Marine navigation signal intelligent identification system
By integrating the intelligent navigation signal recognition system of the external perception layer and the recognition cognitive layer on the intelligent ship, and using multiple sensors and neural network models to identify and process navigation signals, the problem of insufficient recognition capabilities of intelligent ships is solved and navigation safety is improved and accuracy of collision avoidance decisions.
Patent Information
- Application Number
- CN202410950918.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-16
- Publication Date
- 2024-11-26
AI Technical Summary
The lack of perception capabilities of the external perception layer in the prior art has led to the failure of intelligent ships to effectively identify the navigation signals and intentions of surrounding ships in path planning and collision avoidance, which limits the application of intelligent navigation technology.
A marine navigation signal intelligent identification system is designed, including an external sensing layer, an identification cognitive layer and an interactive interface layer. Navigation signal information is collected through acoustic signal sensors, visible light video sensors and infrared video sensors, and the neural network model is used to identify and Comprehensive processing to generate the location and navigation signal information of the source ship, and provide decision-making support for unmanned ships or manned ships.
It realizes intelligent identification and tracking of the navigation signals of surrounding ships, enhances the collision avoidance decision-making capabilities of intelligent ships, improves navigation safety, and complies with the agreements of the "International Maritime Collision Avoidance Rules".
Smart Images

Figure CN119028177A_ABST
Abstract
Description
Technical field:
[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an intelligent recognition system for ship navigation signals. Background technology:
[0002] With the development of artificial intelligence technology, intelligent ships and intelligent shipping have become the future development direction of the shipping industry and have attracted the attention of research institutions around the world. In recent years, unmanned boat technology has developed rapidly, and various types of intelligent ship products have been widely used. At present, there are relatively mature technical means for path planning and active collision avoidance of intelligent ships. However, due to the lack of intelligent means to automatically identify the navigation signals and intentions of encountering ships, the constraints of the International Regulations for Preventing Collisions at Sea are often less considered in path planning and collision avoidance, which limits the implementation of intelligent navigation technology. It can be expected that unmanned ships or intelligent ships will sail together with traditional ships for a long time in the future, which puts forward an urgent need for the intelligent identification of ship navigation signals and navigation intentions.
[0003] The International Maritime Regulations for Preventing Collisions stipulate that ships use lights, signal shapes, and horns to inform surrounding ships of their own status, which can effectively warn surrounding ships. The present invention discloses a ship navigation signal intelligent recognition system, which uses artificial intelligence to recognize the navigation signals of surrounding ships and obtain information about their status and surrounding obstacles. The system can be installed on unmanned ships / boats to provide an information source for their autonomous collision avoidance decisions; it can also be installed on manned ships / boats to enhance the crew's ability to recognize other ships' navigation signals.
[0004] There is an urgent need for an intelligent recognition system for ship navigation signals, which can help solve the technical problem in the prior art of lacking an external perception layer for perception and recognition. Summary of the invention:
[0005] In one embodiment, the present invention provides an intelligent recognition system for ship navigation signals, which helps to solve the technical problem of the lack of an external perception layer for perception and recognition in the prior art by means of perception, recognition and providing an interface.
[0006] The ship navigation signal intelligent recognition system includes an external perception layer, a recognition and cognition layer, and an interactive interface layer;
[0007] The external perception layer is used to sense and collect navigation signals sent by other ships;
[0008] The recognition and cognition layer is used to process the information from the external perception layer, and after recognizing the navigation signal, generate the position of the source ship and the information conveyed by the navigation signal according to the recognition result;
[0009] The interactive interface layer is used to provide an interface for information interaction with other equipment systems on the ship, as well as a human-computer interaction interface and an interface for unmanned ship collision avoidance decision-making.
[0010] In one embodiment, the external sensing layer includes an acoustic signal sensor and a visible light video sensor, and an infrared video sensor;
[0011] An acoustic signal sensor, used for receiving acoustic signals in the surrounding environment;
[0012] A visible light video sensor for capturing visible spectrum video information around the location;
[0013] Infrared video sensor, used to capture infrared spectrum video information around the location.
[0014] In one embodiment, the external perception layer further includes a video sensor motion control module, and the video sensor motion control module is used to control the visible light video sensor and the infrared video sensor to adjust their postures.
[0015] In one embodiment, the posture adjustment includes rotation, pitch, zoom and focus.
[0016] In one embodiment, the recognition layer further includes an acoustic signal recognition module;
[0017] The acoustic signal recognition module is used to receive the acoustic signal obtained from the acoustic signal sensor, identify and extract the horn sounds of other ships in the acoustic signal, and output characteristic information such as the direction of the sound source ship, the length / shortness of the horn sound, and the order of the horn sound.
[0018] In one embodiment, the recognition layer further includes an image recognition module;
[0019] The image recognition module is used to receive video information captured by the visible light video sensor and the infrared video sensor.
[0020] In one embodiment, the recognition layer further includes a navigation signal comprehensive recognition module;
[0021] The navigation signal comprehensive recognition module is used to receive the navigation signal feature information sent from the sound signal recognition module and the image recognition module.
[0022] In one embodiment, the recognition layer further includes a navigation signal comprehensive recognition module;
[0023] The navigation signal comprehensive recognition module is used to receive the navigation signal feature information sent by the sound signal recognition module and the image recognition module.
[0024] In one embodiment, the recognition layer further includes an automatic tracking module;
[0025] The module is used to receive identification information from the image identification module and tracking instructions from the navigation signal comprehensive recognition module, and issue video sensor movement instructions after identification and judgment.
[0026] In one embodiment, the interaction layer includes an AIS communication interface and an electronic chart interface, a radar communication interface, a human-computer interaction display and control module, and an unmanned ship collision avoidance interface. Description of the drawings:
[0027] Figure 1 A schematic diagram of the structure of an intelligent recognition system for ship navigation signals according to an embodiment of the present invention;
[0028] Figure 2 A schematic flow chart of a method for identifying a trumpet sound in a sound signal identification module according to an embodiment of the present invention;
[0029] Figure 3 A schematic flow chart of a method for identifying a signal light in two time periods in an image recognition module according to an embodiment of the present invention;
[0030] Figure 4 A schematic flow chart of a method for recognizing a navigation signal in a navigation signal comprehensive recognition module according to an embodiment of the present invention;
[0031] Figure 5 An example of the present invention provides a schematic diagram of a possible identifiable ship signal light. Specific implementation method:
[0032] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several variations and improvements may be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.
[0033] Figure 1 A schematic diagram of the structure of an intelligent recognition system for ship navigation signals according to an embodiment of the present invention; Figure 2 A schematic flow chart of a method for identifying a trumpet sound in a sound signal identification module according to an embodiment of the present invention; Figure 3 A schematic flow chart of a method for identifying a signal light in two time periods in an image recognition module according to an embodiment of the present invention; Figure 4 A schematic flow chart of a method for recognizing a navigation signal in a navigation signal comprehensive recognition module according to an embodiment of the present invention; Figure 5 An example of the present invention provides a schematic diagram of a possible identifiable ship signal light.
[0034] like Figures 1 to 5As shown, in one embodiment, the present invention provides an intelligent recognition system for ship navigation signals, which helps to solve the technical problem of the lack of an external perception layer for perception and recognition in the prior art by means of perception, recognition and providing an interface.
[0035] The ship navigation signal intelligent recognition system includes an external perception layer, a recognition and cognition layer, and an interactive interface layer;
[0036] The external perception layer is used to sense and collect navigation signals sent by other ships;
[0037] The recognition and cognition layer is used to process the information from the external perception layer, and after recognizing the navigation signal, generate the position of the source ship and the information conveyed by the navigation signal according to the recognition result;
[0038] The interactive interface layer is used to provide an interface for information interaction with other equipment systems on the ship, as well as a human-computer interaction interface and an interface for unmanned ship collision avoidance decision-making.
[0039] In this embodiment, a specific implementation of a ship navigation signal intelligent recognition system is provided. The navigation signal includes a signal light, a signal shape, a signal sound, etc. After the navigation signal is recognized according to the International Regulations for Preventing Collisions at Sea.
[0040] In one embodiment, the external sensing layer includes an acoustic signal sensor and a visible light video sensor, and an infrared video sensor;
[0041] An acoustic signal sensor, used for receiving acoustic signals in the surrounding environment;
[0042] A visible light video sensor for capturing visible spectrum video information around the location;
[0043] Infrared video sensor, used to capture infrared spectrum video information around the location.
[0044] In this embodiment, three specific implementations of sensors are provided.
[0045] In one embodiment, the external perception layer further includes a video sensor motion control module, and the video sensor motion control module is used to control the visible light video sensor and the infrared video sensor to adjust their postures.
[0046] In this embodiment, a specific implementation method of a video sensor motion control module for adjusting posture is provided.
[0047] In one embodiment, the posture adjustment includes rotation, pitch, zoom and focus.
[0048] In this embodiment, a specific implementation method for adjusting the posture is provided.
[0049] In one embodiment, the recognition layer further includes an acoustic signal recognition module;
[0050] The acoustic signal recognition module is used to receive the acoustic signal obtained from the acoustic signal sensor, identify and extract the horn sounds of other ships in the acoustic signal, and output characteristic information such as the direction of the sound source ship, the length / shortness of the horn sound, and the order of the horn sound.
[0051] In this embodiment, a specific implementation of an acoustic signal recognition module is provided.
[0052] In one embodiment, the recognition layer further includes an image recognition module;
[0053] The image recognition module is used to receive video information captured by visible light video sensors and infrared video sensors, identify and extract the position of other ships in the video and the information of their lights and types, output characteristic information such as the position of the source ship, the coordinates and color of the lights, the coordinates and color of the types, and cooperate with the automatic tracking module and the sensor motion control module to control the video sensor to track the ship.
[0054] In this embodiment, a specific implementation of the image recognition module is provided.
[0055] In one embodiment, the recognition layer further includes a navigation signal comprehensive recognition module;
[0056] The navigation signal comprehensive recognition module is used to receive the navigation signal feature information sent from the sound signal recognition module and the image recognition module.
[0057] In this embodiment, a specific implementation of a navigation signal comprehensive recognition module is provided. According to the relevant provisions of the International Regulations for Preventing Collisions at Sea, the type, navigation dynamics, status and relative position of the target ship are judged, and the comprehensive recognition results are matched and integrated and sent to the interaction layer.
[0058] In one embodiment, the recognition layer further includes an automatic tracking module;
[0059] The module is used to receive identification information from the image identification module and tracking instructions from the navigation signal comprehensive recognition module, and issue video sensor movement instructions after identification and judgment.
[0060] In this embodiment, a specific implementation of an automatic tracking module is provided.
[0061] In one embodiment, the interaction layer includes an AIS communication interface and an electronic chart interface, a radar communication interface, a human-computer interaction display and control module, and an unmanned ship collision avoidance interface.
[0062] In this embodiment, a specific implementation method of the interaction layer including multiple interfaces is provided. An AIS communication interface is used to receive the position information of surrounding ships acquired by AIS; an electronic chart and radar communication interface is used to mark the information of the ship navigation signal intelligent recognition system on the electronic chart and radar system. A human-computer interaction display and control module visualizes the output results of the recognition and cognition layer and displays them to the operator, and provides an interface for changing system parameters and controlling system status; an unmanned ship collision avoidance interface is used to send the output results of the recognition and cognition layer to the unmanned ship's collision avoidance decision-making system, and provide surrounding ship navigation signal information for collision avoidance.
[0063] Based on the above system framework:
[0064] This embodiment is aimed at a small number of intelligent ships. Based on the typical facility configuration, it involves a ship navigation signal intelligent recognition system, including three parts: external perception layer, recognition and cognition layer and interaction layer. The specific components and associations of each layer are as follows: Figure 1 As shown. The external perception layer includes an acoustic signal sensor, a visible light video sensor, an infrared video sensor and a video sensor motion control module, which are used to sense and collect navigation signals emitted by surrounding ships. The acoustic signal sensor is an array composed of multiple microphones arranged in a certain pattern, which collects the horn signals emitted by the ship; the visible light video sensor is a ship-mounted rotatable and zoomable camera device that captures the real image information of the surrounding ships; the infrared video sensor is a ship-mounted rotatable and zoomable camera device that captures the infrared spectrum image information of the surrounding ships; the video sensor motion control module is used to control the rotation, pitch, zoom and focus of the two video sensors.
[0065] The recognition and cognition layer includes a sound signal recognition module, an image recognition module and a navigation signal comprehensive recognition module, which are used to process information from the external perception layer, identify the type of navigation signals in the information, and judge the type and status of the source ship based on characteristics such as color, position, quantity and duration.
[0066] A method for identifying a signal in an acoustic signal identification module, such as Figure 2As shown. First, the audio collected by the acoustic signal sensor is passed through a bandpass filter to filter out the audio signal within the frequency range of the ship's horn. Secondly, the filtered audio is preprocessed and converted into a digital signal. Thirdly, the audio signal is feature extracted using the Mel-scale Frequency Cepstral Coefficients (MFCC) analysis method. After that, a neural network model is used to identify the ship's horn. The specific methods used include but are not limited to Deep Belief Nets (DBN), which outputs information such as the shortness / length and order of the horn. At the same time, based on the positioning algorithm of Time Difference of Arrival (TDOA), combined with the arrangement of microphones in the microphone array, the position of the sound source ship is estimated.
[0067] The image recognition module determines whether the environment is day or night based on the local time, such as Figure 3 As shown in the figure, the recognition methods and targets in the two environments are different. The neural network model YOLO v5 (You Only Look Once v5) used for image recognition establishes specific training sets for different detection targets, making it suitable for multi-class target detection. When it is daytime, the ship type can be identified only by visible light video images. The steps are as follows: 1) Identify the ship in the visible light video image through the neural network model; 2) Track the ship with the detection frame and calculate the direction of the target ship relative to the ship; 3) Capture the image in the detection frame; 4) Identify the ship type in the detection frame image through the neural network model, and output the coordinates and type information of each type. When it is dark, the infrared video image and the visible light video image are fused to identify the ship's signal lights. The steps are as follows: 1) Since the shooting positions and angles of the infrared video sensor and the visible light video sensor are slightly different, the two video images are first cropped, offset, lens corrected, etc. to make them completely overlap; 2) The ship in the infrared video image is identified through the neural network model; 3) The detection frame tracks the ship and calculates the position of the target ship relative to the ship; 4) The image in the detection frame is captured and mapped to the visible light video image, and the image at the same position is intercepted; 5) The neural network model is used to identify the ship's signal lights in the intercepted visible light video image, and the coordinates and color information of each signal light are output. After completing the recognition and tracking of the ship's navigation signals in the picture, the video sensor is controlled to rotate, and the navigation signals of all surrounding ships are dynamically identified by scanning.
[0068] A method for recognizing navigation signals in the navigation signal comprehensive recognition module, such as Figure 4As shown. It mainly includes the process of integrating the navigation signal recognition results and the surrounding ship position information transmitted by other systems, matching the source ship and judging the information transmitted by the navigation signal. The position information of the surrounding ships is obtained through the AIS interface, matched with the ship position output by the acoustic signal recognition module and the image recognition module, and the position of the ship that sent the navigation signal is determined.
[0069] According to the relevant provisions of the International Regulations for Preventing Collisions at Sea, the sound signal recognition module identifies the short / long and order of the horn, and judges the status or direction of the output sound source ship. For example, if it is identified that a ship emits short, long, and short horns, it is recognized that the ship is at anchor. The image recognition module identifies the coordinates and type of each signal type, and judges the status of the target ship, the towing combination, the type of operation it is engaged in, and the surrounding obstructions. For example, it identifies that a ship hangs two vertical spherical signal types, and recognizes that the ship is out of control. The coordinates of each signal light recognized by the image recognition module determine the relative position of the signal light on the ship, and the color of the signal light is combined to judge the status of the target ship, the heading range, type, towing combination, type of operation it is engaged in, and the surrounding obstructions. For example, it is detected Figure 4 The relative positions of the signal lights of the ship shown are determined based on the coordinates of the signal lights output by the image recognition module. The perceptible information combined with the colors of the signal lights is as follows: 1) the ship is not a trawling fishing vessel; 2) it is underway and moving in the water; 3) its fishing gear is extended outward on the port side, and the horizontal length of the extension is greater than 150m; 4) the own ship is directly in front of the ship.
[0070] The interaction layer provides an interface for information interaction with other equipment / systems of the ship, as well as a human-computer interaction interface and an interface for collision avoidance decision of unmanned ships. The AIS communication interface provides the navigation signal comprehensive recognition module with the position information of surrounding ships; the electronic chart and radar communication interface provide the navigation signal recognized by the navigation signal comprehensive recognition module to the AIS, electronic chart, radar and other systems, and the above systems can be called on demand. The human-computer interaction display and control module provides a visual interface to the crew, displaying the real-time navigation signal recognition screen and recognition results. The user can manipulate the movement of the image sensor and the tracking of the target ship. A development and maintenance interface is provided, and the system can be modified, maintained, upgraded, tested and other operations can be performed. The intelligent ship collision avoidance interface provides the collision avoidance information obtained from the navigation signal recognition to the collision avoidance decision system of the ship, and provides information such as the status, movement, heading range, type, and surrounding obstruction range of surrounding ships, so as to improve the accuracy of decision-making and reduce the probability of navigation accidents. It can also receive tracking and identification instructions sent by the collision avoidance decision system, focusing on tracking and identifying the navigation signals of a certain ship or ships in a certain area.
[0071] In view of the defects in the prior art, the purpose of the present invention is to provide navigation signal information of surrounding ships for collision avoidance decision-making of unmanned ships / boats, and to provide auxiliary means for crew members of manned ships / boats to identify navigation signals.
[0072] The technical objectives of the present invention are achieved through the following technical solutions:
[0073] The invention discloses an intelligent recognition system for ship navigation signals, which is divided into an external perception layer, a recognition and judgment layer and an interaction layer.
[0074] The external perception layer is composed of an acoustic signal sensor, a visible light video sensor, an infrared video sensor and a video sensor motion control module.
[0075] Preferably, the acoustic signal sensor is an array composed of multiple microphones, so that the time at which the horn sound from the same sound source reaches the microphone pair with a certain position relationship is different.
[0076] Preferably, the visible light video sensor is a marine camera device having but not limited to the following characteristics: 1) variable focal length, with a large variable focal range, which can meet the needs of identifying multiple targets over a large range, identifying and tracking a single target at close / long distances, etc.; 2) pitch adjustment and 360° rotation, which can shoot around the ship and track specific ships; 3) beyond visual range, which can shoot clear ultra-long-distance ship video images, improving recognition accuracy and perception range.
[0077] Preferably, the infrared video sensor is a marine camera device that can capture video images in the infrared spectrum, and its characteristics are the same as those of the visible light video sensor. The infrared video sensor and the visible light video sensor are installed adjacently on the same platform, and the images captured by the two are almost the same.
[0078] The video sensor motion control module receives instructions from the human-computer interaction display and control module and the image recognition module to perform motion control such as rotation, pitch, zoom and focus on the video image sensor.
[0079] The recognition and judgment layer is composed of an acoustic signal recognition module, an image recognition module, a navigation signal comprehensive recognition module and an automatic tracking module.
[0080] The acoustic signal recognition module processes and recognizes the acoustic signal received by the acoustic signal sensor. The acoustic signal in the horn frequency band is filtered out using a bandpass filter, and then preprocessed and converted into a digital signal for subsequent processing. Feature extraction is performed on the preprocessed signal, and the horn in the acoustic signal is identified using a neural network model. The length / shortness and number of horns are output as part of the horn recognition result; the time when each microphone receives the horn is output, the time delay of different microphones receiving the horn is calculated, and the sound source ship is located based on the arrangement of the microphone array.
[0081] The image recognition module processes and recognizes the video signals collected by the visible light video sensor and the infrared video sensor, and is divided into navigation signal recognition in two environments: daytime and nighttime:
[0082] 1) Identify the type of the target ship in the daytime environment. The ship image recognition model is used to identify the ship in the collected visible light video image, the detection frame tracks the identified ship, and the position of the target ship relative to the ship is calculated and output. At the same time, the image in the detection frame is captured, and the ship type in the captured image is recognized again by the ship image recognition model, and the center coordinates of each type image and the type of type are output.
[0083] 2) Identify the target ship’s signal lights in the dark. Perform operations such as cropping, offsetting, and lens correction on the video images captured by the visible light and infrared video image sensors to ensure that the two images can completely overlap. The ship infrared image recognition model is used to identify the ship in the collected infrared video image. The detection frame tracks the identified ship and calculates and outputs the orientation and outline of the target ship relative to the ship. At the same time, the image within the detection frame is captured. The captured infrared image is mapped on the visible light image to extract the visible light image of the corresponding area. The ship signal lights in the extracted visible light image are identified again by the ship signal light image recognition model, and the center coordinates of each signal light image and the color of the signal light are output.
[0084] The navigation signal comprehensive recognition module comprehensively recognizes and judges the navigation signal information obtained in the previous stage, and outputs the movement, status and obstruction information of the corresponding ship. The azimuth information of the surrounding ships is obtained through the AIS communication interface, and matched with the ship azimuth output by the sound signal recognition module and the image recognition module to determine the position of the ship that sent the navigation signal. According to the relevant provisions of the International Regulations for Preventing Collisions at Sea, the state or movement of the output sound source ship is judged by the short / long and sequence of the signal recognized by the sound signal recognition module; the state, towing combination, type of operation and surrounding obstructions of the target ship are judged by the coordinates and type of each signal type recognized by the image recognition module; the relative position of the signal light on the ship is judged by the coordinates of each signal light recognized by the image recognition module, and the state, heading range, type, towing combination, type of operation and surrounding obstructions of the target ship are judged by combining the color of the signal light.
[0085] The automatic tracking module receives the recognition information from the image recognition module and the tracking instructions from the navigation signal comprehensive recognition module, marks the target ship to be tracked, and sends instructions to the sensor motion control module to ensure that the target ship is always within the field of view and the image is clearly visible.
[0086] The interaction layer is composed of an AIS communication interface, an electronic chart and radar communication interface, a human-computer interaction display and control module, and an unmanned ship collision avoidance interface.
[0087] The AIS communication interface provides the position information of surrounding ships to the navigation signal comprehensive recognition module; the electronic chart and radar communication interface provides the navigation signal identified by the navigation signal comprehensive recognition module to the AIS, electronic chart, radar and other systems, and the above systems can be called as needed.
[0088] The human-computer interaction display and control module provides a visual interface for the system users, showing the real-time navigation signal recognition screen and recognition results. The user can manipulate the movement of the image sensor and the tracking of the target ship. A development and maintenance interface is provided to modify the system parameters, perform maintenance upgrades, and test operations.
[0089] The smart ship collision avoidance interface is an optional interface. When the system is installed on an unmanned ship, it provides the unmanned ship's collision avoidance decision system with the results of navigation signal recognition, enriches the information obtained by the collision avoidance decision system, and reduces the probability of navigation accidents. It can also receive tracking and identification instructions sent by the collision avoidance decision system, focusing on tracking and identifying the navigation signals of a certain ship or ships in a certain area.
[0090] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is intended to include these modifications and variations.
Claims
1. An intelligent recognition system for ship navigation signals, It is characterized in that The ship navigation signal intelligent recognition system comprises: An external sensing layer, used to sense and collect navigation signals sent by other ships; An identification and cognition layer is used to process the information from the external perception layer, and after identifying the navigation signal, generate the position of the source ship and the information conveyed by the navigation signal according to the identification result; An interactive interface layer is used to provide an interface for information interaction with other ship equipment systems, as well as a human-computer interaction interface and an interface for unmanned ship collision avoidance decision-making.
2. The intelligent recognition system for ship navigation signals according to claim 1, It is characterized in that The external perception layer includes: An acoustic signal sensor, used for receiving acoustic signals in the surrounding environment; a visible light video sensor for capturing visible spectrum video information around the location; An infrared video sensor is used to capture infrared spectrum video information around the location.
3. The intelligent recognition system for ship navigation signals according to claim 2, It is characterized in that The external perception layer also includes: The video sensor motion control module is used to control the visible light video sensor and the infrared video sensor to adjust their posture.
4. The intelligent recognition system for ship navigation signals according to claim 3, It is characterized in that The posture adjustment includes rotation, pitch, zoom and focus.
5. The intelligent recognition system for ship navigation signals according to claim 4, It is characterized in that The recognition layer also includes: The sound signal recognition module is used to receive the sound signal obtained from the sound signal sensor, identify and extract the sound of other ships in the sound signal, and output characteristic information such as the direction of the sound source ship, the length / shortness of the sound, and the order of the sound.
6. The intelligent recognition system for ship navigation signals according to claim 5, It is characterized in that The recognition layer also includes: An image recognition module is used to receive video information captured by the visible light video sensor and the infrared video sensor.
7. The intelligent recognition system for ship navigation signals according to claim 6, It is characterized in that The recognition layer also includes: A navigation signal comprehensive recognition module is used to receive the navigation signal feature information sent from the sound signal recognition module and the image recognition module.
8. The intelligent recognition system for ship navigation signals according to claim 7, It is characterized in that The recognition layer also includes: A navigation signal comprehensive recognition module is used to receive the navigation signal feature information sent from the sound signal recognition module and the image recognition module.
9. The intelligent recognition system for ship navigation signals according to claim 8, It is characterized in that The recognition layer also includes: An automatic tracking module is used to receive the recognition information from the image recognition module and the tracking instructions from the navigation signal comprehensive recognition module, and issue the video sensor movement instructions after recognition and judgment.
10. The intelligent recognition system for ship navigation signals according to claim 9, It is characterized in that The interaction layer includes an AIS communication interface and an electronic chart interface, a radar communication interface, a human-computer interaction display and control module, and an unmanned ship collision avoidance interface.
