A beacon identification method and device based on video inspection and a storage medium

Abstract

A beacon identification method and device based on video inspection and a storage medium, which comprises: establishing a beacon data table for each beacon; performing multiple image acquisition on each beacon, and all acquisition images of each beacon forming an image material package; recording the first image acquisition data corresponding to each acquisition image, and saving the binding of both to the beacon data table; obtaining current image acquisition data, calculating the difference between the current image acquisition data and the first image acquisition data in each beacon data table; obtaining the minimum value and querying the corresponding beacon as the identification object; traversing the image material package of the identification object to determine the reference image; obtaining the first image acquisition data and the current image acquisition data of the reference image to obtain the shooting parameter; and shooting and identifying the identification object. Through a large number of image acquisition combined with traversing the image material package of the identification object, the reference image can be determined, and then the shooting parameter can be quickly calculated with high accuracy, thereby improving the beacon identification efficiency.

Description

Technical Field

[0001] This invention relates to the field of navigation mark inspection technology, and in particular to a navigation mark identification method, device, and computer-readable storage medium based on video inspection. Background Technology

[0002] Navigational aid inspection refers to a comprehensive and meticulous on-site inspection and maintenance of navigational aids to preventively identify and eliminate potential malfunctions, ensuring that the aids are always in normal working order. Navigational aid inspection is an important part of navigational aid management and a crucial means of continuously maintaining the characteristics of navigational aids and maximizing their navigational assistance effectiveness.

[0003] With the development and progress of science and technology, a large number of new equipment and materials have been applied and promoted in the field of navigation aids, and the methods of navigation aid maintenance and management have also shown a diversified development trend. At present, the technology of navigation aid recognition in video inspection business mainly involves calculating the azimuth of the navigation aid by using the real-time latitude and longitude coordinates of the ship and the preset latitude and longitude coordinates of the navigation aid. Then, the difference between the azimuth of the navigation aid and the azimuth (heading angle) of the ship is used to calculate the rotation angle and direction of the inspection equipment, thereby determining the shooting range.

[0004] However, since navigation marks float within a certain range at sea, the above methods can only calculate the approximate angle that the video inspection equipment needs to rotate. Therefore, the probability of capturing the navigation mark is not high, and even if the navigation mark is captured, it cannot be guaranteed that the captured image of the navigation mark will be identifiable.

[0005] In other words, existing technologies suffer from low accuracy in calculating shooting angles during video inspections, resulting in low efficiency in navigation mark recognition. Summary of the Invention

[0006] The main objective of this invention is to provide a navigation mark identification method, apparatus, device, and storage medium based on video inspection, aiming to solve the technical problem that the calculation accuracy of the shooting angle in existing video inspection-based navigation mark identification methods is not high, resulting in low navigation mark identification efficiency.

[0007] To achieve the above objectives, this invention provides a navigation mark recognition method based on video inspection, comprising the following steps: acquiring all navigation marks to be identified, establishing and initializing a navigation mark data table for each navigation mark; acquiring multiple images of each navigation mark according to a preset image acquisition angle and a preset image acquisition multiplier, and assembling all acquired images of each navigation mark into an image material package; recording the first image acquisition data corresponding to each acquired image, binding the two together, and saving them to the navigation mark data table; acquiring the current image acquisition data, calculating the difference between it and the first image acquisition data in each navigation mark data table; obtaining the minimum value among the differences, querying the corresponding navigation mark, and using that navigation mark as the recognition object; traversing the image material package of the recognition object, determining an acquired image as a reference image according to a preset reference standard; acquiring the first image acquisition data bound to the reference image, and obtaining the shooting parameters of the video inspection module based on the first image acquisition data bound to the reference image and the current image acquisition data; capturing the recognition object according to the shooting parameters to obtain the image to be identified; performing navigation mark recognition on the image to be identified, and uploading the recognition result.

[0008] Optionally, the preset image acquisition angle is specifically: the first range value is divided according to the preset shooting angle interval to obtain multiple preset image acquisition angles; the preset image acquisition magnification is specifically: the second range value is divided according to the preset shooting magnification to obtain multiple preset image acquisition magnifications.

[0009] Optionally, when binding the acquired image and its corresponding first image acquisition data to the navigation mark data table, the navigation mark's ID is used as the foreign key.

[0010] Optionally, the image material package of the object to be identified is traversed, and a captured image is determined as a reference image according to a preset reference standard. This includes at least the following steps: obtaining the image material package of the object to be identified by the ID of the beacon; traversing the image material package of the object to be identified, establishing a coordinate system with the center point of each captured image, using the center point of the beacon in the captured image as the coordinate point of the beacon, and calculating the distance between the beacon and the center point of the image in each captured image based on the coordinate point and the origin of the coordinate system; calculating the pixel percentage of the beacon in each captured image, specifically the proportion of the beacon's pixels among all pixels; sorting the pixel percentages and distances between the beacons and the center points of the images for all beacons, and selecting the captured image with the largest pixel percentage and the closest distance between the beacon and the center point of the image as the reference image; if the captured image with the largest pixel percentage and the captured image with the closest distance between the beacon and the center point of the image are not the same captured image, then the captured image with the largest pixel percentage is selected as the reference image.

[0011] Optionally, the first image acquisition data includes at least: the ship's latitude and longitude coordinates, the ship's heading angle, the turning angle of the image acquisition module, and the preset image acquisition multiplier when the image acquisition is performed; the current image acquisition data includes at least: the ship's current latitude and longitude coordinates, the ship's current heading angle, and the turning angle of the current video inspection module.

[0012] Optionally, the current image acquisition data is obtained, and the difference between it and the first image acquisition data in each navigation mark data table is calculated. Specifically, within a preset recognition range, the latitude and longitude coordinates of the current ship are obtained, and the difference between them and the latitude and longitude coordinates of the ship in each navigation mark data table is calculated.

[0013] Optionally, the first image acquisition data bound to the reference image is obtained. Based on the first image acquisition data bound to the reference image and the current image acquisition data, the shooting parameters of the video inspection module are obtained. Specifically, the heading angle of the ship bound to the reference image and the turning angle of the image acquisition module are obtained; the heading angle of the image acquisition module is obtained by summing the heading angle of the ship and the turning angle of the image acquisition module; the heading angle of the video inspection module is calculated based on the heading angle of the current ship and the turning angle of the current video inspection module; the difference between the heading angle of the video inspection module and the heading angle of the image acquisition module is calculated to obtain the rotation angle of the video inspection module; and the preset image acquisition magnification of the reference image is obtained and used together with the rotation angle as shooting parameters.

[0014] Optionally, the image to be identified is subjected to navigation mark recognition, and the recognition result is uploaded. Specifically, the image to be identified is subjected to navigation mark recognition. If the recognition is successful, the navigation mark is marked as recognized and uploaded. If the recognition fails, the navigation mark is marked as unrecognized and uploaded.

[0015] Corresponding to the aforementioned navigation mark recognition method based on video inspection, this invention provides a navigation mark recognition device based on video inspection, comprising: a navigation mark establishment module, used to acquire all navigation marks to be identified, establish a navigation mark data table for each navigation mark, and initialize it; an image acquisition module, used to acquire multiple images of each navigation mark according to a preset image acquisition angle and a preset image acquisition multiplier, and to assemble all acquired images of each navigation mark into an image material package; a recording module, used to record the first image acquisition data corresponding to each acquired image, and to bind the two and save them to the navigation mark data table; and a calculation module, used to acquire the current image acquisition data, calculate its relationship with each navigation mark data table. The system calculates the difference between the first image acquisition data and the current image acquisition data. It then calculates the minimum difference and queries the corresponding navigation mark, using that navigation mark as the identification object. A traversal module iterates through the image data package of the identification object, determining a reference image based on a preset reference standard. A shooting parameter determination module acquires the first image acquisition data bound to the reference image and, based on the first image acquisition data bound to the reference image and the current image acquisition data, obtains the shooting parameters for the video inspection module. The video inspection module then captures images of the identification object according to the shooting parameters to obtain the image to be identified. Finally, a navigation mark recognition module performs navigation mark recognition on the image to be identified and uploads the recognition results.

[0016] Furthermore, to achieve the above objectives, the present invention also provides a computer-readable storage medium storing a navigation mark identification method based on video inspection, wherein the navigation mark identification method based on video inspection, when executed by a processor, implements the steps of the navigation mark identification method based on video inspection as described above.

[0017] The beneficial effects of this invention are:

[0018] (1) Compared with the prior art, the present invention performs multiple image acquisitions on each navigation mark in the early stage of acquiring the image to be identified, based on the preset image acquisition angle and preset image acquisition multiple, which can cover the multi-angle state of all navigation marks that need to be identified. Then, by traversing the image material package of the object to be identified, the reference image can be determined, and the best shooting parameters of the navigation mark to be identified can be quickly calculated. Since a large amount of image acquisition data is stored in the early stage, the accuracy of the calculation is high, and a high-quality image to be identified can be obtained, which can effectively improve the efficiency of navigation mark identification.

[0019] (2) Compared with the prior art, the present invention divides the first range value by a preset shooting angle interval to obtain multiple preset image acquisition angles; and divides the second range value by a preset shooting magnification factor to obtain multiple preset image acquisition factors; thus ensuring the diversity of image material packages for each navigation mark that needs to be identified and ensuring the accuracy of the reference image.

[0020] (3) Compared with the prior art, the present invention uses the id of the navigation beacon as a foreign key when the acquired image and its corresponding first image acquisition data are bound and saved to the navigation beacon data table, so that the corresponding image material package can be obtained through the id of the navigation beacon in the future.

[0021] (4) Compared with the prior art, the present invention traverses the image material package of the object to be identified and determines a captured image as a reference image according to a preset reference standard, so as to provide a reference for the navigation mark recognition image and facilitate the subsequent calculation of the rotation angle of the video inspection module to improve the recognition efficiency.

[0022] (5) Compared with the prior art, the present invention obtains the rotation angle of the video inspection module by calculating the difference between the heading angle of the video inspection module and the heading angle of the image acquisition module; obtains the preset image acquisition multiple of the reference image, and uses it together with the rotation angle as shooting parameters, which can obtain a high-quality image to be identified, which is convenient for navigation mark identification and can also improve the accuracy of navigation mark identification. Attached Figure Description

[0023] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings:

[0024] Figure 1 This is a simplified flowchart of an embodiment of the navigation mark identification method based on video inspection of the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] like Figure 1As shown, the present invention provides a navigation mark recognition method based on video inspection, which includes the following steps: acquiring all navigation marks to be identified, establishing a navigation mark data table for each navigation mark and initializing it; acquiring multiple images of each navigation mark according to a preset image acquisition angle and a preset image acquisition multiplier, and forming an image material package from all acquired images of each navigation mark; recording the first image acquisition data corresponding to each acquired image, binding the two together and saving them to the navigation mark data table; acquiring the current image acquisition data and calculating the difference between it and the first image acquisition data in each navigation mark data table; acquiring the minimum value among the differences and querying the corresponding navigation mark, and using that navigation mark as the recognition object; traversing the image material package of the recognition object, and determining an acquired image as a reference image according to a preset reference standard; acquiring the first image acquisition data bound to the reference image, and obtaining the shooting parameters of the video inspection module based on the first image acquisition data bound to the reference image and the current image acquisition data; shooting the recognition object according to the shooting parameters to obtain the image to be identified; performing navigation mark recognition on the image to be identified, and uploading the recognition result.

[0027] In the early stage of acquiring the image to be identified, this invention performs multiple image acquisitions on each navigation mark according to a preset image acquisition angle and a preset image acquisition multiplier, which can cover the multi-angle states of all navigation marks that need to be identified. Then, by traversing the image material package of the object to be identified, a reference image can be determined, and the optimal shooting parameters of the navigation mark to be identified can be quickly calculated. Since a large amount of image acquisition data is stored in the early stage, the accuracy of the calculation is high, and high-quality images to be identified can be obtained, which can effectively improve the efficiency of navigation mark identification.

[0028] In this embodiment, the preset image acquisition angle is specifically defined as follows: the first range value is divided according to the preset shooting angle interval to obtain multiple preset image acquisition angles; the preset image acquisition magnification is specifically defined as follows: the second range value is divided according to the preset shooting magnification to obtain multiple preset image acquisition magnifications.

[0029] Preferably, the first range is 0°-360°, with a preset shooting angle interval of 1°; the second range is 10x-30x, with a preset shooting magnification of 1x. That is, the 0°-360° range is divided according to the 1° shooting angle interval to obtain 360 preset image acquisition angles; the second range is divided according to the preset shooting magnification of 1x to obtain 30 preset image acquisition magnifications.

[0030] Preferably, the image material package of the object to be identified will contain images of the same point captured from multiple angles and at multiple magnifications, so it is necessary to traverse all the captured images in the image material package.

[0031] This invention divides a first range value by a preset shooting angle interval to obtain multiple preset image acquisition angles; and divides a second range value by a preset shooting magnification factor to obtain multiple preset image acquisition factors; thus ensuring the diversity of image material packages for each navigation mark to be identified and guaranteeing the accuracy of the reference image.

[0032] In this embodiment, when the acquired image and its corresponding first image acquisition data are bound and saved to the navigation mark data table, the id of the navigation mark is used as the foreign key.

[0033] When the acquired image and its corresponding first image acquisition data are bound together and saved to the navigation mark data table, the present invention uses the navigation mark's ID as a foreign key, which facilitates the subsequent retrieval of the corresponding image material package through the navigation mark's ID.

[0034] In this embodiment, traversing the image material package of the object to be identified and determining a captured image as a reference image according to a preset reference standard includes at least the following steps: obtaining the image material package of the object to be identified by the ID of the beacon; traversing the image material package of the object to be identified, establishing a coordinate system with the center point of each captured image, using the center point of the beacon in the captured image as the coordinate point of the beacon, and calculating the distance between the beacon and the center point of the image in each captured image based on the coordinate point and the origin of the coordinate system; calculating the pixel percentage of the beacon in each captured image, specifically the proportion of the beacon's pixels among all pixels; sorting the pixel percentages and distances between the beacons and the center points of the images for all beacons, and selecting the captured image with the largest pixel percentage and the closest distance between the beacon and the center point of the image as the reference image; if the captured image with the largest pixel percentage and the captured image with the closest distance between the beacon and the center point of the image are not the same captured image, then the captured image with the largest pixel percentage is selected as the reference image.

[0035] This invention traverses the image material package of the object to be identified and determines a captured image as a reference image according to a preset reference standard, providing a reference for the navigation mark identification image and facilitating the subsequent calculation of the rotation angle of the video inspection module to improve the identification efficiency.

[0036] In this embodiment, the first image acquisition data includes at least: the latitude and longitude coordinates of the ship, the heading angle of the ship, the turning angle of the image acquisition module, and the preset image acquisition multiplier when the image acquisition is performed; the current image acquisition data includes at least: the latitude and longitude coordinates of the current ship, the heading angle of the current ship, and the turning angle of the current video inspection module.

[0037] In this embodiment, the current image acquisition data is obtained, and the difference between it and the first image acquisition data in each navigation mark data table is calculated. Specifically, within a preset recognition range, the latitude and longitude coordinates of the current ship are obtained, and the difference between them and the latitude and longitude coordinates of the ship in each navigation mark data table is calculated.

[0038] Preferably, the preset identification range is 1-2 nautical miles. Among them, 2 nautical miles is the effective identification range, and 1 nautical mile is the optimal identification range.

[0039] In this embodiment, the first image acquisition data bound to the reference image is obtained. Based on the first image acquisition data bound to the reference image and the current image acquisition data, the shooting parameters of the video inspection module are obtained. Specifically, the heading angle of the ship bound to the reference image and the turning angle of the image acquisition module are obtained; the heading angle of the image acquisition module is obtained by summing the heading angle of the ship and the turning angle of the image acquisition module; the heading angle of the video inspection module is calculated based on the current heading angle of the ship and the current turning angle of the video inspection module; the difference between the heading angle of the video inspection module and the heading angle of the image acquisition module is calculated to obtain the rotation angle of the video inspection module; and the preset image acquisition magnification of the reference image is obtained and used together with the rotation angle as shooting parameters.

[0040] This invention calculates the difference between the heading angle of the video inspection module and the heading angle of the image acquisition module to obtain the rotation angle of the video inspection module; it obtains the preset image acquisition multiplier of the reference image and uses it together with the rotation angle as shooting parameters, which can obtain high-quality images to be identified, facilitate navigation mark identification, and improve the accuracy of navigation mark identification.

[0041] Preferably, the image acquisition module and the video inspection module are cameras.

[0042] In this embodiment, navigation mark recognition is performed on the image to be recognized, and the recognition result is uploaded. Specifically, the navigation mark recognition is performed on the image to be recognized. If the recognition is successful, the navigation mark is marked as recognized and uploaded. If the recognition fails, the navigation mark is marked as unrecognized and uploaded.

[0043] The present invention also provides a navigation mark identification device based on video inspection, which includes: in this embodiment, acquiring first image acquisition data bound to a reference image; obtaining shooting parameters of the video inspection module based on the first image acquisition data bound to the reference image and the current image acquisition data, specifically: acquiring the heading angle of the ship bound to the reference image and the turning angle of the image acquisition module; summing the heading angle of the ship and the turning angle of the image acquisition module to obtain the heading angle of the image acquisition module; calculating the heading angle of the video inspection module based on the current heading angle of the ship and the current turning angle of the video inspection module; calculating the difference between the heading angle of the video inspection module and the heading angle of the image acquisition module to obtain the rotation angle of the video inspection module; acquiring a preset image acquisition multiplier of the reference image, and using it together with the rotation angle as shooting parameters.

[0044] This invention also provides a computer-readable storage medium, which may be a computer-readable storage medium included in the memory described in the above embodiments; or it may be a standalone computer-readable storage medium not assembled into a device. The computer-readable storage medium stores at least one instruction, which is loaded and executed by a processor to implement... Figure 1 The method for identifying navigational aids based on video inspection is shown. The computer-readable storage medium can be a read-only memory, a hard disk, or an optical disk, etc.

[0045] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the device embodiments, equipment embodiments, and storage medium embodiments, since they are basically similar to the method embodiments, the descriptions are relatively simple, and relevant parts can be referred to the descriptions of the method embodiments.

[0046] Furthermore, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0047] The foregoing description illustrates and describes preferred embodiments of the present invention. It should be understood that the present invention is not limited to the forms disclosed herein and should not be construed as excluding other embodiments. It can be used in various other combinations, modifications, and environments, and can be altered within the scope of the inventive concept by means of the foregoing teachings or techniques or knowledge in related fields. Any modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention should be within the protection scope of the appended claims.

Claims

1. A navigation mark identification method based on video inspection, characterized in that, Includes the following steps: Obtain all navigation marks that need to be identified, create a navigation mark data table for each navigation mark and initialize it; Based on the preset image acquisition angle and preset image acquisition magnification, multiple image acquisitions are performed on each navigation mark, and all acquired images of each navigation mark are combined into an image material package. Record the first image acquisition data corresponding to each acquired image, and save the two data together to the navigation beacon data table; Acquire the current image acquisition data and calculate the difference between it and the first image acquisition data in each navigation mark data table; find the minimum value among the differences and query the corresponding navigation mark, and use that navigation mark as the identification object; Iterate through the image material package of the object to be identified, and determine a captured image as a reference image according to a preset reference standard; Obtain the first image acquisition data bound to the reference image, and obtain the shooting parameters of the video inspection module based on the first image acquisition data bound to the reference image and the current image acquisition data; The object to be identified is photographed according to the shooting parameters to obtain the image to be identified; Perform navigation mark identification on the image to be identified and upload the identification results; When the captured image and its corresponding first image capture data are bound together and saved to the navigation mark data table, the navigator's ID is used as the foreign key; Traversing the image material package of the object to be identified, and determining a captured image as a reference image according to a preset reference standard, includes at least the following steps: The image data package of the target object is obtained by using the beacon's ID. The image data package is then traversed, and a coordinate system is established with the center point of each captured image as the coordinate point of the beacon. The distance between the beacon and the center point of the image is calculated based on this coordinate point and the origin of the coordinate system. The pixel percentage of the beacon in each captured image is calculated, specifically the proportion of the beacon's pixels among all pixels. All beacon pixel percentages and distances to the center point are sorted, and the image with the largest pixel percentage and the closest distance to the center point is selected as the reference image. If the image with the largest pixel percentage and the image with the closest distance to the center point are not the same image, the image with the largest pixel percentage is selected as the reference image. The first image acquisition data includes at least: the ship's latitude and longitude coordinates, the ship's heading angle, the turning angle of the image acquisition module, and the preset image acquisition multiplier during image acquisition; The current image acquisition data includes at least: the current latitude and longitude coordinates of the ship, the current heading angle of the ship, and the current turning angle of the video inspection module; Acquire the current image acquisition data and calculate the difference between it and the first image acquisition data in each navigation mark data table. Specifically, within the preset recognition range, acquire the latitude and longitude coordinates of the current ship and calculate the difference between it and the latitude and longitude coordinates of the ship in each navigation mark data table.

2. The navigation mark identification method based on video inspection according to claim 1, characterized in that: The preset image acquisition angle is specifically defined as follows: the first range value is divided according to the preset shooting angle interval to obtain multiple preset image acquisition angles; The preset image acquisition magnification is specifically defined as follows: the second range value is divided according to the preset shooting magnification to obtain multiple preset image acquisition magnifications.

3. The navigation mark identification method based on video inspection according to claim 1, characterized in that: Obtain the first image acquisition data bound to the reference image. Based on the first image acquisition data bound to the reference image and the current image acquisition data, obtain the shooting parameters of the video inspection module, specifically: Obtain the heading angle of the ship bound to the reference image and the turning angle of the image acquisition module; The heading angle of the image acquisition module is obtained by summing the heading angle of the ship and the turning angle of the image acquisition module. The heading angle of the video inspection module is calculated based on the current heading angle of the ship and the current turning angle of the video inspection module. The difference between the heading angle of the video inspection module and the heading angle of the image acquisition module is calculated to obtain the rotation angle of the video inspection module; Obtain the preset image acquisition magnification of the reference image and use it together with the rotation angle as the shooting parameters.

4. The navigation mark identification method based on video inspection according to claim 1, characterized in that: The image to be identified is subjected to navigation mark recognition, and the recognition results are uploaded. Specifically: The system performs navigation mark recognition on the image to be identified. If the recognition is successful, the navigation mark is marked as identified and uploaded. If the recognition fails, the navigation mark is marked as unrecognized and uploaded.

5. A navigation mark identification device based on video inspection, characterized in that, include: The navigation mark establishment module is used to acquire all navigation marks that need to be identified, establish a navigation mark data table for each navigation mark and initialize it. The image acquisition module is used to acquire multiple images of each navigation mark according to the preset image acquisition angle and preset image acquisition magnification. All the acquired images of each navigation mark are combined into an image material package. The recording module is used to record the first image acquisition data corresponding to each acquired image, and save the two to the navigation mark data table after binding them; when saving the acquired image and its corresponding first image acquisition data to the navigation mark data table, the id of the navigation mark is used as the foreign key; The first image acquisition data includes at least: the ship's latitude and longitude coordinates, the ship's heading angle, the turning angle of the image acquisition module, and the preset image acquisition multiplier during image acquisition; The calculation module is used to acquire the current image acquisition data, calculate the difference between it and the first image acquisition data in each navigation mark data table; obtain the minimum value of the difference, query the corresponding navigation mark, and use the navigation mark as the identification object; the current image acquisition data includes at least: the latitude and longitude coordinates of the current ship, the heading angle of the current ship, and the turning angle of the current video inspection module; acquiring the current image acquisition data and calculating the difference between it and the first image acquisition data in each navigation mark data table specifically involves: within a preset recognition range, acquiring the latitude and longitude coordinates of the current ship, and calculating the difference between it and the latitude and longitude coordinates of the ship in each navigation mark data table; The traversal module is used to traverse the image material package of the object to be identified, and determine a captured image as a reference image according to a preset reference standard. It includes at least the following steps: The image data package of the target object is obtained by using the beacon's ID. The image data package is then traversed, and a coordinate system is established with the center point of each captured image as the coordinate point of the beacon. The distance between the beacon and the center point of the image is calculated based on this coordinate point and the origin of the coordinate system. The pixel percentage of the beacon in each captured image is calculated, specifically the proportion of the beacon's pixels among all pixels. All beacon pixel percentages and distances to the center point are sorted, and the image with the largest pixel percentage and the closest distance to the center point is selected as the reference image. If the image with the largest pixel percentage and the image with the closest distance to the center point are not the same image, the image with the largest pixel percentage is selected as the reference image. The shooting parameter determination module is used to obtain the first image acquisition data bound to the reference image, and to obtain the shooting parameters of the video inspection module based on the first image acquisition data bound to the reference image and the current image acquisition data. The video inspection module is used to capture images of the objects to be identified based on the shooting parameters, thereby obtaining images to be identified. The navigation mark recognition module is used to identify navigation marks in the image to be identified and upload the recognition results.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a navigation mark identification method based on video inspection, which, when executed by a processor, implements the steps of the navigation mark identification method based on video inspection as described in any one of claims 1 to 4.

Images