Vehicle verification method, device, electronic device and medium

Abstract

The present application provides a vehicle verification method, device, electronic device and medium, wherein the method comprises: after receiving a vehicle verification request sent by a user terminal, sending an image upload instruction carrying a shooting magnification rule to the user terminal, so that the user terminal uploads a vehicle image group obtained by shooting the vehicle to be verified; the shooting magnification rule is used to characterize the standard shooting magnification change reflected by multiple vehicle images in the vehicle image group uploaded by the user terminal; according to the multiple vehicle images in the vehicle image group, the actual shooting magnification change is determined; according to the degree of conformity between the actual shooting magnification change and the standard shooting magnification change, the vehicle to be verified is verified, which can effectively intercept false vehicle images shot by cheating behaviors such as screen snapping and paper snapping, ensure the authenticity and validity of the vehicle image, thereby improving the passenger experience and protecting the safety of passengers' lives and property.

Description

Technical Field

[0001] The present application relates to the field of vehicle verification technology, and in particular to a vehicle verification method, device, electronic device and medium. Background Art

[0002] With the rise of online ride-hailing, it has become the main mode of transportation for people to go out and take a taxi. More and more private car owners have also joined the online ride-hailing camp. However, in order to attract drivers to join, online ride-hailing platforms review vehicles by uploading information online for certification.

[0003] Online car-hailing registration requires drivers to upload real vehicle images, but a small number of drivers have vehicles that are not compliant or do not use their own vehicles to accept orders. In this case, the driver will verify the vehicle image by cheating by taking a photo of the screen or paper. The existing vehicle verification scheme can only identify the vehicle. Even if the vehicle image is a fake vehicle image taken by cheating by taking a photo of the screen or paper, the verification will be successful. If a passenger takes a non-compliant vehicle or a vehicle that does not match the person, it will affect the passenger experience and even endanger the safety of the passenger's life and property. Summary of the invention

[0004] In view of this, the purpose of the present application is to provide a vehicle verification method, device, electronic device and medium, which can verify whether a vehicle is a vehicle in a real scene by comparing the consistency between the standard camera magnification change when shooting the vehicle and the actual shooting magnification change of the captured vehicle image group.

[0005] According to a first aspect of the present application, a vehicle verification method is provided, comprising:

[0006] After receiving a vehicle verification request sent by a user terminal, an image upload instruction carrying a shooting magnification rule is sent to the user terminal, so that the user terminal uploads a vehicle image group obtained by shooting the vehicle to be verified; the shooting magnification rule is used to characterize the standard shooting magnification changes reflected by multiple vehicle images in the vehicle image group uploaded by the user terminal;

[0007] Determine actual shooting magnification change according to multiple vehicle images in the vehicle image group;

[0008] The vehicle to be verified is verified according to the degree of conformity between the actual shooting magnification change and the standard shooting magnification change.

[0009] In a possible implementation, the method further includes:

[0010] Determining, according to a plurality of vehicle images in the vehicle image group, a change in image blur levels between the plurality of vehicle images;

[0011] Based on the change in the blur level of the image, adjusting the change in the actual shooting magnification;

[0012] According to the degree of conformity between the actual shooting magnification change and the standard shooting magnification change, the vehicle to be verified is verified, including:

[0013] The vehicle to be verified is verified according to the conformity between the adjusted actual shooting magnification change and the standard shooting magnification change.

[0014] In a possible implementation, determining the actual shooting magnification change according to the multiple vehicle images in the vehicle image group includes:

[0015] Determining vehicle features of the plurality of vehicle images in the vehicle image group according to the plurality of vehicle images in the vehicle image group;

[0016] Based on the vehicle features of the plurality of vehicle images in the vehicle image group, actual shooting magnification changes among the plurality of vehicle images in the vehicle image group are determined.

[0017] In a possible implementation, determining, according to a plurality of vehicle images in the vehicle image group, a change in image blur levels between the plurality of vehicle images includes:

[0018] Based on vehicle features of a plurality of vehicle images in the vehicle image group, detecting a vehicle region position of each vehicle image in the vehicle image group;

[0019] Based on the vehicle region position of each vehicle image in the vehicle image group, performing image segmentation on the vehicle image;

[0020] Based on the multiple target vehicle images obtained after image segmentation, a change in image blur levels between the multiple target vehicle images obtained after image segmentation is determined.

[0021] In a possible implementation, adjusting the actual shooting magnification change based on the image blur level change includes:

[0022] Performing weighted calculation on the change of the image blur degree and the change of the actual shooting magnification;

[0023] The actual shooting magnification change is adjusted according to the result obtained by weighted calculation.

[0024] In a possible implementation manner, the shooting magnification rule is used to characterize the shooting magnification of each vehicle image among a plurality of vehicle images shot in a predetermined shooting order.

[0025] According to a second aspect of the present application, a verification method is provided, comprising:

[0026] After receiving the verification request sent by the user terminal, an image upload instruction carrying a shooting magnification rule is sent to the user terminal, so that the user terminal uploads a feature image group obtained by shooting the feature object to be verified; the shooting magnification rule is used to characterize the standard shooting magnification change reflected by multiple feature images in the feature image group uploaded by the user terminal;

[0027] Determine actual shooting magnification change according to multiple feature images in the feature image group;

[0028] The feature object to be verified is verified according to the degree of conformity between the actual shooting magnification change and the standard shooting magnification change.

[0029] In a possible implementation, the method further includes:

[0030] Determining, according to a plurality of characteristic images in the characteristic image group, a change in image blur degree between the plurality of characteristic images;

[0031] Based on the change in the blur level of the image, adjusting the change in the actual shooting magnification;

[0032] According to the degree of conformity between the actual shooting magnification change and the standard shooting magnification change, the feature object to be verified is verified, including:

[0033] The feature object to be verified is verified according to the conformity between the adjusted actual shooting magnification change and the standard shooting magnification change.

[0034] In a possible implementation, determining the actual shooting magnification change according to the multiple feature images in the feature image group includes:

[0035] Determining target features of the plurality of feature images in the feature image group according to the plurality of feature images in the feature image group;

[0036] Based on the target features of the plurality of feature images in the feature image group, actual shooting magnification changes between the plurality of feature images in the feature image group are determined.

[0037] In a possible implementation, determining, according to a plurality of characteristic images in a characteristic image group, a change in image blur degree between the plurality of characteristic images includes:

[0038] Based on target features of a plurality of feature images in the feature image group, detecting a target region position of each feature image in the feature image group;

[0039] Based on the target area position of each feature image in the feature image group, performing image segmentation on the feature image;

[0040] Based on the multiple target feature images obtained after the image segmentation, the change of the image blur degree between the multiple target feature images obtained after the image segmentation is determined.

[0041] According to a third aspect of the present application, a vehicle verification device is provided, comprising:

[0042] The instruction sending module is used to send an image upload instruction carrying a shooting magnification rule to the user terminal after receiving a vehicle verification request sent by the user terminal, so that the user terminal uploads a vehicle image group obtained by shooting the vehicle to be verified; the shooting magnification rule is used to characterize the standard shooting magnification change reflected by multiple vehicle images in the vehicle image group uploaded by the user terminal;

[0043] A magnification determination module, used to determine actual shooting magnification changes based on a plurality of vehicle images in the vehicle image group;

[0044] The vehicle verification module is used to verify the vehicle to be verified according to the conformity between the actual shooting magnification change and the standard shooting magnification change.

[0045] In a possible implementation, the device further includes:

[0046] A blur level determination module, used to determine the change of image blur levels between the multiple vehicle images in the vehicle image group;

[0047] A magnification adjustment module, used for adjusting the actual shooting magnification change based on the image blur degree change;

[0048] The vehicle verification module comprises:

[0049] The vehicle verification unit is used to verify the vehicle to be verified according to the conformity between the adjusted actual shooting magnification change and the standard shooting magnification change.

[0050] In a possible implementation, the magnification determination module includes:

[0051] a vehicle feature determination unit, configured to determine vehicle features of a plurality of vehicle images in the vehicle image group based on the plurality of vehicle images in the vehicle image group;

[0052] The actual magnification determination unit is used to determine the actual shooting magnification change between the multiple vehicle images in the vehicle image group based on the vehicle features of the multiple vehicle images in the vehicle image group.

[0053] In a possible implementation, the blur level determination module includes:

[0054] a position detection unit, configured to detect a vehicle region position of each vehicle image in the vehicle image group based on vehicle features of a plurality of vehicle images in the vehicle image group;

[0055] An image segmentation unit, configured to perform image segmentation on each vehicle image in the vehicle image group based on a vehicle region position of the vehicle image;

[0056] The blur degree determination unit is used to determine the change of image blur degrees between the multiple target vehicle images obtained after image segmentation based on the multiple target vehicle images obtained after image segmentation.

[0057] In a possible implementation, the magnification adjustment module includes:

[0058] A weighted calculation unit, used for performing weighted calculation on the change of the image blur degree and the change of the actual shooting magnification;

[0059] The magnification adjustment unit is used to adjust the actual shooting magnification change according to the result obtained by weighted calculation.

[0060] In a possible implementation manner, the shooting magnification rule is used to characterize the shooting magnification of each vehicle image among a plurality of vehicle images shot in a predetermined shooting order.

[0061] According to a fourth aspect of the present application, a verification device is provided, including:

[0062] A sending module is used to send an image upload instruction carrying a shooting magnification rule to the user terminal after receiving a verification request sent by the user terminal, so that the user terminal uploads a feature image group obtained by shooting the feature object to be verified; the shooting magnification rule is used to characterize the standard shooting magnification change reflected by multiple feature images in the feature image group uploaded by the user terminal;

[0063] A first determination module, used to determine the actual shooting magnification change according to a plurality of characteristic images in the characteristic image group;

[0064] The verification module is used to verify the feature object to be verified according to the conformity between the actual shooting magnification change and the standard shooting magnification change.

[0065] In a possible implementation, the device further includes:

[0066] A second determination module is used to determine the change of image blur degree between the multiple feature images according to the multiple feature images in the feature image group;

[0067] An adjustment module, used for adjusting the change of the actual shooting magnification based on the change of the image blur degree;

[0068] The verification module comprises:

[0069] The verification unit is used to verify the feature object to be verified according to the conformity between the adjusted actual shooting magnification change and the standard shooting magnification change.

[0070] In a possible implementation manner, the first determining module includes:

[0071] a feature determination unit, configured to determine target features of the plurality of feature images in the feature image group according to the plurality of feature images in the feature image group;

[0072] The magnification determination unit is used to determine the actual shooting magnification change between the multiple feature images in the feature image group based on the target features of the multiple feature images in the feature image group.

[0073] In a possible implementation manner, the second determining module includes:

[0074] A detection unit, configured to detect a target region position of each feature image in the feature image group based on target features of a plurality of feature images in the feature image group;

[0075] a segmentation unit, configured to perform image segmentation on each feature image in the feature image group based on a target region position of the feature image;

[0076] The determination unit is used to determine the change of image blur degree between the multiple target feature images obtained after image segmentation based on the multiple target feature images obtained after image segmentation.

[0077] According to the fifth aspect of the present application, an electronic device is provided, comprising: a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the steps of any possible implementation method described in the first aspect or the second aspect above.

[0078] According to a sixth aspect of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of any possible implementation method described in the first aspect or the second aspect are executed.

[0079] The present application provides a vehicle verification method, device, electronic device and medium. In the method, after receiving a vehicle verification request sent by a user terminal, an image upload instruction carrying a shooting magnification rule is sent to the user terminal, so that the user terminal uploads a vehicle image group obtained by shooting the vehicle to be verified; the shooting magnification rule is used to characterize the standard shooting magnification change reflected by multiple vehicle images in the vehicle image group uploaded by the user terminal; the actual shooting magnification change is determined according to the multiple vehicle images in the vehicle image group; and the vehicle to be verified is verified according to the degree of conformity between the actual shooting magnification change and the standard shooting magnification change. The solution provided by the present application, after receiving a vehicle verification request sent by a user terminal, an image upload instruction carrying a shooting magnification rule is sent to the user terminal, and the user terminal uploads the vehicle image group obtained by shooting the vehicle to be verified. In the process of shooting the vehicle image group, there is no need for the user to manually shoot, thereby ensuring the real-time and authenticity of the vehicle image group in the current scene.

[0080] Specifically, the existing vehicle verification scheme can only identify the vehicle. Even if the vehicle image is a fake vehicle image taken by cheating behaviors such as slapping the screen or slapping paper, the verification will be successful. The existing vehicle verification scheme cannot distinguish whether the vehicle is a vehicle in a real scene. After conducting a lot of research and experiments, the inventors of this application found that the actual shooting magnification change of the feature image obtained by cheating methods such as slapping the screen or slapping paper does not match the standard shooting magnification change; while the actual shooting magnification change of the vehicle image in the real scene is consistent with the standard shooting magnification change. Therefore, by verifying whether the vehicle is a vehicle in a real scene by comparing the standard camera magnification change when shooting the vehicle with the actual shooting magnification change of the vehicle image group, it is possible to effectively intercept fake vehicle images taken by cheating behaviors such as slapping the screen or slapping paper, ensure the authenticity and effectiveness of the vehicle image, thereby improving the passenger experience and protecting the safety of passengers' lives and property. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0082] Figure 1 A flow chart of a vehicle verification method provided by an embodiment of the present application is shown;

[0083] Figure 2A basic structural diagram of a shooting magnification recovery network model provided in an embodiment of the present application is shown;

[0084] Figure 3 A flow chart showing a method for adjusting the actual shooting magnification change in the vehicle verification method provided in an embodiment of the present application is shown;

[0085] Figure 4 The overall structural diagram of the shooting magnification recovery network model provided by the embodiment of the present application is shown;

[0086] Figure 5 A flow chart of a verification method provided by an embodiment of the present application is shown;

[0087] Figure 6 A flow chart of a method for adjusting the actual shooting magnification change in the verification method provided in an embodiment of the present application is shown;

[0088] Figure 7 A schematic diagram of the structure of a vehicle verification device provided in an embodiment of the present application is shown;

[0089] Figure 8 A schematic diagram of the structure of a verification device provided in an embodiment of the present application is shown;

[0090] Fig. 9 A schematic structural diagram of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0091] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the drawings in the present application only serve the purpose of explanation and description and are not used to limit the scope of protection of the present application. In addition, it should be understood that the schematic drawings are not drawn in real proportion. The flowchart used in this application shows the operations implemented according to some embodiments of the present application. It should be understood that the operations of the flowchart can be implemented out of sequence, and the steps without logical context can be reversed in order or implemented simultaneously. In addition, those skilled in the art can add one or more other operations to the flowchart under the guidance of the content of the present application, or remove one or more operations from the flowchart.

[0092] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0093] In order to enable those skilled in the art to use the contents of this application, the following implementation is provided in conjunction with the specific application scenario "online car-hailing verification". For those skilled in the art, the general principles defined herein can be applied to other embodiments and application scenarios without departing from the spirit and scope of this application. Although this application is mainly described around "online car-hailing verification", it should be understood that this is only an exemplary embodiment.

[0094] It should be noted that the term "comprising" will be used in the embodiments of the present application to indicate the existence of the features declared thereafter, but does not exclude the addition of other features.

[0095] Considering that online car-hailing registration requires drivers to upload real vehicle images, but there are a small number of drivers whose vehicles are not compliant or who do not use their own vehicles to accept orders, etc., the drivers will verify the images of the vehicles by cheating by taking pictures of the screen or paper. If a passenger takes a non-compliant vehicle or a vehicle that does not match the person and the vehicle, it will affect the passenger experience and even endanger the safety of the passenger's life and property. The existing technology cannot verify whether the vehicle in the vehicle image uploaded by the driver is the vehicle in the real scene. Based on this, the embodiments of the present application provide a vehicle verification method and device, which are described below through the following specific embodiments.

[0096] Based on the above technical problems, the embodiment of the present application provides a flow chart of a vehicle verification method, such as Figure 1 As shown, the following steps may be included:

[0097] S101. After receiving a vehicle verification request sent by a user terminal, an image upload instruction carrying a shooting magnification rule is sent to the user terminal, so that the user terminal uploads a vehicle image group obtained by shooting the vehicle to be verified; the shooting magnification rule is used to characterize the standard shooting magnification change reflected by multiple vehicle images in the vehicle image group uploaded by the user terminal;

[0098] S102, determining actual shooting magnification change according to the multiple vehicle images in the vehicle image group;

[0099] S103: Verify the vehicle to be verified according to the degree of conformity between the actual shooting magnification change and the standard shooting magnification change.

[0100] In step S101, the user terminal sends a vehicle verification request to the server. After receiving the vehicle verification request, the server sends an image upload instruction with a shooting magnification rule to the user terminal. After receiving the image upload instruction, the user terminal shoots the vehicle to be verified, obtains a vehicle image group, and uploads the vehicle image group to the server.

[0101] The specific words involved in step S101 are explained below:

[0102] User terminal: The device type corresponding to the user terminal may be a mobile device, such as a tablet computer, a smart phone, etc.

[0103] Server: A server is a type of computer that interacts with a client through a network and can process information and / or data related to a service request to perform one or more functions described in this application.

[0104] Vehicle Verification Request: A vehicle verification request is a request for vehicle verification that is triggered after a user account is successfully registered to verify whether the vehicle is real and valid. The user here usually refers to an online car-hailing driver, which can be simplified to a driver.

[0105] Image upload instruction: The image upload instruction includes a magnification rule, instruction information for instructing the user terminal to take a photo of the vehicle to be verified based on the magnification rule, and instruction information for instructing the user terminal to upload the captured vehicle image group to the server.

[0106] Vehicle image group: The user end collects the first frame of vehicle image according to the initial shooting magnification. Every time the shooting magnification changes, the user end continues to collect a frame of vehicle image, and finally forms a group of images, namely, the vehicle image group.

[0107] Shooting magnification rule: The shooting magnification rule is used to characterize the shooting magnification of each vehicle image in multiple vehicle images taken in a predetermined shooting order. For example, the shooting magnification is based on 1.5 to ensure that the shooting magnification can show a downward trend in the initial stage, with a step size of ±0.25, which ensures that the vehicle in the image has a more obvious change, and 1.0 and 2.0 are the upper and lower boundaries of the shooting magnification, 1.0 is the lowest shooting magnification of the mobile phone, and 2.0 is the upper limit to ensure that the vehicle does not exceed the image boundary. Finally, a set of standard shooting magnification sequences (1.5, 1.75, 2.0, 1.75, 1.5, 1.25, 1.5) are obtained to characterize the shooting magnification rule.

[0108] It should be noted that a group of shooting magnifications (1.5, 1.75, 2.0, 1.75, 1.5, 1.25, 1.5) is taken as an example in order to teach those skilled in the art how to implement the present invention. In a specific implementation, a group of shooting magnifications in other sorting methods may also be adopted, such as ascending order, descending order, descending order then ascending order, random sorting, etc., as long as the shooting magnification varies within the upper and lower boundaries of the shooting magnification.

[0109] Changes in standard shooting magnification: The standard shooting magnification sequence (1.5, 1.75, 2.0, 1.75, 1.5, 1.25, 1.5) is used to characterize the shooting magnification pattern. The initial base magnification is coded as 0, the magnification increase is coded as 1, the magnification decrease is coded as -1, and the magnification is equal to 0, forming the final code (0, 1, 1, -1, -1, -1, 1) to characterize the standard shooting magnification coding sequence to reflect the changes in the standard shooting magnification.

[0110] It should be noted that the coding method in which the initial base magnification is encoded as 0, the magnification increase is encoded as 1, the magnification decrease is encoded as -1, and the magnification equal is encoded as 0 is to teach those skilled in the art how to implement the present invention. In the specific implementation, other coding methods can also be used to convert the standard shooting magnification sequence into a standard shooting magnification coding sequence for use in the calculation of subsequent steps.

[0111] In addition, it should be noted that, assuming that the device type corresponding to the user end is a mobile phone, during the shooting process, the user only needs to hold the mobile phone, without shooting action, and shoot by calling the SDK for shooting in the mobile phone to prevent users from uploading false vehicle photos taken in advance. Shooting can be real-time shooting, or it can be started when the driver clicks to start shooting through the mobile phone, so that the driver can adjust the shooting angle before starting shooting.

[0112] Step S102 may include the following sub-steps:

[0113] Step S1021, determining vehicle features of the multiple vehicle images in the vehicle image group according to the multiple vehicle images in the vehicle image group;

[0114] Step S1022: determining actual shooting magnification changes among the multiple vehicle images in the vehicle image group based on the vehicle features of the multiple vehicle images in the vehicle image group.

[0115] It should be noted that the actual shooting magnification change is reflected by the actual shooting magnification coding sequence. For example, taking the first vehicle image as the benchmark, the initial benchmark magnification is coded as 0, the magnification increase is coded as 1, the magnification decrease is coded as -1, and the magnification equal is coded as 0, and the actual shooting magnification coding sequence can be obtained.

[0116] In the specific implementation, Figure 2 As shown, the present application constructs a shooting magnification recovery network model, which may include: a feature recognition network sub-model 10 and a shooting magnification recognition network sub-model 20. The output end of the feature recognition network sub-model 10 is connected to the input end of the shooting magnification recognition network sub-model 20.

[0117] In the above step S1021, when the feature recognition network sub-model 10 is used, the multiple vehicle images in the vehicle image group are input into the feature recognition network sub-model 10 to obtain the vehicle features of the multiple vehicle images in the vehicle image group output by the feature recognition network sub-model 10. In the above step S1022, when the shooting magnification recognition network sub-model 20 is used, the vehicle features of the multiple vehicle images in the vehicle image group are input into the shooting magnification recognition network sub-model 20 to obtain the actual shooting magnification changes between the multiple vehicle images in the vehicle image group output by the shooting magnification recognition network sub-model 20.

[0118] Preferably, the feature recognition network sub-model 10 can be a CNN network model, which is used to identify the vehicle features of multiple vehicle images in the vehicle image group. The shooting magnification recognition network sub-model 20 can be a CNN classification network model, which is used to identify the actual shooting magnification changes between multiple vehicle images in the vehicle image group. In addition, both the CNN network model and the CNN classification network model adopt a CNN lightweight network structure. The detection accuracy of the CNN lightweight network structure is comparable to that of the Resnet network structure, but compared with the Resnet network structure, the use of the CNN lightweight network structure can make the server-side inference speed faster.

[0119] It should be noted that this embodiment uses a shooting magnification recovery network model to identify the actual shooting magnification changes between multiple vehicle images in the vehicle image group in order to teach those skilled in the art how to implement the present invention. The present invention is not limited to this, and other methods may also be used to identify the actual shooting magnification changes.

[0120] In step S103, the actual shooting magnification change is reflected by the actual shooting magnification code sequence, and the standard shooting magnification change is reflected by the standard shooting magnification code sequence. The vehicle to be verified is verified according to the conformity between the actual shooting magnification code sequence and the standard shooting magnification code sequence.

[0121] The degree of conformity can also be understood as similarity. The similarity between the actual shooting magnification coding sequence and the standard shooting magnification coding sequence is used to determine whether the actual shooting magnification coding sequence is close to or even the same as the standard shooting magnification coding sequence. If so, the vehicle verification is successful; if not, the vehicle verification fails.

[0122] After conducting a lot of research and experiments, the inventors of this application found that the actual shooting magnification change of the feature image obtained by cheating methods such as tapping the screen or tapping paper does not match the standard shooting magnification change; while the actual shooting magnification change of the vehicle image in the real scene is consistent with the standard shooting magnification change.

[0123] Based on the above findings, it is possible to verify whether the vehicle is a vehicle in a real scene by comparing the standard camera magnification change when shooting the vehicle with the actual shooting magnification change of the captured vehicle image group. This can effectively intercept false vehicle images taken by cheating behaviors such as tapping the screen or paper, ensure the authenticity and validity of the vehicle images, thereby improving the passenger experience and protecting the safety of passengers' lives and property.

[0124] It should be noted that the "verification" in this application can be defined as "vehicle liveness verification". At present, liveness technology has not been applied to vehicle liveness. Liveness technology is divided into silent liveness technology and motion liveness technology. Silent liveness technology can intercept cheating behaviors such as screen slapping and paper slapping, and requires a large amount of data to train the model. Motion liveness technology requires the user to place the facial area at a specified position on the mobile phone screen and make specified actions (such as opening the mouth, closing the eyes, shaking the head, etc.) for liveness verification. The liveness technology used in the "vehicle liveness verification" in this application is silent liveness technology.

[0125] The inventors of the present application have also discovered after extensive research and experiments that the actual shooting magnification code identified by the above-mentioned embodiment may be affected by up and down jitter, front and back displacement, etc. of the user end, and thus has low accuracy.

[0126] Based on this technical problem, the embodiment of the present application provides a flow chart of a method for adjusting the actual shooting magnification change, such as Figure 3 As shown, the following steps may be included:

[0127] Step S301, determining the change of image blur levels between the multiple vehicle images in the vehicle image group;

[0128] Step S302: adjusting the actual shooting magnification change based on the change in the image blur level.

[0129] Furthermore, after obtaining the adjusted actual shooting magnification change, the vehicle to be verified can be verified according to the degree of conformity between the adjusted actual shooting magnification change and the standard shooting magnification change. As the shooting magnification increases, the collected vehicle image becomes increasingly blurred, so the degree of blur can also reflect the shooting magnification change to a certain extent. By combining the actual shooting magnification change with the image blur degree change to correct the actual shooting magnification change, the accuracy of the actual shooting magnification change can be improved.

[0130] Step S301 may include the following sub-steps:

[0131] Step S3011, based on vehicle features of a plurality of vehicle images in the vehicle image group, detecting a vehicle region position of each vehicle image in the vehicle image group;

[0132] Step S3012: performing image segmentation on each vehicle image in the vehicle image group based on the vehicle region position of the vehicle image;

[0133] Step S3013: based on the multiple target vehicle images obtained after image segmentation, determine the change in image blur degree between the multiple target vehicle images obtained after image segmentation.

[0134] In step S3011, the above-mentioned shooting magnification recovery network model is improved, such as Figure 4 As shown, the shooting magnification restoration network model also includes a vehicle area position detection network sub-model 30. The output end of the feature recognition network sub-model 10 is also connected to the vehicle area position detection network sub-model 30. In the above step S3011, when the vehicle area position detection network sub-model 30 is used, the vehicle features of the multiple vehicle images in the vehicle image group are input into the vehicle area position detection network sub-model 30 to obtain the vehicle area position of each vehicle image in the vehicle image group output by the vehicle area position detection network sub-model 30.

[0135] Preferably, the vehicle area position detection network submodel 30 can be an FPN network model. The FPN network model uses two smaller scales (10*10, 5*5). In this scenario, the proportion of vehicles in the image is not too small. The large scale has little effect on the detection result but will seriously affect the network reasoning speed. Therefore, this embodiment uses 10*10 and 5*5 scales for vehicle area position detection. Optionally, the vehicle area position can be the coordinates of the minimum rectangular box of the vehicle area.

[0136] In step S3012, based on the vehicle region position of each vehicle image in the vehicle image group, the vehicle image is segmented to obtain a target vehicle image.

[0137] The target vehicle image is the target vehicle image located in the vehicle region position in the vehicle image. If the vehicle region position can be the coordinates of the minimum rectangular frame of the vehicle region, the target vehicle image is the target vehicle image located in the minimum rectangular frame in the vehicle image.

[0138] It should be noted that the vehicle image can also be segmented manually. Specifically, the vehicle image is segmented manually using the vehicle region position of each vehicle image in the vehicle image group. For example, the vehicle image is segmented manually using an image segmentation tool according to the vehicle region position to obtain a target vehicle image located in the vehicle region position in the vehicle image.

[0139] In step S3013, the present application also constructs a blur degree regression network, which is used to regress blur images of different degrees, and then process the regressed series of blur degree scores to obtain a blur degree change trend coding sequence.

[0140] In a specific implementation, the multiple target vehicle images obtained after image segmentation are input into a blur degree regression network to obtain the image blur degree variation between the multiple target vehicle images obtained after image segmentation output by the blur degree regression network.

[0141] For example, the vehicle images with different shooting magnifications sent by the user end are input into the blur degree regression network, and the blur degree sequence of the vehicle area at different shooting magnifications output by the blur degree regression network is obtained (0, 0.3, 0.7, 1.0.0.6, 0.2). The initial baseline blur degree is encoded as 0, the blur degree is greater than the previous frame image, encoded as 1; the blur degree is less than the previous frame image, encoded as -1; the blur degree is equal to the previous frame image, encoded as 0, and the corresponding blur degree change trend encoding sequence is (0, 1, 1, 1, -1, -1) to reflect the change of image blur degree.

[0142] It should be noted that the coding method in which the initial baseline blur level is encoded as 0, the increased blur level is encoded as 1, the decreased blur level is encoded as -1, and the equal blur level is encoded as 0 is for the purpose of teaching those skilled in the art how to implement the present invention. In a specific implementation, other coding methods may be used to convert the blur level sequence into a blur level change trend coding sequence, but the present invention is not limited thereto.

[0143] In step S202, a weighted calculation is performed on the change in the image blur level and the change in the actual shooting magnification; and the change in the actual shooting magnification is adjusted according to the result of the weighted calculation.

[0144] In a specific implementation, the blur degree change trend coding sequence and the actual shooting magnification coding sequence are weighted according to their respective preset weights, and the actual shooting magnification change is adjusted according to the result of the weighted calculation. The preset weights here are weights fitted by the inventors in a large number of experiments.

[0145] Since the actual shooting magnification code identified will be affected by the user's up and down shaking, front and back displacement, etc., the accuracy is not high. Therefore, the actual shooting magnification code is combined with the blur degree code to adjust and optimize the actual shooting magnification code, which can improve the accuracy of the actual shooting magnification code.

[0146] Based on the same inventive concept, the present application also provides a flowchart of another verification method, such as Figure 5 As shown, the characteristic image group may include the following steps:

[0147] Step S501: after receiving a verification request sent by a user terminal, an image upload instruction carrying a shooting magnification rule is sent to the user terminal, so that the user terminal uploads a feature image group obtained by shooting a feature object to be verified; the shooting magnification rule is used to characterize the standard shooting magnification change reflected by multiple feature images in the feature image group uploaded by the user terminal;

[0148] Step S502, determining actual shooting magnification change according to a plurality of characteristic images in the characteristic image group;

[0149] Step S503: verifying the feature object to be verified according to the degree of conformity between the actual shooting magnification change and the standard shooting magnification change.

[0150] In step S501, the user terminal sends a verification request to the server. After receiving the verification request, the server sends an image upload instruction carrying a shooting magnification rule to the user terminal. After receiving the image upload instruction, the user terminal shoots the feature object to be verified, obtains a feature image group, and uploads the feature image group to the server.

[0151] The specific words involved in step S501 are explained below:

[0152] Verification request: A verification request can be a request for verifying a feature object that is triggered after a user account is successfully registered, in order to verify whether the feature object is authentic and valid.

[0153] Feature object: A feature object refers to any object in a real scene, such as a vehicle.

[0154] Feature image: Feature image refers to the image of the feature object taken by the user.

[0155] Feature image group: The user end collects the first frame of feature image according to the initial shooting magnification. Every time the shooting magnification changes, the user end continues to collect a frame of feature image, and finally forms a group of images, namely, the feature image group.

[0156] Shooting magnification law: The shooting magnification law is used to characterize the shooting magnification of each feature image in multiple feature images shot in a predetermined shooting order. For example: the shooting magnification is based on 1.5 to ensure that the shooting magnification can show a downward trend in the initial stage, with a step size of ±0.25, which ensures that the feature objects in the image have more obvious changes, and 1.0 and 2.0 are the upper and lower boundaries of the shooting magnification, 1.0 is the lowest shooting magnification of the mobile phone, and 2.0 is the upper limit to ensure that the feature objects will not exceed the image boundary. Finally, a set of standard shooting magnification sequences (1.5, 1.75, 2.0, 1.75, 1.5, 1.25, 1.5) are obtained to characterize the shooting magnification law.

[0157] It should be noted that a group of shooting magnifications (1.5, 1.75, 2.0, 1.75, 1.5, 1.25, 1.5) is taken as an example in order to teach those skilled in the art how to implement the present invention. In a specific implementation, a group of shooting magnifications in other sorting methods may also be adopted, such as ascending order, descending order, descending order then ascending order, random sorting, etc., as long as the shooting magnification varies within the upper and lower boundaries of the shooting magnification.

[0158] Changes in standard shooting magnification: The standard shooting magnification sequence (1.5, 1.75, 2.0, 1.75, 1.5, 1.25, 1.5) is used to characterize the shooting magnification pattern. The initial base magnification is coded as 0, the magnification increase is coded as 1, the magnification decrease is coded as -1, and the magnification is equal to 0, forming the final code (0, 1, 1, -1, -1, -1, 1) to characterize the standard shooting magnification coding sequence to reflect the changes in the standard shooting magnification.

[0159] It should be noted that the coding method in which the initial base magnification is encoded as 0, the magnification increase is encoded as 1, the magnification decrease is encoded as -1, and the magnification equal is encoded as 0 is to teach those skilled in the art how to implement the present invention. In the specific implementation, other coding methods can also be used to convert the standard shooting magnification sequence into a standard shooting magnification coding sequence for use in the calculation of subsequent steps.

[0160] In addition, it should be noted that, assuming that the device type corresponding to the user end is a mobile phone, during the shooting process, the user only needs to hold the mobile phone, without shooting action, and shoot by calling the SDK for shooting in the mobile phone to prevent users from uploading false feature object photos taken in advance. Shooting can be real-time shooting, or it can be started when the driver clicks to start shooting through the mobile phone, so that the driver can adjust the shooting angle before starting shooting.

[0161] In step S502, target features of the multiple feature images in the feature image group are determined according to the multiple feature images in the feature image group; and actual shooting magnification changes between the multiple feature images in the feature image group are determined based on the target features of the multiple feature images in the feature image group.

[0162] It should be noted that the actual shooting magnification change is reflected by a set of actual shooting magnification coding sequences. For example, taking the first feature image as the benchmark, the initial benchmark magnification is coded as 0, the magnification increase is coded as 1, the magnification decrease is coded as -1, and the magnification equal is coded as 0, and the actual shooting magnification coding sequence can be obtained.

[0163] In step S503, the actual shooting magnification change is reflected by the actual shooting magnification code sequence, and the standard shooting magnification change is reflected by the standard shooting magnification code sequence. According to the conformity between the actual shooting magnification code sequence and the standard shooting magnification code sequence, the feature object to be verified is verified.

[0164] The degree of conformity can also be understood as similarity. The similarity between the actual shooting magnification coding sequence and the standard shooting magnification coding sequence is used to determine whether the actual shooting magnification coding sequence is close to or even the same as the standard shooting magnification coding sequence. If so, the feature object verification succeeds; if not, the feature object verification fails.

[0165] After conducting a lot of research and experiments, the inventors of this application found that the actual shooting magnification change of the feature image obtained by cheating methods such as tapping the screen or tapping paper does not match the standard shooting magnification change; while the actual shooting magnification change of the vehicle image in the real scene is consistent with the standard shooting magnification change.

[0166] Based on the above findings, we can verify whether the feature object is the feature object in the real scene by the degree of conformity between the actual shooting magnification change and the standard shooting magnification change. We can effectively intercept false feature images taken by cheating behaviors such as tapping the screen or paper, and ensure the authenticity and validity of the feature images.

[0167] It should be noted that the "verification" in this application can be defined as "liveness verification". Currently, liveness technology is divided into silent liveness technology and motion liveness technology. Silent liveness technology can intercept cheating behaviors such as screen slapping and paper slapping, and requires a large amount of data to train the model. Motion liveness technology requires the user to place the facial area at a specified position on the mobile phone screen and make specified actions (such as opening the mouth, closing the eyes, shaking the head, etc.) for liveness verification. The liveness technology used in "liveness verification" in this application is silent liveness technology.

[0168] The inventors of the present application have also discovered after extensive research and experiments that the actual shooting magnification code identified by the above-mentioned embodiment may be affected by up and down jitter, front and back displacement, etc. of the user end, and thus has low accuracy.

[0169] Based on this technical problem, the embodiment of the present application provides a flow chart of a method for adjusting the actual shooting magnification change, such as Figure 6 As shown, the following steps may be included:

[0170] Step S601, determining the change of image blur degree between the multiple feature images in the feature image group;

[0171] Step S602: adjusting the actual shooting magnification change based on the change in the image blur level.

[0172] Furthermore, after obtaining the adjusted actual shooting magnification change, the feature object to be verified can be verified according to the degree of conformity between the adjusted actual shooting magnification change and the standard shooting magnification change. As the shooting magnification increases, the captured feature image becomes increasingly blurred, so the degree of blur can also reflect the shooting magnification change to a certain extent. By combining the actual shooting magnification change with the image blur degree change to correct the actual shooting magnification change, the accuracy of the actual shooting magnification change can be improved.

[0173] Step S601 may include the following sub-steps:

[0174] Step S6011, based on the target features of multiple feature images in the feature image group, detecting the target area position of each feature image in the feature image group;

[0175] Step S6012, performing image segmentation on each feature image in the feature image group based on the target region position of the feature image;

[0176] Step S6013: based on the multiple target feature images obtained after the image segmentation, determine the change in the image blur degree between the multiple target feature images obtained after the image segmentation.

[0177] In step S6012, based on the target region position of each feature image in the feature image group, the feature image is segmented to obtain a target feature image.

[0178] The target feature image is the target feature image located in the target region position in the feature image. If the target region position can be the coordinates of the minimum rectangular frame of the target object region, the target feature image is the target feature image located in the minimum rectangular frame in the feature image.

[0179] It should be noted that the feature image can also be segmented manually. Specifically, the feature image is segmented manually using the target region position of each feature image in the feature image group. For example, the feature image is segmented manually using an image segmentation tool according to the target region position to obtain a target feature image located in the target region position in the feature image.

[0180] In step S6013, the present application also constructs a blur degree regression network, which is used to regress blur images of different degrees, and then processes the regressed series of blur degree scores to obtain a blur degree change trend coding sequence.

[0181] In a specific implementation, a plurality of target feature images obtained after image segmentation are input into a blur degree regression network to obtain the image blur degree variation between the plurality of target feature images obtained after image segmentation output by the blur degree regression network.

[0182] For example, the feature images of different shooting magnifications sent by the user end are input into the blur degree regression network, and the blur degree sequence of the target area at different shooting magnifications output by the blur degree regression network is obtained (0, 0.3, 0.7, 1.0.0.6, 0.2). The initial baseline blur degree is encoded as 0, the blur degree is greater than the previous frame image, encoded as 1; the blur degree is less than the previous frame image, encoded as -1; the blur degree is equal to the previous frame image, encoded as 0, and the corresponding blur degree change trend encoding sequence is (0, 1, 1, 1, -1, -1) to reflect the change of image blur degree.

[0183] It should be noted that the coding method in which the initial baseline blur level is encoded as 0, the increased blur level is encoded as 1, the decreased blur level is encoded as -1, and the equal blur level is encoded as 0 is for the purpose of teaching those skilled in the art how to implement the present invention. In a specific implementation, other coding methods may be used to convert the blur level sequence into a blur level change trend coding sequence, but the present invention is not limited thereto.

[0184] Since the actual shooting magnification code identified will be affected by the user's up and down shaking, front and back displacement, etc., the accuracy is not high. Therefore, the actual shooting magnification code is combined with the blur degree code to adjust and optimize the actual shooting magnification code, which can improve the accuracy of the actual shooting magnification code.

[0185] Corresponding to the above vehicle verification method, the present application also provides a structural schematic diagram of a vehicle verification device, such as Figure 7 As shown, the device may include:

[0186] The instruction sending module 701 is used to send an image upload instruction carrying a shooting magnification rule to the user terminal after receiving a vehicle verification request sent by the user terminal, so that the user terminal uploads a vehicle image group obtained by shooting the vehicle to be verified; the shooting magnification rule is used to characterize the standard shooting magnification change reflected by multiple vehicle images in the vehicle image group uploaded by the user terminal;

[0187] A magnification determination module 702, used to determine actual shooting magnification changes according to a plurality of vehicle images in the vehicle image group;

[0188] The vehicle verification module 703 is used to verify the vehicle to be verified according to the conformity between the actual shooting magnification change and the standard shooting magnification change.

[0189] In a possible implementation, the device further includes:

[0190] A blur level determination module 704 is used to determine the change of image blur levels between the multiple vehicle images in the vehicle image group;

[0191] A magnification adjustment module 705, configured to adjust the actual shooting magnification change based on the image blur level change;

[0192] The vehicle verification module 703 includes:

[0193] The vehicle verification unit is used to verify the vehicle to be verified according to the conformity between the adjusted actual shooting magnification change and the standard shooting magnification change.

[0194] In a possible implementation, the ratio determination module 702 includes:

[0195] a vehicle feature determination unit, configured to determine vehicle features of a plurality of vehicle images in the vehicle image group based on the plurality of vehicle images in the vehicle image group;

[0196] The actual magnification determination unit is used to determine the actual shooting magnification change between the multiple vehicle images in the vehicle image group based on the vehicle features of the multiple vehicle images in the vehicle image group.

[0197] In a possible implementation, the blur level determination module 704 includes:

[0198] a position detection unit, configured to detect a vehicle region position of each vehicle image in the vehicle image group based on vehicle features of a plurality of vehicle images in the vehicle image group;

[0199] An image segmentation unit, configured to perform image segmentation on each vehicle image in the vehicle image group based on a vehicle region position of the vehicle image;

[0200] The blur degree determination unit is used to determine the change of image blur degrees between the multiple target vehicle images obtained after image segmentation based on the multiple target vehicle images obtained after image segmentation.

[0201] In a possible implementation, the magnification adjustment module 705 includes:

[0202] A weighted calculation unit, used for performing weighted calculation on the change of the image blur degree and the change of the actual shooting magnification;

[0203] The magnification adjustment unit is used to adjust the actual shooting magnification change according to the result obtained by weighted calculation.

[0204] In a possible implementation manner, the shooting magnification rule is used to characterize the shooting magnification of each vehicle image among a plurality of vehicle images shot in a predetermined shooting order.

[0205] Corresponding to the above verification method, the present application also provides a structural diagram of a verification device, such as Figure 8 As shown, the device may include:

[0206] The sending module 801 is used to send an image upload instruction carrying a shooting magnification rule to the user terminal after receiving a verification request sent by the user terminal, so that the user terminal uploads a feature image group obtained by shooting the feature object to be verified; the shooting magnification rule is used to characterize the standard shooting magnification change reflected by multiple feature images in the feature image group uploaded by the user terminal;

[0207] A first determination module 802 is used to determine the actual shooting magnification change according to a plurality of feature images in the feature image group;

[0208] The verification module 803 is used to verify the feature object to be verified according to the conformity between the actual shooting magnification change and the standard shooting magnification change.

[0209] In a possible implementation, the device further includes:

[0210] A second determination module 804 is used to determine the change of image blur degree between the multiple feature images in the feature image group according to the multiple feature images;

[0211] An adjustment module 805, configured to adjust the actual shooting magnification change based on the image blur level change;

[0212] The verification module 803 includes:

[0213] The verification unit is used to verify the feature object to be verified according to the conformity between the adjusted actual shooting magnification change and the standard shooting magnification change.

[0214] In a possible implementation, the first determining module 802 includes:

[0215] a feature determination unit, configured to determine target features of the plurality of feature images in the feature image group according to the plurality of feature images in the feature image group;

[0216] The magnification determination unit is used to determine the actual shooting magnification change between the multiple feature images in the feature image group based on the target features of the multiple feature images in the feature image group.

[0217] In a possible implementation, the second determining module 804 includes:

[0218] A detection unit, configured to detect a target region position of each feature image in the feature image group based on target features of a plurality of feature images in the feature image group;

[0219] a segmentation unit, configured to perform image segmentation on each feature image in the feature image group based on a target region position of the feature image;

[0220] The determination unit is used to determine the change of image blur degree between the multiple target feature images obtained after image segmentation based on the multiple target feature images obtained after image segmentation.

[0221] The present application also provides an electronic device, such as Fig. 9 As shown, it includes: a processor 901, a memory 902 and a bus 903. The memory 902 stores machine-readable instructions executable by the processor 901. When the electronic device is running, the processor 901 communicates with the memory 902 through the bus 903. When the machine-readable instructions are executed by the processor 901, the method described in the previous method embodiment is performed. For specific implementation, please refer to the method embodiment, which will not be repeated here.

[0222] The computer program product provided in the embodiment of the present application includes a computer-readable storage medium storing non-volatile program code executable by the above-mentioned processor 901. The instructions included in the program code can be used to execute the method described in the previous method embodiment. The specific implementation can be found in the method embodiment, which will not be repeated here.

[0223] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the system and device described above can refer to the corresponding process in the method embodiment, and will not be repeated in this application. In the several embodiments provided in this application, it should be understood that the disclosed system, device and method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0224] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0225] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0226] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard drives, ROM, RAM, magnetic disks, or optical disks.

[0227] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A vehicle verification method, It is characterized in that include: After receiving a vehicle verification request sent by a user terminal, an image upload instruction carrying a shooting magnification rule is sent to the user terminal, so that the user terminal uploads a vehicle image group obtained by shooting the vehicle to be verified; the shooting magnification rule is used to represent the shooting magnification of each vehicle image in a plurality of vehicle images shot in a predetermined shooting order, and the plurality of vehicle images are shot in a predetermined shooting order; Determine actual shooting magnification change according to multiple vehicle images in the vehicle image group; Verifying the vehicle to be verified according to the degree of conformity between the actual shooting magnification change and the standard shooting magnification change; The standard shooting magnification change condition is information corresponding to the shooting magnification rule and reflecting the up and down changes of the shooting magnification rule.

2. The method according to claim 1, It is characterized in that Also includes: Determining, based on a plurality of vehicle images in the vehicle image group, changes in image blur levels between the plurality of vehicle images; Based on the change in the blur level of the image, adjusting the change in the actual shooting magnification; According to the degree of conformity between the actual shooting magnification change and the standard shooting magnification change, the vehicle to be verified is verified, including: The vehicle to be verified is verified according to the conformity between the adjusted actual shooting magnification change and the standard shooting magnification change.

3. The method according to claim 2, It is characterized in that According to the multiple vehicle images in the vehicle image group, the actual shooting magnification change is determined, including: Determining vehicle features of the plurality of vehicle images in the vehicle image group according to the plurality of vehicle images in the vehicle image group; Based on the vehicle features of the plurality of vehicle images in the vehicle image group, actual shooting magnification changes among the plurality of vehicle images in the vehicle image group are determined.

4. The method according to claim 3, It is characterized in that Determining, according to a plurality of vehicle images in the vehicle image group, a change in image blur levels between the plurality of vehicle images, comprising: Based on vehicle features of a plurality of vehicle images in the vehicle image group, detecting a vehicle region position of each vehicle image in the vehicle image group; Based on the vehicle region position of each vehicle image in the vehicle image group, performing image segmentation on the vehicle image; Based on the multiple target vehicle images obtained after image segmentation, a change in image blur levels between the multiple target vehicle images obtained after image segmentation is determined.

5. The method according to claim 2, It is characterized in that Based on the change in the blur level of the image, adjusting the change in the actual shooting magnification includes: Performing weighted calculation on the change of the image blur degree and the change of the actual shooting magnification; The actual shooting magnification change is adjusted according to the result obtained by weighted calculation.

6. A verification method, It is characterized in that include: After receiving a verification request sent by a user terminal, an image upload instruction carrying a shooting magnification rule is sent to the user terminal, so that the user terminal uploads a feature image group obtained by shooting the feature object to be verified; the shooting magnification rule is used to represent the shooting magnification of each feature image in a plurality of feature images shot in a predetermined shooting order, and the plurality of feature images are shot in a predetermined shooting order; Determine actual shooting magnification change according to multiple feature images in the feature image group; Verifying the feature object to be verified according to the degree of conformity between the actual shooting magnification change and the standard shooting magnification change; The standard shooting magnification change condition is information corresponding to the shooting magnification rule and reflecting the up and down changes of the shooting magnification rule.

7. The method according to claim 6, It is characterized in that Also includes: Determining, according to a plurality of characteristic images in the characteristic image group, a change in image blur degree between the plurality of characteristic images; Based on the change in the blur level of the image, adjusting the change in the actual shooting magnification; According to the degree of conformity between the actual shooting magnification change and the standard shooting magnification change, the feature object to be verified is verified, including: The feature object to be verified is verified according to the conformity between the adjusted actual shooting magnification change and the standard shooting magnification change.

8. The method according to claim 7, It is characterized in that According to the multiple characteristic images in the characteristic image group, the actual shooting magnification change is determined, including: Determining target features of the plurality of feature images in the feature image group according to the plurality of feature images in the feature image group; Based on the target features of the plurality of feature images in the feature image group, actual shooting magnification changes between the plurality of feature images in the feature image group are determined.

9. The method according to claim 8, It is characterized in that Determining, according to a plurality of characteristic images in the characteristic image group, a change in image blur degree between the plurality of characteristic images, comprising: Based on target features of a plurality of feature images in the feature image group, detecting a target region position of each feature image in the feature image group; Based on the target area position of each feature image in the feature image group, performing image segmentation on the feature image; Based on the multiple target feature images obtained after the image segmentation, the change of the image blur degree between the multiple target feature images obtained after the image segmentation is determined.

10. A vehicle verification device, It is characterized in that include: The instruction sending module is used to send an image upload instruction carrying a shooting magnification rule to the user terminal after receiving a vehicle verification request sent by the user terminal, so that the user terminal uploads a vehicle image group obtained by shooting the vehicle to be verified; the shooting magnification rule is used to represent the shooting magnification of each vehicle image in a plurality of vehicle images shot in a predetermined shooting order, and the plurality of vehicle images are shot in a predetermined shooting order; A magnification determination module, used to determine actual shooting magnification changes based on a plurality of vehicle images in the vehicle image group; A vehicle verification module, used for verifying the vehicle to be verified according to the conformity between the actual shooting magnification change and the standard shooting magnification change; The standard shooting magnification change condition is information corresponding to the shooting magnification rule and reflecting the up and down changes of the shooting magnification rule.

11. The device according to claim 10, It is characterized in that Also includes: A blur level determination module, used to determine the change of image blur levels between the multiple vehicle images in the vehicle image group; A magnification adjustment module, used for adjusting the actual shooting magnification change based on the image blur degree change; The vehicle verification module comprises: The vehicle verification unit is used to verify the vehicle to be verified according to the conformity between the adjusted actual shooting magnification change and the standard shooting magnification change.

12. The device according to claim 11, It is characterized in that The multiplication factor determination module comprises: a vehicle feature determination unit, configured to determine vehicle features of a plurality of vehicle images in the vehicle image group based on the plurality of vehicle images in the vehicle image group; The actual magnification determination unit is used to determine the actual shooting magnification change between the multiple vehicle images in the vehicle image group based on the vehicle features of the multiple vehicle images in the vehicle image group.

13. The device according to claim 12, It is characterized in that The fuzziness degree determination module comprises: a position detection unit, configured to detect a vehicle region position of each vehicle image in the vehicle image group based on vehicle features of a plurality of vehicle images in the vehicle image group; An image segmentation unit, configured to perform image segmentation on each vehicle image in the vehicle image group based on a vehicle region position of the vehicle image; The blur degree determination unit is used to determine the change of image blur degrees between the multiple target vehicle images obtained after image segmentation based on the multiple target vehicle images obtained after image segmentation.

14. The device according to claim 11, It is characterized in that The magnification adjustment module comprises: A weighted calculation unit, used for performing weighted calculation on the change of the image blur degree and the change of the actual shooting magnification; The magnification adjustment unit is used to adjust the actual shooting magnification change according to the result obtained by weighted calculation.

15. A verification device, It is characterized in that include: A sending module, configured to send an image upload instruction carrying a shooting magnification rule to the user terminal after receiving a verification request sent by the user terminal, so that the user terminal uploads a feature image group obtained by shooting the feature object to be verified; the shooting magnification rule is used to represent the shooting magnification of each feature image in a plurality of feature images shot in a predetermined shooting order, and the plurality of feature images are shot in a predetermined shooting order; A first determination module, used to determine the actual shooting magnification change according to a plurality of characteristic images in the characteristic image group; A verification module, used to verify the feature object to be verified according to the conformity between the actual shooting magnification change and the standard shooting magnification change; The standard shooting magnification change condition is information corresponding to the shooting magnification rule and reflecting the up and down changes of the shooting magnification rule.

16. The device according to claim 15, It is characterized in that Also includes: A second determination module is used to determine the change of image blur degree between the multiple feature images according to the multiple feature images in the feature image group; An adjustment module, used for adjusting the change of the actual shooting magnification based on the change of the image blur degree; The verification module comprises: The verification unit is used to verify the feature object to be verified according to the conformity between the adjusted actual shooting magnification change and the standard shooting magnification change.

17. The device according to claim 16, It is characterized in that The first determining module comprises: a feature determination unit, configured to determine target features of the plurality of feature images in the feature image group according to the plurality of feature images in the feature image group; The magnification determination unit is used to determine the actual shooting magnification change between the multiple feature images in the feature image group based on the target features of the multiple feature images in the feature image group.

18. The device according to claim 17, It is characterized in that The second determining module comprises: A detection unit, configured to detect a target region position of each feature image in the feature image group based on target features of a plurality of feature images in the feature image group; a segmentation unit, configured to perform image segmentation on each feature image in the feature image group based on a target region position of the feature image; The determination unit is used to determine the change of image blur degree between the multiple target feature images obtained after image segmentation based on the multiple target feature images obtained after image segmentation.

19. An electronic device, It is characterized in that include: A processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the storage medium communicate via the bus, and the processor executes the machine-readable instructions to perform the steps of any method as claimed in claims 1 to 9.

20. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 9 are executed.

Images