a marine vessel which easily estimates existence reliability of a target.SOLUTION: A target detection system for a marine vessel comprises: a plurality of candidate data generation units each of which generates target candidate data including position data on target candidates existing in the surroundings of the marine vessel; an identical candidate selection unit which selects a plurality of pieces of target candidate data representing identical target candidates on the basis of the position data from the target candidate data generated by each of the plurality of candidate data generation units; and an existence reliability calculation unit which calculates existence reliability of the identical target candidates on the basis of attributes of the plurality of candidate data generation units which generate the plurality of pieces of selected target candidate data of the plurality of candidate data generation units.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to a ship target detection system, a ship target detection method, a reliability estimation device, and a program. [Background technology]

[0002] Patent document 1 discloses that the AIS device of a ship equipped with an ARPA radar device is connected to the ARPA radar device via an interface, and constantly compares the received dynamic information of nearby ships with the radar symbol information obtained from the ARPA radar device, and treats radar symbol information that does not have dynamic information with a specified match as transmission stop warning information, and dynamic information that does not have radar symbol information with a specified match as anonymous report warning information, and displays this warning information on a display device to alert the operator to caution, and also converts the warning information into a report and transmits it to nearby ships. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 3794641 Summary of the Invention [Problem to be solved by the invention]

[0004] By the way, for example, a radar detects ships, buoys, land, icebergs, floating containers, etc. around the ship as targets, but at the same time, it may also detect inappropriate targets such as noise.

[0005] The present invention has been made in consideration of the above-mentioned problems, and a main object of the present invention is to provide a ship target detection system, a ship target detection method, a reliability estimation device, and a program that make it easy to estimate the reliability of a target's presence. [Means for solving the problem]

[0006] In order to solve the above problems, a ship target detection system according to one aspect of the present invention includes a plurality of candidate data generation units, each of which generates target candidate data including position data of target candidates existing around a ship; a same candidate selection unit which selects a plurality of target candidate data representing a same target candidate from the target candidate data generated by each of the plurality of candidate data generation units based on the position data; and a presence reliability calculation unit which calculates a presence reliability of the same target candidate based on attributes of a plurality of candidate data generation units among the plurality of candidate data generation units which generated the selected plurality of target candidate data.

[0007] In addition, a ship target detection method according to another aspect of the present invention generates target candidate data including position data of target candidates existing around the ship, using each of a plurality of candidate data generation units, selects a plurality of target candidate data representing the same target candidate from the target candidate data generated by each of the plurality of candidate data generation units based on the position data, and calculates a presence reliability of the same target candidate based on attributes of a plurality of candidate data generation units among the plurality of candidate data generation units that generated the selected plurality of target candidate data.

[0008] Furthermore, a reliability estimation device according to another aspect of the present invention includes a candidate data acquisition unit that acquires target candidate data including position data of target candidates existing around the ship, generated by each of a plurality of candidate data generation units; a same candidate selection unit that selects a plurality of target candidate data representing a same target candidate from the target candidate data generated by each of the plurality of candidate data generation units based on the position data; and a presence reliability calculation unit that calculates presence reliability of the same target candidate based on attributes of a plurality of candidate data generation units among the plurality of candidate data generation units that generated the selected plurality of target candidate data.

[0009] In addition, a program according to another aspect of the present invention causes a computer to execute the following steps: acquire target candidate data including position data of target candidates existing around the ship, generated by each of a plurality of candidate data generation units; select a plurality of target candidate data representing the same target candidate from the target candidate data generated by each of the plurality of candidate data generation units based on the position data; and calculate a presence reliability of the same target candidate based on attributes of a plurality of candidate data generation units among the plurality of candidate data generation units that generated the selected plurality of target candidate data. Effect of the Invention

[0010] According to the present invention, it becomes easy to estimate the reliability of the presence of a target. [Brief description of the drawings]

[0011] [Figure 1] 1 is a diagram illustrating an example of the configuration of a ship target detection system according to an embodiment. [Diagram 2] FIG. 1 is a diagram illustrating an example of the configuration of a reliability estimation device according to an embodiment. [Diagram 3] FIG. 13 is a diagram illustrating an example of a radar candidate management DB. [Figure 4] A figure showing an example of an AIS candidate management DB. [Diagram 5] FIG. 13 is a diagram illustrating an example of a sonar candidate management DB. [Figure 6] FIG. 13 is a diagram illustrating an example of an image candidate management DB. [Figure 7] FIG. 13 is a diagram illustrating an example of a common candidate management DB. [Figure 8] 5 is a diagram showing an example of a procedure of a ship target detection method according to the embodiment; FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0013] 1 is a block diagram showing an example of the configuration of a ship target detection system 100. The ship target detection system 100 is a system that is mounted on a ship and detects targets that exist around the ship.

[0014] The ship target detection system 100 includes a reliability estimation device 1, multiple radars 2 and 3, a radar 3, an AIS 4, a sonar 5, an image sensor 6, a display device 7, a GNSS receiver 8, a gyrocompass 9, an ECDIS 10, a wireless communication unit 11, and an automatic steering device 12. These devices are connected to a network N such as a LAN, and are capable of network communication with each other.

[0015] The reliability estimation device 1 is a computer including a CPU, a RAM, a ROM, a non-volatile memory, an input / output interface, etc. The CPU of the reliability estimation device 1 executes information processing according to a program loaded from the ROM or the non-volatile memory to the RAM.

[0016] The program may be supplied via an information storage medium such as an optical disk or a memory card, or may be supplied via a communication network such as the Internet or a LAN.

[0017] In this embodiment, the reliability estimation device 1 is an independent device, but is not limited to this, and may be integrated with another device such as the ECDIS 10. In other words, the function of the reliability estimation device 1 may be incorporated into another device such as the ECDIS 10.

[0018] 2 is a block diagram showing an example of the configuration of the reliability estimation device 1. The reliability estimation device 1 includes a candidate data acquisition unit 21, a same candidate selection unit 22, a presence reliability calculation unit 23, a type reliability calculation unit 24, an object sign classification unit 25, an instruction reception unit 26, a collision avoidance calculation unit 27, and a ship control unit 28. These functional units are realized by the CPU of the reliability estimation device 1 executing information processing in accordance with a program.

[0019] The reliability estimation device 1 further includes a radar candidate management DB 31, an AIS candidate management DB 32, a sonar candidate management DB 33, an image candidate management DB 34, and a common candidate management DB 35. These databases are provided in a non-volatile memory of the reliability estimation device 1. Alternatively, the databases may be provided externally and accessed via a communication network.

[0020] Returning to the explanation of Fig. 1, the radars 2 and 3 emit microwaves from their antennas and receive the reflected waves, generating an echo image based on the received signals. The frequencies used by the radars 2 and 3 are, for example, X-band (8 to 12 GHz), S-band (2 to 4 GHz), or Ka-band (26 to 40 GHz). The radars 2 and 3 also detect target candidates appearing in the echo image, and generate target tracking data representing the positions and speed vectors of the target candidates.

[0021] The AIS (Automatic Identification System) 4 receives AIS data from other ships around the ship. The AIS 4 is an example of a communication device that receives other ship information. It is not limited to AIS, and a VDES (VHF Data Exchange System) may be used. The AIS data includes, for example, an identification code, a ship name, a position, a course, a ship speed, and a destination.

[0022] The sonar 5 emits ultrasonic waves in the surrounding direction from an ultrasonic transducer installed on the bottom of the ship, receives the reflected waves, and generates an echo image based on the received signal. The sonar 5 also detects target candidates appearing in the echo image and generates target tracking data indicating the positions of the target candidates. This function may be realized by the reliability estimation device 1.

[0023] The image sensor 6 is, for example, a camera such as a visible light camera or an infrared camera, and captures the outside of the ship to generate an image. The image sensor 6 is installed, for example, on the bridge of the ship, facing the bow direction. The image sensor 6 may be configured to pan, tilt, or zoom in response to a user's operation input.

[0024] The display device 7 is a display device with a touch sensor. The touch sensor detects a position on the screen pointed to by a finger or the like. The pointed position may be input by a trackball or the like, instead of a touch sensor. The display device 7 displays an image captured by the image sensor 6, or an echo image generated by the radars 2 and 3 or the sonar 5, etc.

[0025] The GNSS receiver 8 detects the position of the ship based on radio waves received from a Global Navigation Satellite System (GNSS). The gyrocompass 9 detects the heading of the ship. The gyrocompass is not limited to a gyrocompass, and a GPS compass or a magnetic compass may be used.

[0026] An ECDIS (Electronic Chart Display and Information System) 10 acquires the ship's position from a GNSS receiver 8 and displays the ship's position on an electronic chart. The ECDIS 10 also displays a planned route that is set to follow a number of waypoints in order on the electronic chart. A GNSS plotter may be used instead of an ECDIS.

[0027] The wireless communication unit 11 includes various wireless equipment for realizing communication with other ships or land base stations, such as wireless equipment for the VHF band, the MW band, and the HF band. The wireless communication unit 11 accesses, for example, a MICS (Maritime Information and Communication System) or the like to acquire data such as weather and tides.

[0028] The automatic steering device 12 calculates a target rudder angle for steering the bow to the planned course based on the bow direction obtained from the gyrocompass 9 and the planned course obtained from the ECDIS 10, and drives the steering gear so that the rudder angle of the steering gear approaches the target rudder angle. The automatic steering device 12 may also control the ship's engine.

[0029] By the way, the radars 2 and 3 detect ships, buoys, land, icebergs, floating containers, etc. around the ship as target candidates, but on the other hand, they may detect inappropriate targets such as noise as target candidates. Since it is pointless to perform avoidance based on such inappropriate targets, it is important to properly grasp the presence of targets. Furthermore, if it is possible to grasp not only the presence of targets but also the type of target, more efficient avoidance will be possible.

[0030] Therefore, in this embodiment, a plurality of candidate data generating units such as the radars 2 and 3, the AIS 4, and the sonar 5 are used to calculate the presence reliability and type reliability of the target candidates existing around the ship.

[0031] The radars 2 and 3, the AIS 4, and the sonar 5 are examples of candidate data generators that generate target candidate data. The target candidate data includes position data of the target candidates. Furthermore, the target candidate data may include type data of the target candidates.

[0032] Specifically, the radars 2 and 3 generate target tracking data representing the positions of target candidates as target candidate data. The AIS 4 generates AIS data representing the positions and types of other ships as target candidate data. The sonar 5 generates target tracking data representing the positions of target candidates as target candidate data.

[0033] The combination of the image sensor 6 and the object marker classifying unit 25 is also an example of a candidate data generating unit. The object marker classifying unit 25 identifies the object candidates in the image captured by the image sensor 6, thereby generating the target candidate data. The positions of the target candidates are calculated based on the capturing direction of the image sensor 6 and the positions of the target candidates in the image. The object marker classifying unit 25 also estimates the type of the target candidate.

[0034] The object identification unit 25 uses an image as input data and the range and type of targets in the image as teacher data to use a trained model generated in advance by machine learning to estimate the range and type of target candidates in the image captured by the image sensor 6. The trained model is an object detection model such as YOLO (You Only Look Once) or SSD (Single Shot MultiBox Detector).

[0035] The instruction receiving unit 26 that receives a user's instruction of a target candidate in an image displayed on the display device 7 is also an example of a candidate data generating unit. The target candidate data is generated when the user indicates a target candidate in an image captured by the image sensor 6 and displayed on the display device 7. Alternatively, the target candidate data may be generated when the user indicates a target candidate in an echo image generated by the radars 2 and 3 or the sonar 5 and displayed on the display device 7.

[0036] The reliability estimation device 1 may acquire target candidate data generated by a radar or the like on the other ship from the other ship via the wireless communication unit 11 and use the data to calculate the presence reliability, etc. In this case, the wireless communication unit 11 is an example of a communication device that receives other ship information and is an example of a candidate data generation unit.

[0037] In order to detect target candidates from multiple angles, the multiple candidate data generating unit preferably includes two or more types of radar, AIS, sonar, image sensor, etc. In addition, since radar is excellent at detecting target candidates, the multiple candidate data generating unit preferably includes at least one radar.

[0038] In addition, since radars differ in linearity and attenuation depending on the wavelength of radio waves, it is preferable that the multiple candidate data generation units include multiple radars using different frequency bands. However, the multiple candidate data generation units may be multiple radars using the same frequency band but with antennas installed in different locations.

[0039] Although the target candidate data generated by the radars 2 and 3 or the sonar 5 generally does not include type data, the target candidate data may include type data by combining with the object marker classifying unit 25. In other words, the object marker classifying unit 25 may estimate the type of the target candidate appearing in the echo image generated by the radars 2 and 3 or the sonar 5.

[0040] In this case, the object identification unit 25 uses the echo image as input data and the range and type of targets in the echo image as teacher data to use a trained model generated in advance by machine learning to estimate the range and type of target candidates in the echo image generated by the radars 2, 3 or the sonar 5.

[0041] 3 is a diagram showing an example of the radar candidate management DB 31. The radar candidate management DB 31 is a database for managing target candidate data generated by the radars 2 and 3. The radar candidate management DB 31 may be stored in the computers of the radars 2 and 3.

[0042] The radar candidate management DB31 includes fields such as "candidate identifier," "position," "type," "speed," "course," "trail," "time," "size," "strength," "attributes," and "instructions."

[0043] "Candidate identifier" is an identifier for identifying a target candidate. "Position" indicates the position of the target candidate. "Type" indicates the type of the target candidate. Specifically, "Type" is the type of the target candidate estimated from the echo image by the target identification unit 25.

[0044] "Speed", "Course", and "Track" respectively represent the speed, course, and track of the target candidate. "Time" represents the time elapsed since the detection of the target candidate. "Size" represents the size of the image appearing in the echo image. "Intensity" represents the signal strength of the reflected wave.

[0045] The "attribute" indicates which of the multiple radars 2 and 3 detected the object. The "instruction" indicates an instruction given by the user in the echo image displayed on the display device 7.

[0046] 4 is a diagram showing an example of the AIS candidate management DB 32. The AIS candidate management DB 32 is a database for managing target candidate data generated by the AIS 4. The AIS candidate management DB 32 may be stored in the computer of the AIS 4. The AIS candidate management DB 32 includes fields such as "candidate identifier", "position", "type", "ship name", "course", "speed", and "destination".

[0047] 5 is a diagram showing an example of the sonar candidate management DB 33. The sonar candidate management DB 33 is a database for managing target candidate data generated by the sonar 5. The sonar candidate management DB 33 may be stored in the computer of the sonar 5. The sonar candidate management DB 33 includes fields such as "candidate identifier", "position", "type", "time", "size", "strength", and "instructions".

[0048] 6 is a diagram showing an example of the image candidate management DB 34. The image candidate management DB 34 is a database for managing target candidate data generated by the target identification unit 25 identifying target candidates in an image captured by the image sensor 6. The image candidate management DB 34 includes fields such as "candidate identifier", "position", "type", "speed", "course", "trail", "time", "size", and "instructions".

[0049] 7 is a diagram showing an example of the common candidate management DB 35. The common candidate management DB 35 is a higher-level database for cross-managing the above-mentioned radar candidate management DB 31, AIS candidate management DB 32, sonar candidate management DB 33, and image candidate management DB 34. The common candidate management DB 35 includes fields such as "common candidate identifier", "candidate identifier", "position", "type", "attribute", "instruction", "existence reliability", and "type reliability", for example.

[0050] The "common candidate identifier" is an identifier for identifying a common target candidate. Candidate identifiers representing the same target candidate are grouped together using this common candidate identifier. The "position" indicates the position of the target candidate. Whether or not they represent the same target candidate is determined based on whether or not they have a common position. The "type" indicates the type of the target candidate.

[0051] "Attribute" indicates the attribute of the candidate data generation unit that generated the target candidate data. In other words, "attribute" indicates whether the target candidate data was detected by the radar 2, 3, the AIS 4, the sonar 5, or the image sensor 6. "Existence reliability" and "type reliability" indicate the existence reliability and type reliability of the target candidate (details will be described later).

[0052] 8 is a flow chart showing an example of the procedure of a ship target detection method realized in the ship target detection system 100. The reliability estimation device 1 executes the information processing shown in the figure in accordance with a program.

[0053] First, the reliability estimation device 1 collects target candidate data generated by each of a plurality of candidate data generating units such as the radar 2, 3, the AIS 4, the sonar 5, and the image sensor 6 (S11: processing as the candidate data acquiring unit 21). Specifically, the reliability estimation device 1 integrates the target candidate data managed in each of the radar candidate management DB 31, the AIS candidate management DB 32, the sonar candidate management DB 33, and the image candidate management DB 34 into a common candidate management DB 35.

[0054] Next, the reliability estimation device 1 selects multiple target candidate data representing the same target candidate based on the position data included in the target candidate data (S12: processing as the common candidate selection unit 22). Specifically, the reliability estimation device 1 selects multiple target candidate data with a common position as representing the same target candidate from among the many target candidate data integrated in the common candidate management DB 35, and assigns a common candidate identifier to the target candidate data. The reliability estimation device 1 also assigns a common candidate identifier to a single target candidate data for which there is no other target candidate data with a common position.

[0055] Next, the reliability estimation device 1 calculates the presence reliability of the target candidate based on the attributes of the candidate data generation unit that generated the target candidate data (S13: processing as the presence reliability calculation unit 23). That is, the presence reliability is calculated depending on which of the radars 2 and 3, the AIS 4, the sonar 5, the image sensor 6, etc., generated the target candidate data. The calculated presence reliability is stored in the common candidate management DB 35.

[0056] When multiple target candidate data belong to one common candidate identifier, the reliability estimation device 1 calculates the presence reliability based on the attributes of multiple candidate data generation units that generated the target candidate data. A weight representing the reliability is assigned to each of the radars 2 and 3, the AIS 4, the sonar 5, and the image sensor 6, and the presence reliability is calculated by adding up the weights.

[0057] For example, since the target candidate data (AIS data) generated by the AIS 4 is data relating to other ships received from other ships, it is preferable to set the weight of the AIS 4 higher than the weights of other candidate data generating units such as the radars 2 and 3.

[0058] In addition, since the target candidate data designated by the user in the image displayed on the display device 7 is a target actually recognized by the user, it is preferable to set the weight of the user's designation higher than the weights of other candidate data generation units.

[0059] Furthermore, when the frequency bands used by the radars 2 and 3 are different from each other, the shorter the wavelength, the more linear and directional the radio waves are, and it is also advantageous in terms of gain, so it is preferable to set the weight of the X-band radar higher than the weight of the S-band radar, for example.

[0060] In addition, when the antennas of radars 2 and 3 are installed in different positions, it is preferable to set the weight of the radar higher the antenna is installed, since there are fewer obstructions on the ship the higher the antenna is installed, such as on the main mast.

[0061] Note that the presence reliability may be adjusted based not only on the attributes of the candidate data generating unit but also on the speed, track, time, size, strength, etc. of the target candidate. For example, the presence reliability may be adjusted to be higher as the size of the target candidate increases, the moving speed increases, the time elapsed since detection increases, or the detection signal strength increases.

[0062] In addition, when a target candidate is located on the planned route of the ship, the presence reliability may be adjusted to be higher than when the target candidate is not located on the planned route.

[0063] Furthermore, the reliability of presence may be adjusted based on the nautical chart, weather, time, tides, etc. For example, when the target candidate is in an unnavigable area such as land or a seaweed net on the nautical chart, the reliability of presence may be adjusted to be high, and the weight of the image sensor may be lowered and the weight of the radar may be increased in rainy weather or at night.

[0064] Next, the reliability estimation device 1 calculates the type reliability of the target candidate based on the type data included in the target candidate data and the attributes of the candidate data generation unit that generated the target candidate data (S14: processing as the type reliability calculation unit 24). That is, the type reliability is calculated according to the content of the type data and whether the type data was generated by the radar 2, 3, the AIS 4, the sonar 5, the image sensor 6, etc. The calculated type reliability is stored in the common candidate management DB 35.

[0065] For example, since the target candidate data (AIS data) generated by the AIS 4 includes the identification codes of other ships as type data, it is preferable that the type reliability is evaluated to be higher than that of other candidate data generating units such as the radars 2 and 3.

[0066] The classification data of the radars 2, 3, the sonar 5, and the image sensor 6 are the types of target candidates estimated by the object classification unit 25. The object classification unit 25 may estimate the type of vessel related to size, such as a small vessel, a medium vessel, a large vessel, or a super large vessel, or may estimate the type of vessel related to type, such as a pleasure vessel, a fishing vessel, a commercial vessel, or a tanker vessel. The object classification unit 25 may also estimate the type of floating object on the water, such as a fishing net, driftwood, a person, drift ice, or a floating container.

[0067] In addition to the type data included in the target candidate data and the attributes of the candidate data generator, the type reliability may be adjusted based on the target candidate's track, position on the nautical chart, or tidal current. For example, if a target candidate enters or leaves a fishing port, is wandering, or is unnaturally going against the tidal current, it is highly likely to be a fishing boat, so the type may be determined to be a fishing boat. In addition, if the target candidate's course and destination are significantly different, the type reliability of AIS4 may be lowered.

[0068] Next, the reliability estimation device 1 performs calculations using a collision avoidance maneuvering algorithm based on the target candidate data, the presence reliability, and the type reliability stored in the common candidate management DB 35, and calculates a collision avoidance planned route (S15: processing by the collision avoidance calculation unit 27).

[0069] If the calculation by the collision avoidance maneuvering algorithm indicates that collision avoidance is necessary (S16: YES), the reliability estimation device 1 performs collision avoidance control using the automatic steering device 12 so that the ship navigates along the planned collision avoidance route (S17: processing as the ship control unit 28).

[0070] According to the embodiment described above, it becomes possible to appropriately grasp the presence of targets and perform avoidance by calculating the presence reliability of target candidates existing around the ship using multiple candidate data generating units such as the radars 2 and 3, the AIS 4, and the sonar 5. Furthermore, by calculating the type reliability of the target candidates, more efficient avoidance becomes possible.

[0071] In addition, if target candidate data is obtained individually from multiple candidate data generating units such as the radars 2 and 3, the AIS 4, and the sonar 5, a person must ultimately judge the reliability, and the reliability may vary depending on the situation. Therefore, as in this embodiment, by indexing the target candidate data obtained from multiple candidate data generating units by comprehensively considering attributes, etc., it becomes possible to provide effective support to people and collision avoidance systems.

[0072] Although the embodiment of the present invention has been described above, the present invention is not limited to the above-described embodiment, and it is needless to say that various modifications can be made by those skilled in the art.

[0073] In the above embodiment, the presence reliability and type reliability calculated by the reliability estimation device 1 are used for avoidance purposes, but the present invention is not limited to this and may be used for watch-keeping purposes. Specifically, the reliability estimation device 1 may change the display mode of the target candidate data according to the presence reliability and type reliability to facilitate watch-keeping.

[0074] For example, the higher the reliability of the presence, the brighter the symbol of the target candidate may be, or the symbol of the target candidate may not be displayed when the reliability of the presence is equal to or lower than a threshold. The threshold may be a fixed value, or may be changed according to circumstances such as weather and time.

[0075] The symbols of the target candidates are displayed on the screens of the radars 2 and 3, the ECDIS 10, the display device 7, or the like. In addition, the symbols of the target candidates may be displayed in AR (Augmented Reality) on the image from the image sensor 6 displayed on the display device 7.

[0076] In addition, when calculating the presence reliability, as described above, if a target candidate is located on the ship's planned route, the presence reliability is adjusted to be higher than when the target candidate is not located on the planned route, which is useful for both collision avoidance and lookout purposes.

[0077] (1) a plurality of candidate data generators each generating target candidate data including position data of target candidates existing around the ship; a common candidate selection unit that selects a plurality of target candidate data representing a common target candidate from the target candidate data generated by each of the plurality of candidate data generation units based on the position data; an existence reliability calculation unit that calculates an existence reliability of the same target candidate based on attributes of a plurality of candidate data generation units that generated the selected plurality of target candidate data among the plurality of candidate data generation units; A ship target detection system comprising:

[0078] (2) The target candidate data further includes type data of the target candidate, a type reliability calculation unit that calculates type reliability of the same target candidate based on the type data included in each of the selected target candidate data and attributes of a plurality of candidate data generation units that generate the selected target candidate data, A ship target detection system as described in (1).

[0079] (3) One of the plurality of candidate data generating units is any one of a radar, a communication device for receiving other ship information, a sonar, a combination of an image sensor for photographing the outside of the ship and an identification unit for identifying target candidates in the image, and an instruction receiving unit for receiving an instruction by a user of a target candidate in the image. A ship target detection system according to (1) or (2).

[0080] (4) the plurality of candidate data generating units include two or more of a radar, a communication device for receiving other ship information, a sonar, a set of an image sensor for photographing the outside of the ship and an identification unit for identifying target candidates in the image, and an instruction receiving unit for receiving an instruction by a user of a target candidate in the image; A ship target detection system according to any one of (1) to (3).

[0081] (5) The plurality of candidate data generating units include at least a radar. A ship target detection system according to any one of (1) to (4).

[0082] (6) The plurality of candidate data generating units include a plurality of radars each having a different operating frequency band. A ship target detection system according to any one of (1) to (5).

[0083] (7) One of the plurality of candidate data generation units is A detection unit configured to output detection data, the detection unit being a radar, an image sensor, or a sonar; an identification unit that uses detection data as input data and a trained model that has been generated in advance by machine learning using a target type as teacher data, estimates the type of the target candidate that appears in the detection data output from the detection unit, and generates the type data; Including, A ship target detection system as described in (2).

[0084] (8) The identification unit estimates a size or a type of a ship as a type of the target candidate. A ship target detection system according to (7).

[0085] (9) The identification unit estimates a type of floating object on the water as a type of the target candidate. A ship target detection system according to (7) or (8).

[0086] (10) the presence reliability calculation unit calculates the presence reliability based on one or more of size data representing a size of the target candidate, speed data representing a moving speed, track data representing a track, time data representing an elapsed time from detection, and intensity data representing an intensity of a detection signal, in addition to the attribute; A ship target detection system according to any one of (1) to (9).

[0087] (11) the presence reliability calculation unit calculates the presence reliability to be higher when the target candidate is located on a planned route than when the target candidate is not located on a planned route; A ship target detection system according to any one of (1) to (10).

[0088] (12) The presence reliability calculation unit calculates the presence reliability based on one or more of nautical chart data, tide data, and weather data in addition to the attributes. A ship target detection system according to any one of (1) to (11).

[0089] (13) and a collision avoidance calculation unit that calculates a collision avoidance planned route based on the target candidate data and the presence reliability of the target candidate. A ship target detection system according to any one of (1) to (12).

[0090] (14) a vessel control unit that controls the vessel according to the planned avoidance route, A ship target detection system according to (13).

[0091] (15) generating target candidate data including position data of target candidates existing around the ship by each of a plurality of candidate data generating units; selecting a plurality of target candidate data representing the same target candidate from the target candidate data generated by each of the plurality of candidate data generating units based on the position data; calculating a presence reliability of the same target candidate based on attributes of a plurality of candidate data generation units that generated the selected plurality of target candidate data among the plurality of candidate data generation units; A method for detecting targets on ships.

[0092] (16) a candidate data acquisition unit that acquires target candidate data including position data of target candidates existing around the ship, the target candidate data being generated by each of the plurality of candidate data generation units; a common candidate selection unit that selects a plurality of target candidate data representing a common target candidate from the target candidate data generated by each of the plurality of candidate data generation units based on the position data; an existence reliability calculation unit that calculates an existence reliability of the same target candidate based on attributes of a plurality of candidate data generation units that generated the selected plurality of target candidate data among the plurality of candidate data generation units; A reliability estimation device comprising:

[0093] (17) acquiring target candidate data including position data of target candidates existing around the ship, the target candidate data being generated by each of a plurality of candidate data generating units; selecting a plurality of target candidate data representing the same target candidate from the target candidate data generated by each of the plurality of candidate data generating units based on the position data; and calculating a presence reliability of the same target candidate based on attributes of a plurality of candidate data generation units that generated the selected plurality of target candidate data among the plurality of candidate data generation units; A program for causing a computer to execute the following. [Explanation of symbols]

[0094] 1 Reliability estimation device, 2 Radar, 3 Radar, 4 AIS, 5 Sonar, 6 Image sensor, 7 Display device, 8 GNSS receiver, 9 Gyrocompass, 10 ECDIS, 11 Wireless communication unit, 12 Automatic steering device, 21 Candidate data acquisition unit, 22 Same candidate selection unit, 23 Presence reliability calculation unit, 24 Type reliability calculation unit, 25 Object sign classification unit, 26 Instruction reception unit, 27 Avoidance calculation unit, 28 Ship control unit, 31 Radar candidate management DB, 32 AIS candidate management DB, 33 Sonar candidate management DB, 34 Image candidate management DB, 35 Common candidate management DB, 100 Ship target detection system

Claims

1. a plurality of candidate data generators each generating target candidate data including position data of target candidates existing around the ship; a common candidate selection unit that selects a plurality of target candidate data representing a common target candidate from the target candidate data generated by each of the plurality of candidate data generation units based on the position data; an existence reliability calculation unit that calculates an existence reliability of the same target candidate based on attributes of a plurality of candidate data generation units that generated the selected plurality of target candidate data among the plurality of candidate data generation units; Equipped with the presence reliability calculation unit calculates the presence reliability based on one or more of size data representing a size of the target candidate, speed data representing a moving speed, track data representing a track, and time data representing an elapsed time from detection, in addition to the attribute; the presence reliability calculation unit adjusts the presence reliability to be higher as the size of the target candidate increases, as the moving speed of the target candidate increases, or as the elapsed time from the detection of the target candidate increases; Marine target detection system.

2. The presence reliability calculation unit calculates the presence reliability based on one or two of nautical chart data and tidal current data in addition to the attribute.

2. A marine target detection system according to claim 1.

3. generating target candidate data including position data of target candidates existing around the ship by each of a plurality of candidate data generating units; selecting a plurality of target candidate data representing the same target candidate from the target candidate data generated by each of the plurality of candidate data generating units based on the position data; calculating a presence reliability of the same target candidate based on attributes of a plurality of candidate data generation units that generated the selected plurality of target candidate data among the plurality of candidate data generation units; A vessel target detection method, comprising: The calculation of the presence reliability is based on one or more of size data representing a size of the target candidate, speed data representing a moving speed, track data representing a track, and time data representing an elapsed time from detection, in addition to the attribute, and The calculation of the presence reliability is adjusted to be higher as the size of the target candidate increases, as the moving speed of the target candidate increases, or as the elapsed time from the detection of the target candidate increases. A method for detecting targets on ships.

4. a candidate data acquisition unit that acquires target candidate data including position data of target candidates existing around the ship, the target candidate data being generated by each of the plurality of candidate data generation units; a common candidate selection unit that selects a plurality of target candidate data representing a common target candidate from the target candidate data generated by each of the plurality of candidate data generation units based on the position data; an existence reliability calculation unit that calculates an existence reliability of the same target candidate based on attributes of a plurality of candidate data generation units that generated the selected plurality of target candidate data among the plurality of candidate data generation units; Equipped with the presence reliability calculation unit calculates the presence reliability based on one or more of size data representing a size of the target candidate, speed data representing a moving speed, track data representing a track, and time data representing an elapsed time from detection, in addition to the attribute; the presence reliability calculation unit adjusts the presence reliability to be higher as the size of the target candidate increases, as the moving speed of the target candidate increases, or as the elapsed time from the detection of the target candidate increases; Reliability Estimation Device.

5. acquiring target candidate data including position data of target candidates existing around the ship, the target candidate data being generated by each of a plurality of candidate data generating units; selecting a plurality of target candidate data representing the same target candidate from the target candidate data generated by each of the plurality of candidate data generating units based on the position data; and calculating a presence reliability of the same target candidate based on attributes of a plurality of candidate data generation units that generated the selected plurality of target candidate data among the plurality of candidate data generation units; Have your computer run The calculation of the presence reliability is based on one or more of size data representing a size of the target candidate, speed data representing a moving speed, track data representing a track, and time data representing an elapsed time from detection, in addition to the attribute, and The calculation of the presence reliability is adjusted to be higher as the size of the target candidate increases, as the moving speed of the target candidate increases, or as the elapsed time from the detection of the target candidate increases. program.

Images