A method and system for identifying vehicle documents

Abstract

The present invention provides a method and system for identifying vehicle-carrying documents, wherein the identification system for vehicle-carrying documents comprises a fuselage, a plurality of detection windows, a plurality of cameras, a processor and a storage device, wherein the detection window is arranged on the top of the fuselage; the camera is arranged inside the fuselage, the camera corresponds to the detection window one by one, and the camera collects the image of the vehicle-carrying document on the detection window; the processor extracts the vehicle-carrying data from the image of the vehicle-carrying document and performs identification and comparison; the storage device is used to store the identification data and the comparison result. The identification system for vehicle-carrying documents of the present invention can detect all the contents of a plurality of vehicle-carrying documents at the same time, thereby improving the detection efficiency and having high detection accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle document recognition, and particularly to a method and a system for recognizing vehicle documents. Background Art

[0002] Vehicle documents are important identity and product certificates of automobiles and the only proof of factory. Vehicle documents include four kinds of files: vehicle qualification certificates, COC conformity certificates, energy consumption labels, and environmental protection information lists. Currently, manual sampling inspection is used to judge whether the content of the files is correct. Due to production capacity requirements, only a very short detection time is allowed for personnel, but there is a lot of information in the four files and it is completely impossible to detect within the specified time, so only sampling inspection can be adopted. Therefore, there are many disadvantages, such as the inability to achieve full content inspection, easy occurrence of detection errors, and low detection efficiency. Summary of the Invention

[0003] The purpose of the present invention is to overcome the deficiencies in the prior art and provide a method and a system for recognizing vehicle documents.

[0004] An embodiment of the present invention provides a method for recognizing vehicle documents, including the following steps:

[0005] Receiving images of multiple vehicle documents; wherein, the images of the vehicle documents are obtained by collecting the vehicle documents on multiple detection windows through multiple cameras respectively;

[0006] Respectively detecting the outer frames of the images of each vehicle document, and correspondingly extracting the content within the outer frames as vehicle materials;

[0007] Performing image preprocessing on the vehicle materials;

[0008] Using deep learning CNN to recognize the content of the complex background area of the vehicle materials; using OCR to recognize the content of the single background area of the vehicle materials;

[0009] Collecting recognition data and making comparisons, and sending the obtained comparison results to a display device.

[0010] Compared with the prior art, the method for recognizing vehicle documents of the present invention can simultaneously detect all the content in multiple vehicle documents, improve the detection efficiency, and have high detection accuracy.

[0011] Further, after extracting the content within the outer frames as vehicle materials, the title position of the vehicle materials is also detected, and the position of the vehicle materials is corrected and positioned according to the position of the title to improve the recognition accuracy.

[0012] Further, deep learning CNN is used to recognize the content of the complex background area of the vehicle documents; OCR is used to recognize the content of the single-background area of the vehicle documents, which specifically includes the following steps: selecting the content of the complex background area from the vehicle documents through a preset number of first recognition areas, and using deep learning CNN to recognize the content selected by the first recognition areas; using OCR to recognize the content in the vehicle documents that is not selected by the first recognition areas;

[0013] The position and selection range of the first recognition area correspond to the content of the title of the vehicle documents, and the position of the first recognition area is relatively fixed with respect to the title position. The area for recognizing the vehicle documents is divided to improve the recognition accuracy.

[0014] Further, the recognition data is collected and compared, and the result obtained from the comparison is sent to the display device, which specifically includes: finding the corresponding vehicle information in the mes system according to the recognition data and performing a comparison, and sending the result of the comparison to the display device. By finding the corresponding vehicle information in the mes system and performing a comparison, the detection quality is improved.

[0015] Further, after the outer frames of the images of the vehicle certificates are respectively detected, it is also determined whether the vehicle certificates correspond to the detection windows according to the sizes of the outer frames of the images of the vehicle certificates detected. This prevents the vehicle certificates from being placed on the non-corresponding detection windows.

[0016] An embodiment of the invention also provides a recognition system for vehicle certificates, including:

[0017] A fuselage;

[0018] A plurality of detection windows, which are arranged on the top of the fuselage;

[0019] A plurality of cameras, which are arranged inside the fuselage, and the cameras correspond to the detection windows one by one. The cameras collect the images of the vehicle certificates on the detection windows;

[0020] A memory, which stores a computer program;

[0021] A processor, when the processor executes the computer program, implements the steps of the recognition method for vehicle certificates. The processor sends the collected recognition data and the result of the comparison to the memory for storage;

[0022] A display screen, which is detachably connected to the fuselage, and the display screen receives and displays the recognition data and the result of the comparison of the processor.

[0023] Compared with the prior art, the on-vehicle document recognition system of the present invention can detect all the contents of multiple on-vehicle documents simultaneously, improving the detection efficiency and having high detection accuracy.

[0024] Further, it further includes a voice playback device, which is fixedly connected to the body. The voice playback device receives and plays the comparison result of the processor. The comparison result is notified to the user for viewing through voice broadcast.

[0025] Further, several support rods are provided at the top of the body, and the display screen is detachably connected to the support rods. The support rods are used to support the display screen.

[0026] Further, a sliding rod group is provided inside the body, and the multiple cameras are respectively movably connected to the sliding rods, and the multiple cameras can move along the sliding rods. This facilitates the user to adjust the position of the cameras.

[0027] Further, the detection window is provided with a transparent support plate for placing the on-vehicle document, and the lens of the camera faces the transparent support plate. This ensures that the camera can accurately collect the image of the on-vehicle document.

[0028] In order to understand the present invention more clearly, the specific embodiments of the present invention will be described below in conjunction with the accompanying drawings. Description of the Drawings

[0029] Figure 1 It is a flowchart of the on-vehicle document recognition method according to an embodiment of the present invention.

[0030] Figure 2 It is a flowchart of step S2 of the on-vehicle document recognition method according to an embodiment of the present invention.

[0031] Figure 3 It is a flowchart of step S3 of the on-vehicle document recognition method according to an embodiment of the present invention.

[0032] Figure 4 It is a schematic structural diagram of the on-vehicle document recognition system from the first angle according to an embodiment of the present invention.

[0033] Figure 5 It is a schematic structural diagram of the on-vehicle document recognition system from the second angle according to an embodiment of the present invention.

[0034] Figure 6 It is a schematic structural diagram of the on-vehicle document recognition system from the third angle according to an embodiment of the present invention. Specific Embodiments

[0035] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0036] Please refer to Figure 1 , which is a flowchart of a method for identifying vehicle-related certificates in an embodiment of the present invention. This identification method can simultaneously detect all the contents in multiple vehicle-related certificates, and it includes the following steps:

[0037] S1: Receive images of multiple vehicle-related certificates; among them, the images of the vehicle-related certificates are obtained by collecting the vehicle-related certificates on multiple detection windows through multiple cameras respectively;

[0038] S2: Detect the outer frames of the images of each of the vehicle-related certificates respectively, and correspondingly extract the contents within the outer frames as vehicle-related materials;

[0039] S3: Perform image preprocessing on the vehicle-related materials;

[0040] S4: Use deep learning CNN (Convolutional Neural Networks) to identify the contents of the complex background areas of the vehicle-related materials; use OCR (optical character recognition) to identify the contents of the single-background areas of the vehicle-related materials;

[0041] S5: Collect identification data and perform comparison, and send the comparison results to the display device.

[0042] Among them, in this embodiment, the number of the vehicle-related certificates is 4, including a vehicle certificate of conformity, a COC certificate of conformity, an energy consumption label, and an environmental protection information list. The number of the detection windows is 4, including a first detection window, a second detection window, a third detection window, and a fourth detection window. Among them, the first detection window is used to place the vehicle certificate of conformity, the second detection window is used to place the COC certificate of conformity, the third detection window is used to place the energy consumption label, and the fourth detection window is used to place the environmental protection information list. Preferably, the first detection window, the second detection window, the third detection window, and the fourth detection window are arranged horizontally in sequence on one side of the top of the fuselage, which is convenient for operators, vehicle owners or third-party personnel to place the vehicle-related certificates on the detection windows.

[0043] In other embodiments, the positional order of the first detection window, the second detection window, the third detection window, and the fourth detection window is not limited, and those skilled in the art can freely change the positional order of the first detection window, the second detection window, the third detection window, and the fourth detection window.

[0044] In other feasible embodiments, the number of the vehicle documents and the detection windows can be set according to the detection requirements, and is not limited to these 4.

[0045] There are 4 cameras, including a first camera corresponding to the first detection window, a second camera corresponding to the second detection window, a third camera corresponding to the third detection window, and a fourth camera corresponding to the fourth detection window. The specific process of receiving images of multiple vehicle documents is as follows: the first camera, the second camera, the third camera, and the fourth camera capture images of the vehicle documents on the first detection window, the second detection window, the third detection window, and the fourth detection window.

[0046] Please refer to Figure 2 , the step S2 specifically includes:

[0047] S201: Detect the outer frames of the images of each of the vehicle documents respectively, and determine whether the vehicle document corresponds to the detection window according to the size of the outer frame of the detected image of the vehicle document; if the vehicle document corresponds to the detection window, execute step S202, otherwise notify the user that the placement of the vehicle document is abnormal.

[0048] Among them, since the camera captures the image of the vehicle document and will simultaneously capture the environmental content outside the vehicle document, therefore, by using the edges of each of the vehicle documents as the outer frames of the images of each of the vehicle documents for detection, it is determined whether the vehicle document corresponds to the detection window. Taking the second detection window as an example, the second detection window is used to place the COC conformity certificate, that is, the COC conformity certificate corresponds to the second detection window. If other vehicle documents are placed in the second detection window, such as the energy consumption label, it can be detected that the size of the outer frame of the image of the energy consumption label is different from the outer frame of the COC conformity certificate, and the result that the placement of the vehicle document is abnormal is obtained.

[0049] S202: Correspondingly extract the content within its outer frame as vehicle information.

[0050] Since the environmental content outside the vehicle document captured by the camera will increase the recognition difficulty, only the content within its outer frame is extracted, so as to obtain the image data containing only the vehicle document as the vehicle information.

[0051] S203: Detect the title position of the vehicle-related materials, and correct and position the vehicle-related materials according to the position of the title.

[0052] The specific correction and positioning are as follows: If the vehicle-related materials are the front images of the vehicle-related certificates that are flipped or tilted, confirm the direction of the image by detecting the position of the title, and use the direction of the image as the reference direction for recognition, so as to improve the accuracy of recognition.

[0053] In this embodiment, step S2 can detect whether the vehicle-related certificates are placed correctly, whether the images of the vehicle-related certificates can be successfully collected and the vehicle-related materials can be extracted. And correct and position the vehicle-related materials.

[0054] Please refer to Figure 3 , step S3 specifically includes:

[0055] S301: Perform image enhancement on the vehicle-related materials; the image enhancement is: selectively highlight the text content in the vehicle-related materials and suppress the background features in the vehicle-related materials.

[0056] S302: Perform grayscale processing on the vehicle-related materials; the grayscale processing is: set the RGB values of the vehicle-related materials to R = G = B.

[0057] S303: Perform binarization processing on the vehicle-related materials; select the grayscale values of the points on the vehicle-related materials to be 0 or 255, so that the vehicle-related materials present an obvious black-and-white effect, make the vehicle-related materials simpler, and reduce the data volume of the vehicle-related materials.

[0058] Step S4 specifically includes: Select the content of the background complex area from the vehicle-related materials through a preset number of first recognition areas, and use deep learning CNN to recognize the content selected by the first recognition areas; use OCR to recognize the content in the vehicle-related materials that is not selected by the first recognition areas;

[0059] The complex background area refers to the part of the accompanying document where the text density or graphic density exceeds the preset density value, such as the background pattern in the accompanying document, the text close to or overlapping with the background pattern, and the text close to the edge of the accompanying document. The text close to the edge of the accompanying document will be difficult to identify because the edge part is irregularly lifted when the image of the accompanying document is captured. The single background area refers to the part of the accompanying document where the text density or graphic density is lower than the preset density value, that is, all other contents in the accompanying document except the complex background area, including the text far away from the background pattern and the text located at the center of the accompanying document. The preset density value is initially set to a default value, and the user can change the preset density value according to actual usage.

[0060] The position and selection range of the first identification area are preset and correspond to the content of the title of the vehicle-attached information. That is, since the types of the vehicle-attached information are different, their titles are also different, so the preset position and selection range of the first identification area correspond to each of the vehicle-attached information. Preferably, the position of the first identification area is relatively fixed to the position of the title. The vehicle-attached information is divided into identification areas to improve the accuracy of identification. Among them, the position and selection range of the first identification area correspond to the content of the title of the vehicle certificate, COC conformity certificate, energy consumption label and environmental protection information list.

[0061] The deep learning CNN is a convolutional neural network with the ability to represent learning, and the deep learning CNN is also equipped with a database, which stores a large amount of data on text and patterns, text angles, and text stretching and deformation. The deep learning CNN uses the data in the database to identify the content selected in the first recognition area.

[0062] The OCR is optical character recognition, which uses a character recognition method to translate the text shapes in the vehicle-mounted data into computer text.

[0063] By dividing different on-board data into different recognition areas, the recognition accuracy is improved.

[0064] In one embodiment, step S5 specifically includes: searching for corresponding vehicle information in the MES system (manufacturing execution system) according to the recognition data, comparing them, and sending the comparison result to a display device. Preferably, the comparison result can also be sent to a voice playback device. The MES system is a production information management system for the execution layer of a manufacturing enterprise workshop, which stores the vehicle information. The vehicle information includes all or part of the information in the vehicle certificate of conformity, COC conformity certificate, energy consumption label, and environmental protection information list.

[0065] In this embodiment, taking the COC conformity certificate as an example, in this embodiment, according to one or more of the vehicle conformity certificate number, vehicle identification number, engine number, and vehicle model in the vehicle documents of the COC conformity certificate, corresponding vehicle information is searched in the MES system, and then other contents in the vehicle documents of the COC conformity certificate are compared with the contents of the vehicle information. If the comparison results are completely consistent, the completely consistent comparison results are sent to the display device and the voice playback device. If there is one or several items in the comparison results that are inconsistent, the comparison results and the inconsistent contents are sent to the display device and the voice playback device.

[0066] Compared with the prior art, the method for identifying vehicle documents of the present invention can detect all the contents in multiple vehicle documents simultaneously, improving the detection efficiency and having high detection accuracy.

[0067] In this embodiment, the vehicle information can be searched for and compared, improving the detection quality and visually displaying the comparison results for the user to view.

[0068] Please refer to Figure 4 and Figure 5 , the vehicle document identification system according to an embodiment of the present invention includes:

[0069] Airframe 1;

[0070] Multiple detection windows 2, the detection windows 2 are arranged on the top of the airframe 1;

[0071] Multiple cameras 3, the cameras 3 are arranged inside the airframe 1, the cameras 3 correspond to the detection windows 2 one by one, and the cameras 3 collect images of vehicle documents on the detection windows 2;

[0072] A storage, the storage stores a computer program;

[0073] A processor, when the processor executes the computer program, implements the steps of the method for identifying vehicle documents. The processor sends the collected identification data and the comparison results to the storage for saving;

[0074] A display screen 6, the display screen 6 is detachably connected to the fuselage 1, and the display screen 6 receives and displays the identification data and comparison results of the processor.

[0075] Preferably, a workbench extends from one side of the top of the fuselage 1, facilitating the operator to place other office supplies.

[0076] The number of the vehicle documents is 4, including a vehicle certificate of conformity, a COC conformity certificate, an energy consumption label, and an environmental protection information list. The number of the detection windows 2 is 4, including a first detection window, a second detection window, a third detection window, and a fourth detection window. Among them, the first detection window is used to place the vehicle certificate of conformity, the second detection window is used to place the COC conformity certificate, the third detection window is used to place the energy consumption label, and the fourth detection window is used to place the environmental protection information list. Preferably, the first detection window, the second detection window, the third detection window, and the fourth detection window are arranged horizontally in sequence on one side of the top of the fuselage, facilitating the operator, the vehicle owner, or a third-party person to place the vehicle documents on the detection windows.

[0077] In other embodiments, the position order of the first detection window, the second detection window, the third detection window, and the fourth detection window is not limited, and those skilled in the art can freely change the position order of the first detection window, the second detection window, the third detection window, and the fourth detection window.

[0078] In other feasible embodiments, the number of the vehicle documents and the detection windows 2 can be set according to the detection requirements, and is not limited to these 4.

[0079] The number of the cameras 3 is 4, including a first camera corresponding to the first detection window, a second camera corresponding to the second detection window, a third camera corresponding to the third detection window, and a fourth camera corresponding to the fourth detection window. The specific process of receiving images of multiple vehicle documents is: the images of the vehicle documents on the first detection window, the second detection window, the third detection window, and the fourth detection window collected by the first camera, the second camera, the third camera, and the fourth camera.

[0080] Preferably, the number of the display screens 6 is 2, and the 2 display screens 6 are located on both sides of the fuselage 1. One of the display screens 6 is for the operator to view, and the other display screen 6 is for the vehicle owner or a third-party person to view.

[0081] In other feasible embodiments, the number of the display screens 6 can be set according to the detection requirements, and is not limited to 2.

[0082] In this embodiment, the user places the vehicle documents on the detection window 2, the camera 3 captures an image of the vehicle documents on the detection window 2, and sends the image of the vehicle documents to the processor. After the processor detects the outer frame of the image of the vehicle documents, correspondingly extracts the content within the outer frame as vehicle information and performs image preprocessing on the vehicle information, the processor identifies the vehicle information to obtain identification data, and the processor compares the identification data; the storage receives the identification data and the result of the comparison and saves them. The user can directly view the identification data and the result of the comparison from the display screen 6, or can call and view the identification data and the result of the comparison saved by the storage.

[0083] Compared with the prior art, the vehicle document identification system of the present invention can simultaneously detect all the content in multiple vehicle documents, improving the detection efficiency and having high detection accuracy.

[0084] In one embodiment, a voice playback device (not shown in the figure) is further included. The voice playback device is fixedly connected to the body, and the voice playback device receives and plays the result of the comparison by the processor. Through the voice playback device, the result of the comparison can be notified to nearby personnel at one time.

[0085] In one embodiment, a plurality of support rods 101 are provided at the top of the body 1, and the display screen 6 is detachably connected to the support rods 101. Specifically, the number of the support rods 101 corresponds to the number of the display screens 6, and is bolted to the corresponding display screen 6. Preferably, the support rods 101 are pivotally connected to the top of the body 1 and can rotate around the pivot connection. The support rods 101 are used to support the display screen 6, preventing the display screen 6 from occupying the position on the outer surface of the body 1.

[0086] In one embodiment, a plurality of sliding rods 103 corresponding to the cameras 3 one by one are provided inside the body 1. Each camera 3 is movably connected to each sliding rod 103, and the camera 3 can move along the sliding rod 103, facilitating the user to adjust the position of the camera 3.

[0087] Optionally, the sliding rods in the sliding rods 103 are vertically arranged, and the camera 3 can move up and down along the sliding rod 103, facilitating the user to adjust the height of the camera 3. Specifically, after the camera 3 is adjusted to a suitable height, the camera 3 and the sliding rod 103 can be relatively fixed through a bolt structure to prevent the camera 3 from sliding down naturally due to gravity.

[0088] Optionally, the sliding rod 103 is arranged in the horizontal direction, and the camera 3 can move along the sliding rod 103 in the horizontal direction, which is convenient for the user to adjust the horizontal position of the camera 3. Specifically, after adjusting the camera 3 to a suitable horizontal position, the camera 3 and the sliding rod 103 can be relatively fixed through a bolt structure to prevent the position of the camera 3 from shifting due to collision.

[0089] Preferably, the camera 3 is movably connected to the sliding rod through a sliding member 7. The sliding member 7 can move along the sliding rod 103, and the camera 3 is pivotally connected to the sliding member 7 and rotates around the pivot joint, which is convenient for the user to adjust the angle of the camera 3. Specifically, when adjusting the angle of the camera 3, a spirit level can be used or the image effect of the camera 3 can be directly viewed, and the camera 3 is gradually rotated around the pivot joint with the sliding member 7 to achieve a better adjustment effect.

[0090] In one embodiment, a plurality of auxiliary lights 8 are further included. The auxiliary lights 8 are arranged inside the fuselage 1, and the auxiliary lights 8 correspond to the detection windows 2 one by one. The auxiliary lights 8 irradiate light towards the detection windows 2 to improve the image accuracy of the camera 3 for collecting the vehicle documents. Preferably, the light intensity, color, and light range of the auxiliary lights 8 correspond to the types of the detection windows 2, that is, the light intensity, color, and light range of different auxiliary lights 8 are not necessarily the same. Specifically, the auxiliary lights 8 include a first auxiliary light, a second auxiliary light, a third auxiliary light, and a fourth auxiliary light, and the light intensity, color, and light range of the first auxiliary light, the second auxiliary light, the third auxiliary light, and the fourth auxiliary light correspond to the first detection window, the second detection window, the third detection window, and the fourth detection window respectively.

[0091] Please refer to Figure 6 , in one embodiment, the detection window 2 is provided with a transparent support plate for placing the vehicle documents, and the lens of the camera 3 faces the transparent support plate. Ensure that the camera 3 can accurately collect the images of the vehicle documents. Among them, the material of the transparent support plate can be transparent glass or a transparent composite material. The user places the vehicle documents on the transparent support plate, and the front side of the vehicle documents faces the transparent support plate. The camera 3 can accurately collect the images of the vehicle documents through the transparent support plate.

[0092] In one embodiment, a support member 105 is provided at the bottom of the fuselage 1 to prevent the bottom of the fuselage 1 from directly contacting the ground and protect the bottom of the fuselage 1. Among them, the support member 105 can be a leg group 1051, a pulley group 1052 or a combination of both. Preferably, the support member 105 is a combination of the leg group 1051 and the pulley group 1052, wherein the height of the leg group 1051 is adjustable. When it is necessary to move the fuselage 1, the leg group 1051 can be adjusted to move away from the ground, and then the fuselage 1 can be moved through the pulley group 1052; when it is necessary to fix the fuselage 1, the leg group 1051 can be adjusted to contact the ground.

[0093] Although the embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for identifying vehicle - accompanying certificates, characterized in that, it includes the following steps: Receiving images of multiple vehicle - accompanying certificates; wherein, the images of the vehicle - accompanying certificates are collected by multiple cameras respectively for the vehicle - accompanying certificates on multiple detection windows; Respectively detecting the outer frames of the images of each of the vehicle - accompanying certificates, and correspondingly extracting the content within the outer frames as vehicle - accompanying materials; Performing image pre - processing on the vehicle - accompanying materials; Using deep - learning CNN to identify the content in the complex - background areas of the vehicle - accompanying materials; using OCR to identify the content in the single - background areas of the vehicle - accompanying materials; Collecting identification data and making comparisons, including: searching for corresponding vehicle information in the mes system according to the identification data and making comparisons; sending the comparison results to a display device.

2. The method for identifying vehicle - accompanying certificates according to claim 1, characterized in that: After extracting the content within the outer frame as vehicle - accompanying materials, the position of the title of the vehicle - accompanying materials is also detected, and the position of the vehicle - accompanying materials is corrected and positioned according to the title position.

3. The method for identifying vehicle - accompanying certificates according to claim 2, characterized in that: Using deep - learning CNN to identify the content in the complex - background areas of the vehicle - accompanying materials; using OCR to identify the content in the single - background areas of the vehicle - accompanying materials, specifically including the following steps: selecting the content in the complex - background areas from the vehicle - accompanying materials through a preset number of first identification areas, and using deep - learning CNN to identify the content selected by the first identification areas; Using OCR to identify the content in the vehicle - accompanying materials that is not selected by the first identification areas; The position and selection range of the first identification area correspond to the content of the title of the vehicle - accompanying materials, and the position of the first identification area is relatively fixed with respect to the title position.

4. The method for identifying vehicle - accompanying certificates according to claim 1, characterized in that: After respectively detecting the outer frames of the images of each of the vehicle - accompanying certificates, it is also determined whether the vehicle - accompanying certificate corresponds to the detection window according to the size of the outer frame of the detected image of the vehicle - accompanying certificate.

5. A system for identifying vehicle - accompanying certificates, characterized in that, it includes: A fuselage; Multiple detection windows, which are arranged on the top of the fuselage; Multiple cameras, which are arranged inside the fuselage, and the cameras correspond to the detection windows one by one, and the cameras collect images of the vehicle - accompanying certificates on the detection windows; A memory, which stores a computer program; A processor, when the processor executes the computer program, it realizes the steps of the method according to any one of claims 1 to 4, and the processor sends the collected identification data and comparison results to the memory for storage; A display screen, which is detachably connected to the fuselage, and the display screen receives and displays the identification data and comparison results of the processor.

6. The system for identifying vehicle - accompanying certificates according to claim 5, characterized in that: It further includes a voice - playing device, which is fixedly connected to the fuselage, and the voice - playing device receives and plays the comparison results of the processor.

7. An identification system for vehicle documents according to claim 5, characterized in that: a plurality of support rods are provided at the top of the fuselage, and the display screen is detachably connected to the support rods.

8. An identification system for vehicle documents according to claim 5, characterized in that: a sliding rod group is provided inside the fuselage, the plurality of cameras are respectively movably connected to the sliding rods, and the plurality of cameras can move along the sliding rods.

9. An identification system for vehicle documents according to claim 5, characterized in that: the detection window is provided with a transparent support plate for placing vehicle documents, and the lens of the camera faces the transparent support plate.

Images