Ship trajectory identification method and system based on photoelectric blind compensation, and medium

Abstract

The invention provides a ship trajectory recognition method and system based on photoelectric blind compensation and a medium, and the method comprises the steps: S1, detecting a ship in an unshielded region through a radar, obtaining the radar position data, and simulating a ship trajectory; s2, a photoelectric station is arranged in a blind area where the radar is likely to be shielded, photoelectric detection is conducted through the photoelectric station, photoelectric position data of the ship are obtained, and channel monitoring is guaranteed to be uninterrupted; s3, transmitting the radar position data, the ship track and the photoelectric position data to a data processing center, and constructing a complete ship navigation track; and S4, outputting the ship trajectory to a multi-element sensing system to realize real-time monitoring and management of the ship. According to the method, the photoelectric station is arranged in the complex terrain area where the radar is prone to being shielded to form a blind area, the limitation of radar monitoring is successfully broken through, channel monitoring is ensured to be uninterrupted, monitoring continuity is guaranteed in an all-around mode, and the monitoring blank caused by shielding is effectively filled up.

Description

Technical Field

[0001] The present invention relates to the technical field of ship track recognition, and in particular to a ship track recognition method, system and medium based on photoelectric blindness compensation. Background Art

[0002] As a major transportation country, my country's water transportation occupies a pivotal position in the country's total transportation volume, accounting for as much as 90%. As important components of water transportation, ocean transportation and inland river transportation are key supporting forces for promoting my country's economic development.

[0003] With the booming shipping industry and the increasing number of ships, the importance of water transport safety management has become increasingly prominent. In the key link of monitoring the trajectory of ships in waterways, traditional radar systems have always played an important role. They can track the movement trajectory of ships and provide certain guarantees for shipping safety.

[0004] However, traditional radar systems have limitations that are difficult to ignore. When radar waves encounter obstructions such as mountains and tall buildings, they will be blocked, forming monitoring blind spots. In these blind spots, the real-time dynamics of ships cannot be effectively captured, resulting in missing ship trajectory information, which seriously affects the integrity of the ship's trajectory. This not only creates loopholes in the supervision of ship navigation, but also greatly increases the risk of collisions between ships, posing a serious threat to shipping safety.

[0005] At the same time, shipping density continues to increase, and the waterway environment becomes increasingly complex. Ships of different types and tonnages shuttle back and forth, which puts higher demands on the accuracy and comprehensiveness of ship trajectory monitoring. Traditional radar systems are limited by their own technical principles and have gradually become unable to meet actual needs in the face of these new challenges.

[0006] Optoelectronic equipment has significant advantages such as high resolution and the ability to provide intuitive images. It can effectively supplement radar monitoring blind spots and obtain ship information that radar cannot detect. Therefore, the organic combination of optoelectronic equipment and radar, giving full play to the advantages of both, and achieving comprehensive and accurate identification of ship trajectories is of great practical significance for improving the level of water transport safety management and ensuring shipping safety.

[0007] Through searching patent documents, it was found that the invention patent with publication number CN110376593A discloses a target perception method and device based on laser radar. The method includes the following steps: Step 1: The laser radar perception module detects the waters around the unmanned boat and obtains the location information of the target; Step 2: The pan-tilt control module is used to calculate the pan-tilt parameters of the optoelectronic system according to the location information of the target, and control the optoelectronic system to accurately locate the target; Step 3: The video analysis module is used to detect and identify the target monitoring screen frame by frame, obtain the target category and the position of the target in the image, and extract the contour, size and color of the target; Step 4: Upload the perceived target data to the unmanned boat control center. This patent is limited to the waters around the unmanned boat, and the detection method focuses on positioning calculation, lacks multi-stage processing and parameter adjustment, and the training model is single.

[0008] In summary, in response to the above-mentioned problems of the prior art, studying a ship trajectory recognition method, system and medium based on photoelectric blindness has become a key task that needs to be solved urgently. Summary of the invention

[0009] In view of the defects in the prior art, the purpose of the present invention is to provide a ship trajectory recognition method, system and medium based on photoelectric blindness compensation.

[0010] A ship trajectory recognition method based on photoelectric blindness compensation provided by the present invention comprises the following steps:

[0011] Step S1, using radar to detect ships in an unobstructed area, obtaining radar position data and simulating ship trajectories;

[0012] Step S2, setting up an optoelectronic station in a blind area where the radar is easily blocked, performing optoelectronic detection through the optoelectronic station, and obtaining optoelectronic position data of the ship to ensure uninterrupted channel monitoring;

[0013] Step S3, transmitting the radar position data, the ship track and the photoelectric position data to the data processing center to construct a complete ship navigation track;

[0014] Step S4, outputting the ship trajectory to the multi-sensing system to achieve real-time monitoring and management of the ship.

[0015] Preferably, step S2 includes the following sub-steps:

[0016] Step S2.1, image acquisition: the camera in the optical power station monitors the blind area in real time and collects images of the ship;

[0017] Step S2.2, target detection: the collected ship images are processed in real time using the Yolov5 target detection algorithm to obtain the ship target recognition result to ensure accurate recognition of the ship target;

[0018] Step S2.3, photoelectric position data calculation: Based on the ship target recognition result, combined with the geographic coordinates, height of the photovoltaic power station and the pitch angle and field of view of the camera, the photoelectric position data is calculated through geometric relationships.

[0019] Preferably, step S2.1 includes: constructing a photovoltaic power station: deploying a photovoltaic power station in a blind area, configuring the photovoltaic power station with high-precision dual-spectrum optoelectronic equipment, the site selection of the photovoltaic power station is based on the scope of the blocked area and the frequency of ship traffic, and the height of the photovoltaic power station is reasonably designed based on the surrounding environment and the field of view of the camera.

[0020] Preferably, in step S2.1, the camera of the photovoltaic power station has two working modes:

[0021] Visible light working mode: During the day, that is, from 4 a.m. to 8 p.m., the visible light camera is automatically enabled to capture the color information of the ship. The visible light camera uses the Yolov5 target detection algorithm to analyze and process the collected ship images to complete the detection of ship targets in the image;

[0022] Infrared light working mode: At night, that is, from 5 pm to 7 am, the infrared camera is automatically enabled. The infrared camera uses the Yolov5 target detection algorithm to analyze and process the collected ship images to complete the detection of ship targets in the image;

[0023] Mode overlap processing: When the two working modes overlap, that is, from 4:00 a.m. to 7:00 a.m. and from 5:00 p.m. to 8:00 p.m., the two working modes are started at the same time to perform target detection processing on the ship images collected by the visible light camera and the infrared camera respectively.

[0024] Preferably, step S2.2 includes the following sub-steps:

[0025] Step S2.2.1, training of target detection network: based on the Yolov5 deep network, the ship image collected in step S2.1 is trained to obtain a detection network for ship target detection;

[0026] Step S2.2.2, ship target detection optimization: Based on the detection network, the accuracy of ship target recognition is improved by expanding the training data and optimizing the target detection algorithm.

[0027] Preferably, step S2.3 includes the following sub-steps:

[0028] Step S2.3.1, calculate the sight angle of the ship target relative to the center of the camera according to the pixel position of the ship target in the picture:

[0029] Horizontal deviation angle (Δφ):

[0030]

[0031] Vertical deviation angle (Δθ):

[0032]

[0033] Ship target azimuth (atarget) and pitch angle (θtarget):

[0034] a target =a+Δa

[0035] θ target =θ+Δθ

[0036] Among them, θ: the pitch angle of the camera, positive value means upward, negative value means downward; a: the azimuth angle of the camera, clockwise relative to the true north; αh: the horizontal field of view of the camera; αv: the vertical field of view of the camera; pixel coordinates (xtarget, ytarget); picture resolution (Wframe, Hframe);

[0037] Step S2.3.2, set the ship to be on the horizontal plane, and calculate the horizontal distance Dtarget of the ship using the geometric relationship according to the altitude Hcam of the photovoltaic power station and the pitch angle θtarget of the ship target:

[0038]

[0039] Where, (Lcam, λcam): latitude and longitude of the photovoltaic station; Hcam: altitude of the photovoltaic station;

[0040] Step S2.3.3, using the longitude and latitude of the photovoltaic power station (Lcam, λcam) and the ship target distance Dtarget, calculate the longitude and latitude of the ship, that is, the photoelectric position data:

[0041] Longitude change (Δλ):

[0042]

[0043] Latitude change (ΔL):

[0044]

[0045] Latitude and longitude of the ship target:

[0046] L target =L cam +ΔL

[0047] λ target =λ cam +Δλ

[0048] Where, (Lcam, λcam): latitude and longitude of the photovoltaic station; Hcam: altitude of the photovoltaic station; R: radius of the earth.

[0049] Preferably, in step S3, the data processing center first establishes a unified coordinate system, and then uses a Kalman filter algorithm to match and integrate the radar position data, the ship trajectory and the photoelectric position data to obtain a complete ship navigation trajectory.

[0050] Preferably, in step S4, the multi-sensing system displays in real time the ship target identified by radar detection and optoelectronic detection of the optoelectronic station, and simultaneously presents the trajectory information of the ship target.

[0051] The present invention also provides a ship track recognition system based on photoelectric blindness compensation, comprising:

[0052] Module M1 uses radar to detect ships in unobstructed areas, obtains radar position data and simulates ship trajectories;

[0053] Module M2, in the blind area where the radar is easily blocked, sets up an optoelectronic station to perform photoelectric detection through the optoelectronic station to obtain the optoelectronic position data of the ship to ensure uninterrupted channel monitoring;

[0054] Module M3 transmits radar position data, ship trajectory and optoelectronic position data to the data processing center to construct a complete ship navigation trajectory;

[0055] Module M4 outputs the ship trajectory to the multi-sensing system to achieve real-time monitoring and management of the ship.

[0056] The present invention also provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned ship trajectory recognition method based on photoelectric blindness compensation are implemented.

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] 1. The present invention successfully breaks through the limitations of radar monitoring by setting up photoelectric stations in complex terrain areas where radars are easily blocked and form blind spots, ensuring uninterrupted channel monitoring, all-round protection of monitoring continuity, and effectively filling the monitoring gaps caused by obstruction.

[0059] 2. This camera adopts Yolov5 deep learning target detection algorithm, which not only accurately identifies the ship in the image, but also calculates the actual position of the ship through the precise positioning parameters of the optoelectronic station, ensuring that the detection accuracy reaches a high standard.

[0060] 3. The present invention integrates the ship position information detected by the photoelectric station with the trajectory data recorded by the radar system to generate a complete and coherent ship navigation trajectory, which greatly improves the comprehensiveness and coherence of trajectory monitoring and fully presents the ship's navigation trajectory.

[0061] 4. The present invention has excellent environmental adaptability and can be easily applied in various scenarios such as oceans, lakes, rivers, etc., perfectly meeting various needs of ship monitoring in different geographical environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0063] Figure 1 The present invention is a flow chart of a method for ship trajectory recognition based on photoelectric blindness compensation in an embodiment of the present invention. DETAILED DESCRIPTION

[0064] 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 changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0065] The present invention is based on a combination of optoelectronic monitoring technology and a radar system, and uses a deep learning target detection algorithm to accurately identify the position of a ship within the radar blind spot.

[0066] Embodiment 1:

[0067] Figure 1 The present invention is a flow chart of a method for ship trajectory recognition based on photoelectric blindness compensation in an embodiment of the present invention.

[0068] like Figure 1 As shown, this embodiment provides a ship track recognition method based on photoelectric blindness compensation, comprising the following steps:

[0069] Step S1, using radar to detect ships in an unobstructed area, obtaining radar position data and simulating ship tracks.

[0070] Step S2, setting up an optoelectronic station in a blind area where the radar is easily blocked, performing photoelectric detection through the optoelectronic station, and obtaining the optoelectronic position data of the ship to ensure uninterrupted channel monitoring.

[0071] Specifically, step S2 includes the following sub-steps:

[0072] Step S2.1, image acquisition: The camera in the optical power station monitors the blind area in real time and acquires images of the ship.

[0073] Further specifically, step S2.1 includes: constructing a photoelectric station: deploying a photoelectric station in the blind area, the photoelectric station is equipped with high-precision dual-spectrum photoelectric equipment, the site selection of the photoelectric station is based on the scope of the blocked area and the frequency of ship passage, and the height of the photoelectric station is reasonably designed based on the surrounding environment and the field of view of the camera to ensure that the geographical location of the photoelectric station can cover the ship passage area in the blind area to the greatest extent, and ensure good detection effect.

[0074] Before image acquisition, the camera parameters need to be accurately determined. The camera's pitch angle and field of view need to be accurately adjusted according to the location and height of the photovoltaic power station and the possible location range of the ship. In addition, the camera's resolution, frame rate and other parameters also need to be selected according to actual needs to meet the accuracy and real-time requirements of ship detection.

[0075] Furthermore, the camera of the photovoltaic power station has two working modes:

[0076] Visible light working mode: During the day, that is, from 4 a.m. to 8 p.m., the visible light camera is automatically enabled to capture the color information of the ship. The visible light camera analyzes and processes the collected ship images based on the Yolov5 target detection algorithm to complete the detection of ship targets in the image.

[0077] Infrared light working mode: At night, that is, from 5 pm to 7 am, the infrared camera is automatically enabled. The infrared camera analyzes and processes the collected ship images based on the Yolov5 target detection algorithm to complete the detection of ship targets in the image.

[0078] Mode overlap processing: When the two working modes overlap, that is, from 4:00 a.m. to 7:00 a.m. and from 5:00 p.m. to 8:00 p.m., the two working modes are started at the same time to perform ship target detection processing on the ship images collected by the visible light camera and the infrared camera respectively.

[0079] In this embodiment, in order to obtain the ship information more clearly, the lens needs to be automatically controlled, and the specific steps are as follows:

[0080] 1. Based on the result of target detection (rectangular frame), calculate the pixel width of the target, and then calculate its ratio to the screen width. Then, calculate the change in lens parameters relative to the current parameters when the target is adjusted to occupy 80% of the screen width. This change is used as the basis for controlling the lens.

[0081] 2. Adjust the lens according to the adjustment parameters calculated above, so as to make the target (ship) appear in the picture as much as possible, increase the proportion of the ship in the picture, and minimize the proportion of the background area.

[0082] 3. If the target size is too large and exceeds the current field of view, you need to increase the field of view to fully include the target.

[0083] Subsequent trajectory fusion operations will be based on the detection results of the two modes and can be performed as long as a ship is detected in one of the channels.

[0084] Step S2.2, target detection: The collected ship images are processed in real time using the Yolov5 target detection algorithm to obtain the ship target recognition result to ensure accurate recognition of the ship target.

[0085] Specifically, step S2.2 includes the following sub-steps:

[0086] Step S2.2.1, training of target detection network: Based on the Yolov5 deep network, the ship images collected in step S2.1 are trained to obtain a detection network for ship target detection.

[0087] In order to improve the reasoning speed, the Yolov5 network needs to be deployed on the NVIDIA GPU. During the deployment process, tensorRT technology is used as the basis. Through the optimization of this technology, the Yolov5 network can reach the extreme reasoning speed on the NVIDIA GPU hardware.

[0088] Step S2.2.2, ship target detection optimization: Based on the detection network, the accuracy of ship target recognition is improved by expanding the training data (including collecting more ship images in different scenarios) and optimizing the target detection algorithm.

[0089] Step S2.3, photoelectric position data calculation: Based on the ship target recognition result, combined with the geographic coordinates, height of the photovoltaic power station and the pitch angle and field of view of the camera, the photoelectric position data is calculated through geometric relationships.

[0090] Specifically, step S2.3 includes the following sub-steps:

[0091] Step S2.3.1, calculate the sight angle of the ship target relative to the center of the camera according to the pixel position of the ship target in the picture:

[0092] Horizontal deviation angle (Δa):

[0093]

[0094] Vertical deviation angle (Δθ):

[0095]

[0096] Ship target azimuth (φtarget) and pitch angle (θtarget):

[0097] φ target =φ+Δφ

[0098] θ target =θ+Δθ

[0099] Among them, θ: the pitch angle of the camera (unit: degree), positive value means upward, negative value means downward; φ: the azimuth angle of the camera (unit: degree), clockwise relative to the north; αh: the horizontal field of view of the camera (unit: degree); αv: the vertical field of view of the camera (unit: degree); pixel coordinates (xtarget, ytarget); picture resolution (Wframe, Hframe);

[0100] Step S2.3.2, set the ship to be on the horizontal plane (height is 0m, relative to the altitude of the photovoltaic power station), and calculate the horizontal distance Dtarget of the ship using the geometric relationship according to the altitude Hcam of the photovoltaic power station and the pitch angle θtarget of the ship target:

[0101]

[0102] Wherein, (Lcam, λcam): latitude and longitude of the photovoltaic station (unit: degree); Hcam: altitude of the photovoltaic station (unit: meter).

[0103] Note: The negative sign is because the pitch angle is negative downwards and needs to be a positive value.

[0104] Step S2.3.3, using the longitude and latitude of the photovoltaic power station (Lcam, λcam) and the ship target distance Dtarget, calculate the longitude and latitude of the ship, that is, the photoelectric position data:

[0105] Longitude change (Δλ):

[0106]

[0107] Latitude change (ΔL):

[0108]

[0109] Latitude and longitude of the ship target:

[0110] L target =L cam +ΔL

[0111] λ target =λ cam +Δλ

[0112] Where, (Lcam, λcam): latitude and longitude of the photovoltaic power station (unit: degree); Hcam: altitude of the photovoltaic power station (unit: meter); R: radius of the earth (about 6371km).

[0113] Step S3, transmitting the radar position data, the ship track and the optoelectronic position data to the data processing center to construct a complete ship navigation track.

[0114] Specifically, the data processing center first establishes a unified coordinate system, and then uses the Kalman filter algorithm to match and integrate the radar position data, ship trajectory and optoelectronic position data. Specifically, the Kalman filter algorithm is used to fuse multi-source data. The algorithm can weightedly fuse the data based on the uncertainty, measurement error and other factors of radar position data, ship trajectory data and optoelectronic position data, thereby effectively ensuring the smoothness and continuity of the trajectory, preventing trajectory deviations caused by data conflicts or inconsistencies, and finally obtaining a complete ship navigation trajectory.

[0115] Step S4, outputting the ship trajectory to the multi-sensing system to achieve real-time monitoring and management of the ship.

[0116] Specifically, in step S4, the multi-sensing system displays in real time the ship target identified by radar detection and optoelectronic detection of the optoelectronic station, and presents the trajectory information of the ship target.

[0117] To be more specific, when the multi-sensing system is turned on, the home page will display all identified ship targets. The user can right-click the ship target to be viewed and choose to view the tracking trajectory, tracking time and other information of the ship target.

[0118] Embodiment 2:

[0119] The present invention also provides a ship trajectory recognition system based on photoelectric blindness compensation. The ship trajectory recognition system based on photoelectric blindness compensation can be realized by executing the process steps of the ship trajectory recognition method based on photoelectric blindness compensation, that is, those skilled in the art can understand the ship trajectory recognition method based on photoelectric blindness compensation as a preferred implementation of the ship trajectory recognition system based on photoelectric blindness compensation.

[0120] Specifically, the ship trajectory identification system includes:

[0121] Module M1 uses radar to detect ships in unobstructed areas, obtains radar position data and simulates ship trajectories.

[0122] Module M2 sets up an optoelectronic station in the blind area where the radar is easily blocked. The optoelectronic detection is carried out through the optoelectronic station to obtain the optoelectronic position data of the ship to ensure uninterrupted channel monitoring.

[0123] Module M3 transmits radar position data, ship trajectory and optoelectronic position data to the data processing center to construct a complete ship navigation trajectory.

[0124] Module M4 outputs the ship trajectory to the multi-sensing system to achieve real-time monitoring and management of the ship.

[0125] Embodiment 3:

[0126] This embodiment provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the steps of a ship trajectory recognition method based on photoelectric blindness compensation in the above-mentioned embodiment 1 are implemented.

[0127] The present invention has a wide range of application scenarios, mainly including:

[0128] (1) Coastal and inland waterways: In ocean transportation, there are many islands, coastal tall buildings and other objects that may block radar signals. For example, in the waterways near busy ports, there are many tall buildings and radar signals are easily blocked. This method can be used to establish an optical power station at a suitable location on the shore to ensure complete monitoring of the ship's trajectory.

[0129] (2) Lakes and inland waters: There may be mountains or lakeside buildings around wide lakes. Due to the complex terrain or changeable climate, radar monitoring has limitations when ships travel near these obstructions. The present invention can effectively compensate for this limitation.

[0130] (3) Port and wharf areas: There are a large number of buildings and equipment in the port area, which can easily interfere with radar signals. The optoelectronic system can improve the accuracy of ship identification.

[0131] Those skilled in the art know that, in addition to realizing the system and its various devices, modules, and units provided by the present invention in a purely computer-readable program code, it is entirely possible to realize the same functions in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system and its various devices, modules, and units provided by the present invention can be considered as a hardware component, and the devices, modules, and units included therein for realizing various functions can also be regarded as structures within the hardware component; the devices, modules, and units for realizing various functions can also be regarded as both software modules for realizing the method and structures within the hardware component.

[0132] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A ship trajectory recognition method based on photoelectric blind spot filling, characterized in that: The following steps are involved: Step S1, using radar to detect ships in an unobstructed area, obtaining radar position data and simulating ship trajectories; Step S2, setting up an optoelectronic station in a blind area where the radar is easily blocked, performing optoelectronic detection through the optoelectronic station to obtain optoelectronic position data of the ship to ensure uninterrupted channel monitoring; Step S3, transmitting the radar position data, the ship track and the optoelectronic position data to a data processing center to construct a complete ship navigation track; Step S4, outputting the ship trajectory to a multi-sensing system to achieve real-time monitoring and management of the ship.

2. A ship trajectory recognition method based on photoelectric blindness compensation according to claim 1, characterized in that: The step S2 includes the following sub-steps: Step S2.1, image acquisition: the camera in the photovoltaic power station monitors the blind area in real time and acquires images of the ship; Step S2.2, target detection: the collected ship images are processed in real time using the Yolov5 target detection algorithm to obtain the ship target recognition result to ensure accurate recognition of the ship target; Step S2.3, photoelectric position data calculation: Based on the ship target recognition result, combined with the geographical coordinates and height of the photovoltaic power station and the pitch angle and field angle of the camera, the photoelectric position data is calculated through geometric relationships.

3. A ship trajectory recognition method based on photoelectric blindness compensation according to claim 2, characterized in that: The step S2.1 includes: constructing a photoelectric station: deploying a photoelectric station in the blind area, the photoelectric station is equipped with high-precision dual-spectrum photoelectric equipment, the site selection of the photoelectric station is based on the scope of the blocked area and the frequency of ship traffic, and the height of the photoelectric station is reasonably designed based on the surrounding environment and the field of view of the camera.

4. The ship trajectory recognition method based on photoelectric blindness compensation according to claim 2 is characterized in that: In step S2.1, the camera of the photovoltaic power station has two working modes: Visible light working mode: During the day, that is, from 4 am to 8 pm, the visible light camera is automatically enabled to capture the color information of the ship. The visible light camera analyzes and processes the collected ship images using the Yolov5 target detection algorithm to complete the detection of the ship target in the image; Infrared light working mode: At night, that is, from 5 pm to 7 am, the infrared camera is automatically enabled. The infrared camera analyzes and processes the collected ship images using the Yolov5 target detection algorithm to complete the detection of ship targets in the image; Mode overlap processing: When the two working modes overlap, that is, from 4:00 a.m. to 7:00 a.m. and from 5:00 p.m. to 8:00 p.m., the two working modes are started at the same time to perform target detection processing on the ship images collected by the visible light camera and the infrared camera respectively.

5. The ship trajectory recognition method based on photoelectric blindness compensation according to claim 2 is characterized in that: The step S2.2 includes the following sub-steps: Step S2.2.1, training of target detection network: based on the Yolov5 deep network, the ship image collected in step S2.1 is trained to obtain a detection network for ship target detection; Step S2.2.2, ship target detection optimization: Based on the detection network, the accuracy of ship target recognition is improved by expanding the training data and optimizing the target detection algorithm.

6. The ship track recognition method based on photoelectric blindness compensation according to claim 2 is characterized in that: The step S2.3 includes the following sub-steps: Step S2.3.1, calculate the sight angle of the ship target relative to the center of the camera according to the pixel position of the ship target in the picture: Horizontal deviation angle (Δφ): Vertical deviation angle (Δθ): Ship target azimuth (φtarget) and pitch angle (θtarget): f target =φ+Δφ i target =θ+Δθ Among them, θ: the pitch angle of the camera, positive value means upward, negative value means downward; φ: the azimuth angle of the camera, clockwise relative to the north; αh: the horizontal field of view of the camera; αv: the vertical field of view of the camera; pixel coordinates (xtarget, ytarget); picture resolution (Wframe, Hframe); Step S2.3.2, set the ship to be on the horizontal plane, and calculate the horizontal distance Dtarget of the ship using the geometric relationship according to the altitude Hcam of the photovoltaic power station and the pitch angle θtarget of the ship target: Where, (Lcam, λcam): latitude and longitude of the photovoltaic station; Hcam: altitude of the photovoltaic station; Step S2.3.3, using the longitude and latitude of the photovoltaic power station (Lcam, λcam) and the ship target distance Dtarget, calculate the longitude and latitude of the ship, that is, the photoelectric position data: Longitude change (Δλ): Latitude change (ΔL): Latitude and longitude of the ship target: L target =L cam +ΔL l target =λ cam +Dl Where, (Lcam, λcam): latitude and longitude of the photovoltaic station; Hcam: altitude of the photovoltaic station; R: radius of the earth.

7. The ship track recognition method based on photoelectric blindness compensation according to claim 1 is characterized in that: In step S3, the data processing center first establishes a unified coordinate system, and then uses a Kalman filter algorithm to match and integrate the radar position data, the ship trajectory and the photoelectric position data to obtain a complete ship navigation trajectory.

8. The ship track recognition method based on photoelectric blindness compensation according to claim 1 is characterized in that: In step S4, the multi-sensing system displays in real time the ship target identified by radar detection and optoelectronic detection of the optoelectronic station, and presents the trajectory information of the ship target.

9. A ship track recognition system based on photoelectric blindness compensation, using a ship track recognition method based on photoelectric blindness compensation as claimed in any one of claims 1 to 8, characterized in that: include: Module M1 uses radar to detect ships in unobstructed areas, obtains radar position data and simulates ship trajectories; Module M2, in the blind area where the radar is easily blocked, an optoelectronic station is set up to perform optoelectronic detection through the optoelectronic station to obtain the optoelectronic position data of the ship to ensure uninterrupted channel monitoring; Module M3, transmitting the radar position data, the ship track and the optoelectronic position data to a data processing center to construct a complete ship navigation track; Module M4 outputs the ship trajectory to the multi-sensing system to achieve real-time monitoring and management of the ship.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of a ship trajectory recognition method based on photoelectric blindness compensation according to any one of claims 1 to 8 are implemented.

Images