Fake-licensed ship identification method, device and equipment, storage medium and program product

Abstract

The invention provides a fake-licensed ship identification method and device, equipment, a storage medium and a program product. The method comprises the following steps: according to initial ship image data and position information of a to-be-verified ship, obtaining an image of the to-be-verified ship and a ship number of the to-be-verified ship; obtaining a comparison image of the recorded ship corresponding to the ship number according to the ship number and the pose data; calculating the two-dimensional feature similarity of the to-be-verified ship image and the comparison image by adopting a preset algorithm; obtaining three-dimensional point cloud data of the ship to be verified according to the image of the ship to be verified and the internal and external parameters of the first equipment; calculating three-dimensional feature similarity according to the three-dimensional point cloud data and the three-dimensional point cloud comparison data; according to the two-dimensional feature similarity and the three-dimensional feature similarity, comprehensive similarity is obtained through calculation; if it is judged that the comprehensive similarity does not exceed a preset similarity threshold value, a recognition result that the to-be-verified ship is the fake-licensed ship is generated. The accuracy and reliability of fishing boat identity recognition are improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, device, equipment, storage medium and program product for identifying counterfeit ships. Background Art

[0002] In recent years, due to insufficient manpower and backward evidence collection methods, the problem of counterfeit fishing vessels has become serious. Therefore, it is urgent to adopt modern scientific and technological means to improve the efficiency of verifying counterfeit fishing vessels.

[0003] At present, the main identification schemes for counterfeit fishing vessels are: using the AIS data sequence of the fishing vessel to identify the ship's identity, or judging whether the fishing vessel is a counterfeit by the ship's movement trajectory.

[0004] However, existing identification schemes all have identification loopholes and are unable to accurately identify counterfeit fishing vessels. Summary of the invention

[0005] The present application provides a method, device, equipment, storage medium and program product for identifying counterfeit vessels, which are used to solve the technical problem that the existing identification solutions all have identification loopholes and cannot accurately identify counterfeit fishing vessels.

[0006] In a first aspect, the present application provides a method for identifying a counterfeit ship, which is applied to a service device, comprising:

[0007] Receive initial ship image data, position information of the ship to be verified, and internal and external parameters of the first device sent by a data acquisition device, wherein the initial ship image data includes: an initial ship image and position data of the data acquisition device, wherein the position information of the ship to be verified is obtained by the data acquisition device in response to a user's operation on a ship selection page to be verified, wherein the ship selection page to be verified is generated by the data acquisition device according to the initial ship image, and wherein the internal and external parameters of the first device are generated by the data acquisition device acquiring its own internal and external parameters;

[0008] Obtaining the image of the vessel to be verified and the license plate number of the vessel to be verified according to the initial vessel image data and the position information of the vessel to be verified;

[0009] According to the ship license plate number and the position data, a comparison image of the recorded ship corresponding to the ship license plate number and the image of the ship to be verified at the same shooting angle is obtained;

[0010] Calculating the similarity of the two-dimensional features between the image of the vessel to be verified and the comparison image using a preset algorithm;

[0011] Obtaining three-dimensional point cloud data of the vessel to be verified according to the image of the vessel to be verified and internal and external parameters of the first device;

[0012] According to the ship license plate number, query is performed in a preset three-dimensional point cloud database to obtain three-dimensional point cloud comparison data corresponding to the ship license plate number;

[0013] Calculating three-dimensional feature similarity based on the three-dimensional point cloud data and the three-dimensional point cloud comparison data;

[0014] Calculating a comprehensive similarity based on the two-dimensional feature similarity and the three-dimensional feature similarity;

[0015] If it is determined that the comprehensive similarity does not exceed the preset similarity threshold, a recognition result that the vessel to be verified is a counterfeit vessel is generated.

[0016] Optionally, the method as described above, wherein, based on the license plate number and the posture data, a comparison image of the recorded vessel corresponding to the license plate number and the image of the vessel to be verified at the same shooting angle is obtained, comprising: performing query and matching in a preset image multi-angle generation model library based on the license plate number to obtain a preset image multi-angle generation model corresponding to the license plate number; and inputting the posture data into the image multi-angle generation model to obtain a comparison image of the recorded vessel corresponding to the license plate number and the image of the vessel to be verified at the same shooting angle.

[0017] Optionally, in the method as described above, the using a preset algorithm to calculate the two-dimensional feature similarity between the image of the ship to be verified and the comparison image includes: using a preset algorithm to calculate the cosine similarity between the image of the ship to be verified and the comparison image; and converting the cosine similarity into two-dimensional feature similarity.

[0018] Optionally, in the method as described above, obtaining the three-dimensional point cloud data of the vessel to be verified based on the image of the vessel to be verified and the internal and external parameters of the first device includes: inputting the image of the vessel to be verified into a preset deep learning model to generate depth information of the vessel to be verified; and performing three-dimensional reconstruction of the vessel to be verified based on the depth information and the internal and external parameters of the first device to obtain the three-dimensional point cloud data of the vessel to be verified.

[0019] Optionally, in the method as described above, the three-dimensional feature similarity is calculated based on the three-dimensional point cloud data and the three-dimensional point cloud comparison data, including: obtaining the nearest point distance and the two-way distance of the three-dimensional point cloud data and the three-dimensional point cloud comparison data; converting the nearest point distance and the two-way distance to obtain the nearest point distance similarity and the two-way distance similarity; and calculating the three-dimensional feature similarity based on the nearest point distance similarity and the two-way distance similarity.

[0020] Optionally, in the method as described above, the three-dimensional point cloud comparison data includes internal and external parameters of the second device; accordingly, obtaining the closest point distance and two-way distance between the three-dimensional point cloud data and the three-dimensional point cloud comparison data includes: projecting the three-dimensional point cloud data and the three-dimensional point cloud comparison data to the same coordinate system according to the internal and external parameters of the first device and the internal and external parameters of the second device; and calculating the closest point distance and two-way distance between the three-dimensional point cloud data and the three-dimensional point cloud comparison data in the coordinate system.

[0021] Optionally, in the method as described above, obtaining the image of the ship to be verified and the license plate number of the ship to be verified based on the initial ship image data and the position information of the ship to be verified includes: performing image segmentation on the initial ship image according to the position information of the ship to be verified to obtain the image of the ship to be verified; and using a preset recognition algorithm to recognize the image of the ship to be verified to obtain the license plate number of the ship to be verified.

[0022] Optionally, in the method as described above, if it is determined that the comprehensive similarity does not exceed a preset similarity threshold, then after generating an identification result that the vessel to be verified is a counterfeit vessel, it also includes: outputting an alarm prompt, and sending the comprehensive similarity and the identification result to the platform end, so that the platform end displays a page of identification results of the vessel to be verified in response to the user clicking to view the result, wherein the page of identification results of the vessel to be verified includes the comprehensive similarity and the identification result.

[0023] Optionally, in the method described above, the conversion formula for converting the cosine similarity into two-dimensional feature similarity is:

[0024]

[0025] Where P img is the two-dimensional feature similarity, and A is the cosine similarity.

[0026] Optionally, in the method described above, the closest point distance and the bidirectional distance are converted to obtain the closest point distance similarity and the bidirectional distance similarity, and the conversion formula is:

[0027] S(d) n =exp(-α*d n )

[0028] Where S(d) n is the nearest point distance similarity, d n is the nearest point distance, exp is the exponential decay function, and α is the positive dynamic scaling factor;

[0029] S(d) d=exp(-α*d d )

[0030] Where S(d) d is the bidirectional distance similarity, d d is the two-way distance, exp is the exponential decay function, and α is a positive dynamic scaling factor.

[0031] Optionally, in the above method, the three-dimensional feature similarity is calculated based on the closest point distance similarity and the bidirectional distance similarity, and the calculation formula is:

[0032]

[0033] Where P voc is the three-dimensional feature similarity, S(d) n is the nearest point distance similarity, S(d) d is the two-way distance similarity.

[0034] Optionally, in the method described above, the calculation formula for calculating the comprehensive similarity based on the two-dimensional feature similarity and the three-dimensional feature similarity is:

[0035] W=aP img +bP voc , 0 <a<1,0<b<1,a+b=1

[0036] Where W is the comprehensive similarity, P img is the two-dimensional feature similarity, P voc is the three-dimensional feature similarity, a and b are dynamic weight parameters.

[0037] In a second aspect, the present application provides a method for identifying counterfeit ships, which is applied to a data collection device, comprising:

[0038] Collecting initial ship image data, wherein the initial ship image data includes: an initial ship image and position and posture data of the data acquisition device;

[0039] Generating a to-be-verified vessel selection page according to the initial vessel image;

[0040] In response to the user's operation on the to-be-verified vessel selection page, obtaining the location information of the to-be-verified vessel;

[0041] Acquire its own internal and external parameters to generate the internal and external parameters of the first device;

[0042] The initial ship image data, the location information of the ship to be verified, and the internal and external parameters of the first device are sent to the service device, so that the service device obtains the image of the ship to be verified and the license plate number of the ship to be verified according to the initial ship image data and the location information of the ship to be verified, obtains the recorded ship corresponding to the license plate number and the comparison image with the same shooting angle as the image of the ship to be verified according to the license plate number and the posture data, calculates the two-dimensional feature similarity between the image of the ship to be verified and the comparison image by using a preset algorithm, obtains the three-dimensional point cloud data of the ship to be verified according to the image of the ship to be verified and the internal and external parameters of the first device, performs a query in a preset three-dimensional point cloud database according to the license plate number to obtain the three-dimensional point cloud comparison data corresponding to the license plate number, calculates the three-dimensional feature similarity according to the three-dimensional point cloud data and the three-dimensional point cloud comparison data, calculates the comprehensive similarity according to the two-dimensional feature similarity and the three-dimensional feature similarity, and generates a recognition result that the ship to be verified is a fake license plate ship if it is determined that the comprehensive similarity does not exceed a preset similarity threshold.

[0043] Optionally, in the method as described above, obtaining the location information of the vessel to be verified in response to the user's operation on the vessel selection page to be verified includes: obtaining a prompt word in response to the user's input operation on the vessel selection page to be verified; inputting the prompt word into a preset target detection algorithm to obtain the location information of the vessel to be verified; or, obtaining the location information of the vessel to be verified in response to the user's selection operation on the vessel selection page to be verified.

[0044] In a third aspect, the present application provides a device for identifying counterfeit ships, which is applied to service equipment, including:

[0045] A receiving module, configured to receive initial ship image data, position information of a ship to be verified, and internal and external parameters of a first device sent by a data acquisition device, wherein the initial ship image data includes: an initial ship image and position data of the data acquisition device, wherein the position information of the ship to be verified is obtained by the data acquisition device in response to an operation of a user on a ship selection page to be verified, wherein the ship selection page to be verified is generated by the data acquisition device according to the initial ship image, and wherein the internal and external parameters of the first device are generated by the data acquisition device acquiring its own internal and external parameters;

[0046] An acquisition module, used for obtaining the image of the vessel to be verified and the license plate number of the vessel to be verified according to the initial vessel image data and the position information of the vessel to be verified;

[0047] The acquisition module is further used to obtain, based on the ship license number and the position data, a comparison image of the recorded ship corresponding to the ship license number and the image of the ship to be verified at the same shooting angle;

[0048] A calculation module, used for calculating the similarity of the two-dimensional features between the image of the vessel to be verified and the comparison image by using a preset algorithm;

[0049] The acquisition module is further used to obtain the three-dimensional point cloud data of the vessel to be verified according to the image of the vessel to be verified and the internal and external parameters of the first device;

[0050] A query module, used to query a preset three-dimensional point cloud database according to the ship license plate number to obtain three-dimensional point cloud comparison data corresponding to the ship license plate number;

[0051] The calculation module is further used to calculate the three-dimensional feature similarity based on the three-dimensional point cloud data and the three-dimensional point cloud comparison data;

[0052] The calculation module is further used to calculate the comprehensive similarity based on the two-dimensional feature similarity and the three-dimensional feature similarity;

[0053] The generating module is used for generating a recognition result that the vessel to be verified is a counterfeit vessel if it is determined that the comprehensive similarity does not exceed a preset similarity threshold.

[0054] In a fourth aspect, the present application provides a device for identifying counterfeit ships, which is applied to a data collection device, comprising:

[0055] An acquisition module, used for acquiring initial ship image data, wherein the initial ship image data comprises: an initial ship image and position and posture data of the data acquisition device;

[0056] A generating module, used for generating a to-be-verified vessel selection page according to the initial vessel image;

[0057] An acquisition module, configured to obtain the location information of the vessel to be verified in response to the user's operation on the vessel selection page to be verified;

[0058] The acquisition module is further used to acquire its own internal and external parameters to generate the internal and external parameters of the first device;

[0059] a sending module, configured to send the initial ship image data, the location information of the ship to be verified, and the internal and external parameters of the first device to the service device, so that the service device obtains the image of the ship to be verified and the license plate number of the ship to be verified according to the initial ship image data and the location information of the ship to be verified, obtains the comparison image of the recorded ship corresponding to the license plate number and the same shooting angle as the image of the ship to be verified according to the license plate number and the posture data, calculates the two-dimensional feature similarity between the image of the ship to be verified and the comparison image by using a preset algorithm, obtains the three-dimensional point cloud data of the ship to be verified according to the image of the ship to be verified and the internal and external parameters of the first device, performs a query in a preset three-dimensional point cloud database according to the license plate number to obtain the three-dimensional point cloud comparison data corresponding to the license plate number, calculates the three-dimensional feature similarity according to the three-dimensional point cloud data and the three-dimensional point cloud comparison data, calculates the comprehensive similarity according to the two-dimensional feature similarity and the three-dimensional feature similarity, and generates a recognition result that the ship to be verified is a fake license plate ship if it is determined that the comprehensive similarity does not exceed a preset similarity threshold.

[0060] In a fifth aspect, the present application provides a service device, comprising: a processor, and a memory communicatively connected to the processor;

[0061] The memory stores computer-executable instructions;

[0062] The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor performs the method for identifying counterfeit ships as described in the first aspect and various possible designs of the first aspect.

[0063] In a sixth aspect, the present application provides a data acquisition device, comprising: a processor, and a memory communicatively connected to the processor;

[0064] The memory stores computer-executable instructions;

[0065] The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor performs the method for identifying counterfeit ships as described in the second aspect and various possible designs of the second aspect.

[0066] In the seventh aspect, an embodiment of the present application provides a computer-readable storage medium, which stores computer execution instructions. When the computer execution instructions are executed by a processor, they are used to implement the method described in the first aspect and various possible designs of the first aspect or the second aspect and various possible designs of the second aspect.

[0067] In an eighth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the method described in the first aspect and various possible designs of the first aspect or the method described in the second aspect and various possible designs of the second aspect.

[0068] The method, device, equipment, storage medium and program product for identifying counterfeit vessels provided by the present application calculate the similarity of the two-dimensional features between the image of the vessel to be verified and the comparison image by using a preset algorithm; calculate the three-dimensional feature similarity based on the three-dimensional point cloud data and the three-dimensional point cloud comparison data; calculate the comprehensive similarity based on the two-dimensional feature similarity and the three-dimensional feature similarity; if it is determined that the comprehensive similarity does not exceed the preset similarity threshold, then generate an identification result that the vessel to be verified is a counterfeit vessel. The accuracy and reliability of fishing vessel identification are improved by combining the two-dimensional features and the three-dimensional features for comprehensive analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0070] Figure 1 A schematic diagram of a scenario of a method for identifying counterfeit ships provided in an embodiment of the present application;

[0071] Figure 2 A schematic diagram of a flow chart of a method for identifying counterfeit licensed vessels provided in one embodiment of the present application;

[0072] Figure 3 A schematic diagram of a flow chart of a method for identifying counterfeit ships provided in another embodiment of the present application;

[0073] Figure 4 A schematic diagram of the structure of a device for identifying counterfeit ships provided in an embodiment of the present application;

[0074] Figure 5 A schematic diagram of the structure of another device for identifying counterfeit ships provided in an embodiment of the present application;

[0075] Figure 6 A schematic diagram of the hardware structure of the service device provided in the embodiment of the present application;

[0076] Figure 7 A schematic diagram of the hardware structure of the data acquisition device provided in an embodiment of the present application.

[0077] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0078] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0079] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0080] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0081] Figure 1 A schematic diagram of a scenario of a method for identifying counterfeit ships provided in an embodiment of the present application. Figure 1 As shown, the scenario provided in this embodiment includes a data collection device 101 , a service device 102 and a platform end 103 .

[0082] The data acquisition device 101 may be a PDA (Personal Digital Assistant) device, an industrial camera, a mobile phone, etc.

[0083] The service device 102 may be a server. Optionally, it may be one server or a cluster consisting of multiple servers.

[0084] The platform end 103 may be a network management platform.

[0085] Specifically, the data acquisition device 101 sends the initial ship image data, the location information of the ship to be verified, and the internal and external parameters of the first device to the service device 102; the service device 102 processes the initial ship image data, the location information of the ship to be verified, and the internal and external parameters of the first device to obtain a comprehensive similarity and a recognition result; the service device 102 sends the comprehensive similarity and the recognition result to the platform end 103.

[0086] Figure 2 A flowchart of a method for identifying counterfeit ships provided in one embodiment of the present application. The execution subject of this embodiment may be Figure 1 The service device 102 shown in FIG. 1 may also be other computer devices, which is not particularly limited in this embodiment. Figure 2 As shown, the method includes:

[0087] S201: Receive initial ship image data, position information of the ship to be verified, and internal and external parameters of the first device sent by a data acquisition device, wherein the initial ship image data includes: an initial ship image and position data of the data acquisition device, wherein the position information of the ship to be verified is obtained by the data acquisition device in response to a user's operation on a ship selection page to be verified, wherein the ship selection page to be verified is generated by the data acquisition device based on the initial ship image, and wherein the internal and external parameters of the first device are generated by the data acquisition device obtaining its own internal and external parameters.

[0088] S202: Obtaining the image of the vessel to be verified and the license plate number of the vessel to be verified according to the initial vessel image data and the position information of the vessel to be verified.

[0089] Specifically, S202 specifically includes S2021-S2022:

[0090] S2021: Perform image segmentation on the initial ship image according to the position information of the ship to be verified to obtain the ship image to be verified.

[0091] Specifically, a preset image segmentation model is used to perform image segmentation on the initial ship image according to the position information of the ship to be verified, so as to obtain the ship image to be verified.

[0092] The preset image segmentation model may be an interactive segmentation model.

[0093] The interactive segmentation model includes but is not limited to the EdgeSAM model.

[0094] S2022: Using a preset recognition algorithm, the image of the vessel to be verified is recognized to obtain the license plate number of the vessel to be verified.

[0095] The preset recognition algorithm may be an OCR (Optical Character Recognition) algorithm, a graphic recognition algorithm, or the like.

[0096] Among them, OCR algorithms include but are not limited to paddleOCR.

[0097] S203: According to the ship license plate number and the position and posture data, a comparison image of the recorded ship corresponding to the ship license plate number and the image of the ship to be verified at the same shooting angle is obtained.

[0098] Among them, the posture data includes: position data and direction data.

[0099] The direction data includes: yaw angle and pitch angle.

[0100] Specifically, a query is performed in a preset comparison image library according to the ship license plate number to obtain a preset comparison image corresponding to the ship license plate number; according to the posture data, it is determined whether the shooting angle of the preset comparison image and the image of the ship to be verified is the same; if it is determined that the shooting angles of the preset comparison image and the image of the ship to be verified are different, then according to the image of the ship to be verified, the preset comparison image is image registered and the perspective is transformed to obtain a comparison image of the entered ship corresponding to the ship license plate number and the same shooting angle as the image of the ship to be verified.

[0101] S204: Calculate the two-dimensional feature similarity between the image of the vessel to be verified and the comparison image using a preset algorithm.

[0102] Specifically, S204 includes S2041 to S2042:

[0103] S2041: Using a preset algorithm, calculate the cosine similarity between the image of the vessel to be verified and the comparison image.

[0104] The preset algorithm may be an image similarity comparison algorithm.

[0105] Among them, the image similarity comparison algorithm includes but is not limited to Siamese Network.

[0106] It should be noted that the cosine similarity usually ranges from -1 to 1, where 1 indicates complete similarity (the vector directions are consistent), -1 indicates complete oppositeness (the vector directions are exactly opposite), and 0 indicates irrelevance (vertical relationship).

[0107] S2042: Convert the cosine similarity into two-dimensional feature similarity.

[0108] Among them, the conversion formula for converting cosine similarity into two-dimensional feature similarity is:

[0109]

[0110] Where P img is the two-dimensional feature similarity, and A is the cosine similarity.

[0111] S205: Obtaining three-dimensional point cloud data of the ship to be verified according to the image of the ship to be verified and internal and external parameters of the first device.

[0112] S206: According to the ship license plate number, query is performed in a preset three-dimensional point cloud database to obtain three-dimensional point cloud comparison data corresponding to the ship license plate number.

[0113] S207: Calculate the three-dimensional feature similarity based on the three-dimensional point cloud data and the three-dimensional point cloud comparison data.

[0114] Specifically, S207 specifically includes S2071 to S2073:

[0115] S2071: Obtain the nearest point distance and bidirectional distance of the three-dimensional point cloud data and the three-dimensional point cloud comparison data.

[0116] The three-dimensional point cloud comparison data includes internal and external parameters of the second device.

[0117] Specifically, S2071 specifically includes Sa~Sb:

[0118] Sa: According to the internal and external parameters of the first device and the internal and external parameters of the second device, the three-dimensional point cloud data and the three-dimensional point cloud comparison data are projected into the same coordinate system.

[0119] Sb: In the coordinate system, the nearest point distance and bidirectional distance of the 3D point cloud data and the 3D point cloud comparison data are calculated.

[0120] Specifically, the average distance from the 3D point cloud data to the nearest point in the 3D point cloud comparison data is calculated to obtain the nearest point distance. The nearest point distance from the 3D point cloud data to the 3D point cloud comparison data and the nearest point distance from the 3D point cloud comparison data to the 3D point cloud data are calculated, and the average or maximum value of the two nearest point distances is taken to obtain the bidirectional distance.

[0121] S2072: Convert the closest point distance and the bidirectional distance to obtain the closest point distance similarity and the bidirectional distance similarity.

[0122] Among them, the conversion formula from the closest point distance to the closest point distance similarity is:

[0123] S(d) n =exp(-α*d n )

[0124] Where S(d) n is the nearest point distance similarity, d n is the nearest point distance, exp is the exponential decay function, and α is a positive dynamic scaling factor.

[0125] It should be noted that S(d) nThe value range is (0,1); α is used to adjust the decay speed of the exponential decay function. The choice of α depends on the specific application scenario and the magnitude of the distance difference. In practical applications, this parameter can be adjusted through experiments to achieve an ideal evaluation effect.

[0126] Among them, the conversion formula from bidirectional distance to bidirectional distance similarity is:

[0127] S(d) d =exp(-α*d d )

[0128] Where S(d) d is the bidirectional distance similarity, d d is the two-way distance, exp is the exponential decay function, and α is a positive dynamic scaling factor.

[0129] It should be noted that S(d) d The value range is (0,1).

[0130] S2073: Calculate the three-dimensional feature similarity based on the nearest point distance similarity and the bidirectional distance similarity.

[0131] Among them, the calculation formula of three-dimensional feature similarity is:

[0132]

[0133] Where P voc is the three-dimensional feature similarity, S(d) n is the nearest point distance similarity, S(d) d is the two-way distance similarity.

[0134] S208: Calculate the comprehensive similarity based on the two-dimensional feature similarity and the three-dimensional feature similarity.

[0135] Among them, the calculation formula of comprehensive similarity is:

[0136] W=aP img +bP voc , 0 <a<1,0<b<1,a+b=1

[0137] Where W is the comprehensive similarity, P img is the two-dimensional feature similarity, P voc is the three-dimensional feature similarity, a and b are dynamic weight parameters.

[0138] S209: If it is determined that the comprehensive similarity does not exceed the preset similarity threshold, a recognition result is generated that the vessel to be verified is a counterfeit vessel.

[0139] The preset similarity threshold may be 90%.

[0140] It should be noted that, in practical applications, the preset similarity threshold can be adjusted through experiments to achieve an ideal evaluation effect.

[0141] From the above description, it can be seen that the present application calculates the two-dimensional feature similarity between the image of the vessel to be verified and the comparison image by using a preset algorithm; calculates the three-dimensional feature similarity based on the three-dimensional point cloud data and the three-dimensional point cloud comparison data; calculates the comprehensive similarity based on the two-dimensional feature similarity and the three-dimensional feature similarity; if it is determined that the comprehensive similarity does not exceed the preset similarity threshold, then generates an identification result that the vessel to be verified is a fake licensed vessel. The comprehensive analysis combined with the two-dimensional features and the three-dimensional features improves the accuracy and reliability of fishing vessel identification.

[0142] In one embodiment of the present application, based on the above embodiment, it further includes: if it is determined that the comprehensive similarity exceeds a preset similarity threshold, an identification result is generated that the ship to be verified is not a counterfeit ship.

[0143] In one embodiment of the present application, based on the above embodiment, it further includes: saving the comprehensive similarity and recognition results in a preset database.

[0144] In one embodiment of the present application, based on the above embodiment, the specific process of step S203 is described in detail as follows:

[0145] S2031: According to the ship license plate number, query and match are performed in a preset image multi-angle generation model library to obtain a preset image multi-angle generation model corresponding to the ship license plate number.

[0146] The preset image multi-angle generation model includes but is not limited to the NeRF model.

[0147] It should be noted that each ship license plate number corresponds to a preset image multi-angle generation model.

[0148] S2032: Input the posture data into the image multi-angle generation model to obtain a comparison image of the recorded vessel corresponding to the license plate number and the image of the vessel to be verified at the same shooting angle.

[0149] Among them, the posture data includes: position data and direction data.

[0150] The direction data includes: yaw angle and pitch angle.

[0151] It should be noted that the image multi-angle generation model uses the representation previously learned from the known perspective image set to render and project the new perspective, and generate a comparison image of the registered ship corresponding to the license plate number and the image of the ship to be verified at the same shooting angle.

[0152] From the above description, it can be seen that in this application, the posture data is input into the multi-angle image generation model to obtain a comparison image of the recorded ship corresponding to the license plate number and the image of the ship to be verified at the same shooting angle, which solves the problem of inaccurate feature matching caused by differences in shooting angles and improves the accuracy of feature matching, thereby improving the accuracy and reliability of fishing vessel identity recognition.

[0153] In one embodiment of the present application, based on the above embodiment, the specific process of step S205 is described in detail as follows:

[0154] S2051: Inputting the image of the vessel to be verified into a preset depth estimation model to generate depth information of the vessel to be verified.

[0155] The preset depth estimation model includes but is not limited to DepthAnything.

[0156] The depth information may be a depth image, a depth point cloud, or the like.

[0157] S2052: Perform three-dimensional reconstruction on the vessel to be inspected according to the depth information and internal and external parameters of the first device to obtain three-dimensional point cloud data of the vessel to be inspected.

[0158] Specifically, the depth information and internal and external parameters of the first device are input into a preset three-dimensional reconstruction algorithm to obtain three-dimensional point cloud data of the vessel to be verified.

[0159] Among them, the preset 3D reconstruction algorithm can be structured light 3D reconstruction, multi-view geometry method, etc.

[0160] From the above description, it can be seen that the present application generates depth information of the vessel to be verified by inputting the image of the vessel to be verified into a preset deep learning model, thereby reducing the need for a large amount of training data and improving the efficiency and accuracy of the model depth information estimation.

[0161] In one embodiment of the present application, based on the above embodiment, after step S209, a process of sending the comprehensive similarity and recognition result to the platform is further included, which is described in detail as follows:

[0162] S210: Output an alarm prompt, and send the comprehensive similarity and recognition results to the platform end, so that the platform end responds to the user's operation of clicking to view the results, and displays the ship identification result page to be verified, wherein the ship identification result page to be verified includes the comprehensive similarity and recognition results.

[0163] It should be noted that the platform can provide users with a user interaction interface to view and manage comprehensive similarities and recognition results, and can summarize historical comprehensive similarities and recognition results, and display the overall analysis situation through statistics and analysis of historical records.

[0164] From the above description, it can be seen that the present application can promptly notify the user of the situation of counterfeit ships, thus facilitating the management of counterfeit ships.

[0165] Figure 3 A flowchart of a method for identifying counterfeit ships provided in another embodiment of the present application. The execution subject of this embodiment may be Figure 1 The data acquisition device 101 shown in the figure may also be other computer devices, which is not particularly limited in this embodiment. Figure 3 As shown, the method includes:

[0166] S301: Acquire initial ship image data, wherein the initial ship image data includes: an initial ship image and position and posture data of a data acquisition device.

[0167] The initial vessel image data includes metadata.

[0168] The metadata includes timestamp, geographic location, shooting equipment information, shooting angle information, etc.

[0169] Among them, the posture data includes: position data and direction data.

[0170] The direction data includes: yaw angle and pitch angle.

[0171] S302: Generate a page for selecting a vessel to be verified based on the initial vessel image.

[0172] S303: In response to the user's operation on the to-be-verified vessel selection page, the location information of the to-be-verified vessel is obtained.

[0173] Specifically, in response to a selection or click operation by the user on the to-be-verified vessel selection page, the location information of the to-be-verified vessel is obtained.

[0174] S304: Acquire its own internal and external parameters to generate the first device internal and external parameters.

[0175] S305: sending the initial ship image data, the location information of the ship to be verified and the internal and external parameters of the first device to the service device, so that the service device obtains the image of the ship to be verified and the license plate number of the ship to be verified according to the initial ship image data and the location information of the ship to be verified, obtains the recorded ship corresponding to the license plate number according to the license plate number and the posture data, and a comparison image of the ship to be verified at the same shooting angle as the image of the ship to be verified by using a preset algorithm to calculate the two-dimensional feature similarity between the image of the ship to be verified and the comparison image, obtains the three-dimensional point cloud data of the ship to be verified according to the image of the ship to be verified and the internal and external parameters of the first device, and queries the preset three-dimensional point cloud database according to the license plate number to obtain the three-dimensional point cloud comparison data corresponding to the license plate number, calculates the three-dimensional feature similarity according to the three-dimensional point cloud data and the three-dimensional point cloud comparison data, and calculates the comprehensive similarity according to the two-dimensional feature similarity and the three-dimensional feature similarity. If it is determined that the comprehensive similarity does not exceed the preset similarity threshold, then a recognition result is generated that the ship to be verified is a fake license plate ship.

[0176] From the above description, it can be seen that the present application acquires initial ship image data; obtains the location information of the ship to be verified in response to the user's operation on the ship selection page to be verified; obtains the internal and external parameters of the first device; and sends the initial ship image data, the location information of the ship to be verified, and the internal and external parameters of the first device to the service device, providing a data basis for identifying counterfeit ships.

[0177] In one embodiment of the present application, based on the above embodiment, step S303 further provides another implementation method, and the specific process is described in detail as follows:

[0178] S3031: In response to the user's input operation on the to-be-verified vessel selection page, a prompt word is obtained; the prompt word is input into a preset target detection algorithm to obtain the location information of the to-be-verified vessel; or.

[0179] Among them, the prompt words may be "find the independent boat closest to the center of the picture", "find the boat with a blue fishing net hanging on the right side of the image", "find the boat with the largest pixel area in the picture", etc.

[0180] Among them, the preset target detection algorithm can be a target detection algorithm based on reasoning.

[0181] Among them, the inference-based target detection algorithm includes but is not limited to DetGPT.

[0182] S3032: In response to the user's selection operation on the to-be-verified vessel selection page, the location information of the to-be-verified vessel is obtained.

[0183] Specifically, in response to the user clicking a certain point or selecting a certain area on the to-be-verified vessel selection page, the location information of the to-be-verified vessel is obtained.

[0184] It should be noted that it is necessary to click any point on the vessel to be verified or select the minimum circumscribed rectangular frame of the vessel to be verified.

[0185] It should be noted that the user can choose to enter a prompt word on the ship selection page to determine the location information of the ship to be verified. When the light is dim or the surrounding environment is too complex and the target detection algorithm cannot accurately understand the prompt word, the user can choose to click any point on the ship to be verified or select the minimum circumscribed rectangular frame of the ship to be verified on the ship selection page to determine the location information of the ship to be verified.

[0186] From the above description, it can be seen that the present application provides a data basis for obtaining the image of the vessel to be verified through subsequent image segmentation, solving the problem of being unable to determine the vessel to be verified when there are a large number of vessels on the sea and / or the shooting angle causes the ships to obstruct each other.

[0187] Figure 4 This is a schematic diagram of the structure of the device for identifying counterfeit ships provided in the embodiment of the present application. Figure 4 As shown, the device 40 for identifying counterfeit ships includes: a receiving module 401 , an acquiring module 402 , a calculating module 403 , a query module 404 and a generating module 405 .

[0188] The receiving module 401 is used to receive the initial ship image data, the position information of the ship to be verified and the internal and external parameters of the first device sent by the data acquisition device, wherein the initial ship image data includes: the initial ship image and the position data of the data acquisition device, wherein the position information of the ship to be verified is obtained by the data acquisition device in response to the user's operation on the ship selection page to be verified, wherein the ship selection page to be verified is generated by the data acquisition device according to the initial ship image, and wherein the internal and external parameters of the first device are generated by the data acquisition device acquiring its own internal and external parameters;

[0189] The acquisition module 402 is used to obtain the image of the vessel to be verified and the license plate number of the vessel to be verified according to the initial vessel image data and the position information of the vessel to be verified;

[0190] The acquisition module 402 is further configured to obtain, based on the ship license number and the position data, a comparison image of the recorded ship corresponding to the ship license number and the image of the ship to be verified at the same shooting angle;

[0191] A calculation module 403 is used to calculate the two-dimensional feature similarity between the image of the ship to be verified and the comparison image using a preset algorithm;

[0192] The acquisition module 402 is further used to obtain the three-dimensional point cloud data of the vessel to be verified according to the image of the vessel to be verified and the internal and external parameters of the first device;

[0193] A query module 404 is used to query a preset three-dimensional point cloud database according to the ship license number to obtain three-dimensional point cloud comparison data corresponding to the ship license number;

[0194] The calculation module 403 is further used to calculate the three-dimensional feature similarity based on the three-dimensional point cloud data and the three-dimensional point cloud comparison data;

[0195] The calculation module 403 is further used to calculate the comprehensive similarity according to the two-dimensional feature similarity and the three-dimensional feature similarity;

[0196] The generating module 405 is used to generate a recognition result that the vessel to be verified is a counterfeit vessel if it is determined that the comprehensive similarity does not exceed a preset similarity threshold.

[0197] In a possible design, the acquisition module 402 is specifically used to: perform query matching in a preset image multi-angle generation model library according to the license plate number to obtain a preset image multi-angle generation model corresponding to the license plate number; input the posture data into the image multi-angle generation model to obtain a comparison image of the entered ship corresponding to the license plate number and the same shooting angle as the image of the ship to be verified.

[0198] In a possible design, the calculation module 403 is specifically used to: use a preset algorithm to calculate the cosine similarity between the image of the ship to be verified and the comparison image; and convert the cosine similarity into a two-dimensional feature similarity.

[0199] In a possible design, the query module 404 is specifically used to: input the image of the ship to be verified into a preset deep learning model to generate depth information of the ship to be verified; and perform three-dimensional reconstruction of the ship to be verified based on the depth information and internal and external parameters of the first device to obtain three-dimensional point cloud data of the ship to be verified.

[0200] In one possible design, the calculation module 403 is specifically used to: obtain the nearest point distance and the two-way distance of the three-dimensional point cloud data and the three-dimensional point cloud comparison data; transform the nearest point distance and the two-way distance to obtain the nearest point distance similarity and the two-way distance similarity; and calculate the three-dimensional feature similarity based on the nearest point distance similarity and the two-way distance similarity.

[0201] In a possible design, the three-dimensional point cloud comparison data includes internal and external parameters of the second device; accordingly, the calculation module 403, when "obtaining the nearest point distance and two-way distance between the three-dimensional point cloud data and the three-dimensional point cloud comparison data", is specifically used to: project the three-dimensional point cloud data and the three-dimensional point cloud comparison data to the same coordinate system according to the internal and external parameters of the first device and the internal and external parameters of the second device; and calculate the nearest point distance and two-way distance between the three-dimensional point cloud data and the three-dimensional point cloud comparison data in the coordinate system.

[0202] In a possible design, the acquisition module 402 is specifically used to: perform image segmentation on the initial ship image according to the position information of the ship to be verified to obtain the image of the ship to be verified; and use a preset recognition algorithm to identify the image of the ship to be verified to obtain the license plate number of the ship to be verified.

[0203] In a possible design, the counterfeit vessel identification device 40 further includes:

[0204] The sending module 406 is used to output an alarm prompt and send the comprehensive similarity and the recognition result to the platform end, so that the platform end displays the identification result page of the ship to be verified in response to the user's operation of clicking to view the result, wherein the identification result page of the ship to be verified includes the comprehensive similarity and the recognition result.

[0205] In a possible design, the conversion formula for converting the cosine similarity into the two-dimensional feature similarity is:

[0206]

[0207] Where P img is the two-dimensional feature similarity, and A is the cosine similarity.

[0208] In a possible design, the closest point distance and the bidirectional distance are transformed to obtain the closest point distance similarity and the bidirectional distance similarity, and the transformation formula is:

[0209] S(d) n =exp(-α*d n )

[0210] Where S(d) n is the nearest point distance similarity, d n is the nearest point distance, exp is the exponential decay function, and α is the positive dynamic scaling factor;

[0211] S(d) d =exp(-α*d d )

[0212] Where S(d) d is the bidirectional distance similarity, d d is the two-way distance, exp is the exponential decay function, and α is a positive dynamic scaling factor.

[0213] In a possible design, the three-dimensional feature similarity is calculated based on the closest point distance similarity and the bidirectional distance similarity, and the calculation formula is:

[0214]

[0215] Where P voc is the three-dimensional feature similarity, S(d) n is the nearest point distance similarity, S(d) d is the two-way distance similarity.

[0216] In a possible design, the comprehensive similarity is calculated based on the two-dimensional feature similarity and the three-dimensional feature similarity, and the calculation formula is:

[0217] W=aP img +bP voc , 0 <a<1,0<b<1,a+b=1

[0218] Where W is the comprehensive similarity, P img is the two-dimensional feature similarity, P voc is the three-dimensional feature similarity, a and b are dynamic weight parameters.

[0219] The device provided in this embodiment can be used to execute the technical solution of the above method embodiment. Its implementation principle and technical effect are similar, and this embodiment will not be repeated here.

[0220] Figure 5 This is a schematic diagram of the structure of another device for identifying counterfeit ships provided in an embodiment of the present application. Figure 5 As shown, the device 50 for identifying counterfeit ships includes: a collection module 501 , a generation module 502 , an acquisition module 503 and a sending module 504 .

[0221] The acquisition module 501 is used to acquire initial ship image data, wherein the initial ship image data includes: an initial ship image and position data of the data acquisition device;

[0222] A generating module 502, for generating a to-be-verified vessel selection page according to the initial vessel image;

[0223] An acquisition module 503 is used to obtain the location information of the vessel to be verified in response to the user's operation on the vessel selection page to be verified;

[0224] The acquisition module 503 is further used to acquire its own internal and external parameters to generate the internal and external parameters of the first device;

[0225] The sending module 504 is used to send the initial ship image data, the location information of the ship to be verified, and the internal and external parameters of the first device to the service device, so that the service device obtains the image of the ship to be verified and the license plate number of the ship to be verified according to the initial ship image data and the location information of the ship to be verified, obtains the recorded ship corresponding to the license plate number and the comparison image of the same shooting angle as the image of the ship to be verified according to the license plate number and the posture data, and calculates the two-dimensional feature similarity between the image of the ship to be verified and the comparison image using a preset algorithm. According to the image of the vessel to be verified and the internal and external parameters of the first device, the three-dimensional point cloud data of the vessel to be verified is obtained; according to the license plate number, a query is performed in a preset three-dimensional point cloud database to obtain the three-dimensional point cloud comparison data corresponding to the license plate number; according to the three-dimensional point cloud data and the three-dimensional point cloud comparison data, the three-dimensional feature similarity is calculated; according to the two-dimensional feature similarity and the three-dimensional feature similarity, the comprehensive similarity is calculated; if it is determined that the comprehensive similarity does not exceed a preset similarity threshold, a recognition result that the vessel to be verified is a fake-license vessel is generated.

[0226] In a possible design, the acquisition module 503 is specifically used to: obtain a prompt word in response to the user's input operation on the ship selection page to be verified; input the prompt word into a preset target detection algorithm to obtain the location information of the ship to be verified; or, obtain the location information of the ship to be verified in response to the user's selection operation on the ship selection page to be verified.

[0227] The device provided in this embodiment can be used to execute the technical solution of the above method embodiment. Its implementation principle and technical effect are similar, and this embodiment will not be repeated here.

[0228] Figure 6 Schematic diagram of the hardware structure of the service device provided in the embodiment of the present application. Figure 6 As shown, the service device 60 of this embodiment includes: at least one processor 601 and a memory 602; the memory stores computer-executable instructions; at least one processor executes the computer-executable instructions stored in the memory, so that at least one processor executes the above-mentioned counterfeit ship identification method;

[0229] Optionally, the memory 602 may be independent or integrated with the processor 601 .

[0230] When the memory 602 is independently provided, the service device further includes a bus 603 for connecting the memory 602 and the processor 601 .

[0231] Figure 7 Schematic diagram of the hardware structure of the data acquisition device provided in the embodiment of the present application. Figure 7 As shown, the data acquisition device 70 of this embodiment includes: at least one processor 701 and a memory 702; the memory stores computer-executable instructions; at least one processor executes the computer-executable instructions stored in the memory, so that at least one processor executes the above-mentioned counterfeit ship identification method;

[0232] Optionally, the memory 702 may be independent or integrated with the processor 701 .

[0233] When the memory 702 is independently provided, the data acquisition device further includes a bus 703 for connecting the memory 702 and the processor 701 .

[0234] The embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions. When a processor executes the computer-executable instructions, the method for identifying counterfeit ships as described above is implemented.

[0235] The embodiment of the present application also provides a computer program product, including a computer program, and when the computer program is executed by a processor, the method for identifying counterfeit ships as described above is implemented.

[0236] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the present application.

[0237] It should be further noted that, although the various steps in the flowchart are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowchart may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0238] It should be understood that the above-mentioned device embodiments are only illustrative, and the device of the present application can also be implemented in other ways. For example, the division of units / modules in the above-mentioned embodiments is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units, modules or components can be combined, or can be integrated into another system, or some features can be ignored or not executed.

[0239] In addition, unless otherwise specified, each functional unit / module in each embodiment of the present application may be integrated into one unit / module, each unit / module may exist physically separately, or two or more units / modules may be integrated together. The above-mentioned integrated unit / module may be implemented in the form of hardware or in the form of a software program module.

[0240] If the integrated unit / module is implemented in the form of hardware, the hardware may be a digital circuit, an analog circuit, etc. The physical implementation of the hardware structure includes but is not limited to transistors, memristors, etc. If not otherwise specified, the processor may be any appropriate hardware processor, such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an FPGA (Field-Programmable Gate Array), a DSP (Digital Signal Processor), and an ASIC (Application-Specific Integrated Circuit). If not otherwise specified, the storage unit may be any appropriate magnetic storage medium or magneto-optical storage medium, such as RRAM (Resistive Random Access Memory), DRAM (Dynamic Random Access Memory), SRAM (Static Random-Access Memory), EDRAM (Enhanced Dynamic Random Access Memory), HBM (High-Bandwidth Memory), HMC (Hybrid Memory Cube), etc.

[0241] If the integrated unit / module is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of each embodiment method of the present application. The aforementioned memory includes: U disk, ROM (Read-Only Memory), RAM (Random Access Memory), mobile hard disk, magnetic disk or optical disk, etc., various media that can store program codes.

[0242] In the above embodiments, the description of each embodiment has its own emphasis. For the part not described in detail in a certain embodiment, please refer to the relevant description of other embodiments. The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0243] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0244] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A method for identifying counterfeit ships, characterized in that: Applied to service equipment, including: Receive initial ship image data, position information of the ship to be verified, and internal and external parameters of the first device sent by a data acquisition device, wherein the initial ship image data includes: an initial ship image and position data of the data acquisition device, wherein the position information of the ship to be verified is obtained by the data acquisition device in response to a user's operation on a ship selection page to be verified, wherein the ship selection page to be verified is generated by the data acquisition device according to the initial ship image, and wherein the internal and external parameters of the first device are generated by the data acquisition device acquiring its own internal and external parameters; Obtaining the image of the vessel to be verified and the license plate number of the vessel to be verified according to the initial vessel image data and the position information of the vessel to be verified; According to the ship license plate number and the position data, a comparison image of the recorded ship corresponding to the ship license plate number and the image of the ship to be verified at the same shooting angle is obtained; Calculating the similarity of the two-dimensional features between the image of the vessel to be verified and the comparison image using a preset algorithm; Obtaining three-dimensional point cloud data of the vessel to be verified according to the image of the vessel to be verified and internal and external parameters of the first device; According to the ship license plate number, query is performed in a preset three-dimensional point cloud database to obtain three-dimensional point cloud comparison data corresponding to the ship license plate number; Calculating three-dimensional feature similarity based on the three-dimensional point cloud data and the three-dimensional point cloud comparison data; Calculating a comprehensive similarity based on the two-dimensional feature similarity and the three-dimensional feature similarity; If it is determined that the comprehensive similarity does not exceed the preset similarity threshold, a recognition result that the vessel to be verified is a counterfeit vessel is generated.

2. The method according to claim 1, characterized in that The step of obtaining, according to the ship license plate number and the position data, a comparison image of the recorded ship corresponding to the ship license plate number and the image of the ship to be verified at the same shooting angle as that of the ship to be verified includes: According to the ship license plate number, query and match are performed in a preset image multi-angle generation model library to obtain a preset image multi-angle generation model corresponding to the ship license plate number; The posture data is input into the image multi-angle generation model to obtain a comparison image of the recorded vessel corresponding to the license plate number and the image of the vessel to be verified at the same shooting angle.

3. The method according to claim 1, characterized in that The using of a preset algorithm to calculate the two-dimensional feature similarity between the image of the vessel to be verified and the comparison image includes: Using a preset algorithm, calculating the cosine similarity between the image of the vessel to be verified and the comparison image; The cosine similarity is converted into a two-dimensional feature similarity.

4. The method according to claim 1, characterized in that: The step of obtaining the three-dimensional point cloud data of the vessel to be verified according to the image of the vessel to be verified and the internal and external parameters of the first device includes: Inputting the image of the vessel to be verified into a preset deep learning model to generate depth information of the vessel to be verified; The vessel to be verified is three-dimensionally reconstructed according to the depth information and internal and external parameters of the first device to obtain three-dimensional point cloud data of the vessel to be verified.

5. The method according to claim 1, characterized in that The calculating the three-dimensional feature similarity according to the three-dimensional point cloud data and the three-dimensional point cloud comparison data includes: Obtaining the closest point distance and the bidirectional distance between the three-dimensional point cloud data and the three-dimensional point cloud comparison data; Convert the closest point distance and the bidirectional distance to obtain the closest point distance similarity and the bidirectional distance similarity; The three-dimensional feature similarity is calculated based on the closest point distance similarity and the bidirectional distance similarity.

6. The method according to claim 5, characterized in that The three-dimensional point cloud comparison data includes internal and external parameters of the second device; Accordingly, the obtaining of the closest point distance and the bidirectional distance between the three-dimensional point cloud data and the three-dimensional point cloud comparison data includes: Projecting the three-dimensional point cloud data and the three-dimensional point cloud comparison data into the same coordinate system according to the internal and external parameters of the first device and the internal and external parameters of the second device; In the coordinate system, the closest point distance and the bidirectional distance between the three-dimensional point cloud data and the three-dimensional point cloud comparison data are calculated.

7. The method according to claim 1, characterized in that The obtaining the image of the vessel to be verified and the license plate number of the vessel to be verified according to the initial vessel image data and the position information of the vessel to be verified comprises: According to the position information of the vessel to be verified, the initial vessel image is segmented to obtain the vessel image to be verified; The image of the vessel to be verified is identified using a preset recognition algorithm to obtain the license plate number of the vessel to be verified.

8. The method according to claim 1, characterized in that If it is determined that the comprehensive similarity does not exceed the preset similarity threshold, after generating the identification result that the vessel to be verified is a counterfeit vessel, the method further includes: Output an alarm prompt, and send the comprehensive similarity and the recognition result to the platform end, so that the platform end displays the identification result page of the ship to be verified in response to the user's operation of clicking to view the result, wherein the identification result page of the ship to be verified includes the comprehensive similarity and the recognition result.

9. The method according to claim 3, characterized in that: The conversion formula for converting the cosine similarity into two-dimensional feature similarity is: Where P img is the two-dimensional feature similarity, and A is the cosine similarity.

10. The method according to claim 5, characterized in that The closest point distance and the two-way distance are transformed to obtain the closest point distance similarity and the two-way distance similarity, and the transformation formula is: S(d) n =exp(-α*d n ) Where S(d) n is the nearest point distance similarity, d n is the nearest point distance, exp is the exponential decay function, and α is the positive dynamic scaling factor; S(d) d =exp(-α*d d ) Where S(d) d is the bidirectional distance similarity, d d is the two-way distance, exp is the exponential decay function, and α is a positive dynamic scaling factor.

11. The method according to claim 5, characterized in that The three-dimensional feature similarity is calculated based on the closest point distance similarity and the bidirectional distance similarity, and the calculation formula is: Where P voc is the three-dimensional feature similarity, S(d) n is the nearest point distance similarity, S(d) d is the two-way distance similarity.

12. The method according to any one of claims 1 to 11, characterized in that: The calculation formula for calculating the comprehensive similarity based on the two-dimensional feature similarity and the three-dimensional feature similarity is: W=aP img +bP voc ,0<a<1,0<b<1,a+b=1 Where W is the comprehensive similarity, P img is the two-dimensional feature similarity, P voc is the three-dimensional feature similarity, a and b are dynamic weight parameters.

13. A method for identifying counterfeit ships, characterized in that: Applied to data acquisition equipment, including: Collecting initial ship image data, wherein the initial ship image data includes: an initial ship image and position and posture data of the data acquisition device; Generating a page for selecting a vessel to be verified according to the initial vessel image; In response to the user's operation on the to-be-verified vessel selection page, obtaining the location information of the to-be-verified vessel; Acquire its own internal and external parameters to generate the internal and external parameters of the first device; The initial ship image data, the location information of the ship to be verified, and the internal and external parameters of the first device are sent to the service device, so that the service device obtains the image of the ship to be verified and the license plate number of the ship to be verified according to the initial ship image data and the location information of the ship to be verified, obtains the recorded ship corresponding to the license plate number and the comparison image with the same shooting angle as the image of the ship to be verified according to the license plate number and the posture data, calculates the two-dimensional feature similarity between the image of the ship to be verified and the comparison image by using a preset algorithm, obtains the three-dimensional point cloud data of the ship to be verified according to the image of the ship to be verified and the internal and external parameters of the first device, performs a query in a preset three-dimensional point cloud database according to the license plate number to obtain the three-dimensional point cloud comparison data corresponding to the license plate number, calculates the three-dimensional feature similarity according to the three-dimensional point cloud data and the three-dimensional point cloud comparison data, calculates the comprehensive similarity according to the two-dimensional feature similarity and the three-dimensional feature similarity, and generates a recognition result that the ship to be verified is a fake license plate ship if it is determined that the comprehensive similarity does not exceed a preset similarity threshold.

14. The method according to claim 13, characterized in that The obtaining, in response to the user's operation on the to-be-verified vessel selection page, the location information of the to-be-verified vessel includes: In response to the user's input operation on the to-be-verified vessel selection page, a prompt word is obtained; the prompt word is input into a preset target detection algorithm to obtain the location information of the to-be-verified vessel; or, In response to the user's selection operation on the to-be-verified vessel selection page, the location information of the to-be-verified vessel is obtained.

15. A device for identifying counterfeit ships, characterized in that: Applied to service equipment, including: A receiving module, configured to receive initial ship image data, position information of a ship to be verified, and internal and external parameters of a first device sent by a data acquisition device, wherein the initial ship image data includes: an initial ship image and position data of the data acquisition device, wherein the position information of the ship to be verified is obtained by the data acquisition device in response to an operation of a user on a ship selection page to be verified, wherein the ship selection page to be verified is generated by the data acquisition device according to the initial ship image, and wherein the internal and external parameters of the first device are generated by the data acquisition device acquiring its own internal and external parameters; An acquisition module, used for obtaining the image of the vessel to be verified and the license plate number of the vessel to be verified according to the initial vessel image data and the position information of the vessel to be verified; The acquisition module is further used to obtain, based on the ship license number and the position data, a comparison image of the recorded ship corresponding to the ship license number and the image of the ship to be verified at the same shooting angle; A calculation module, used for calculating the similarity of the two-dimensional features between the image of the vessel to be verified and the comparison image by using a preset algorithm; The acquisition module is further used to obtain the three-dimensional point cloud data of the vessel to be verified according to the image of the vessel to be verified and the internal and external parameters of the first device; A query module, used to query a preset three-dimensional point cloud database according to the ship license plate number to obtain three-dimensional point cloud comparison data corresponding to the ship license plate number; The calculation module is further used to calculate the three-dimensional feature similarity based on the three-dimensional point cloud data and the three-dimensional point cloud comparison data; The calculation module is further used to calculate the comprehensive similarity based on the two-dimensional feature similarity and the three-dimensional feature similarity; The generating module is used for generating a recognition result that the vessel to be verified is a counterfeit vessel if it is determined that the comprehensive similarity does not exceed a preset similarity threshold.

16. A device for identifying counterfeit ships, characterized in that: Applied to data acquisition equipment, including: An acquisition module, used for acquiring initial ship image data, wherein the initial ship image data comprises: an initial ship image and position and posture data of the data acquisition device; A generating module, used for generating a to-be-verified vessel selection page according to the initial vessel image; An acquisition module, configured to obtain the location information of the vessel to be verified in response to the user's operation on the vessel selection page to be verified; The acquisition module is further used to acquire its own internal and external parameters to generate the internal and external parameters of the first device; a sending module, configured to send the initial ship image data, the location information of the ship to be verified, and the internal and external parameters of the first device to the service device, so that the service device obtains the image of the ship to be verified and the license plate number of the ship to be verified according to the initial ship image data and the location information of the ship to be verified, obtains the comparison image of the recorded ship corresponding to the license plate number and the same shooting angle as the image of the ship to be verified according to the license plate number and the posture data, calculates the two-dimensional feature similarity between the image of the ship to be verified and the comparison image by using a preset algorithm, obtains the three-dimensional point cloud data of the ship to be verified according to the image of the ship to be verified and the internal and external parameters of the first device, performs a query in a preset three-dimensional point cloud database according to the license plate number to obtain the three-dimensional point cloud comparison data corresponding to the license plate number, calculates the three-dimensional feature similarity according to the three-dimensional point cloud data and the three-dimensional point cloud comparison data, calculates the comprehensive similarity according to the two-dimensional feature similarity and the three-dimensional feature similarity, and generates a recognition result that the ship to be verified is a fake license plate ship if it is determined that the comprehensive similarity does not exceed a preset similarity threshold.

17. A service device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method for identifying counterfeit ships according to any one of claims 1 to 12.

18. A data acquisition device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method for identifying counterfeit ships as claimed in claim 13 or 14.

19. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method for identifying counterfeit ships as described in any one of claims 1 to 12, or the method for identifying counterfeit ships as described in claim 13 or 14.

20. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for identifying counterfeit ships according to any one of claims 1 to 12, or the method for identifying counterfeit ships according to claim 13 or 14 is implemented.

Images