CN118155264A - Vehicle inspection method, device, terminal and storage medium - Google Patents

Abstract

The invention provides a vehicle inspection method, a vehicle inspection device, a terminal and a storage medium. The method comprises the following steps: when a vehicle is about to drive into an inspection area, acquiring a face snap image; performing scene detection on the face snapshot image according to a preset scene detection network, and determining a snapshot scene corresponding to the face snapshot image; performing face recognition on the face snap image according to a preset face recognition network to obtain an initial face recognition result corresponding to the face snap image; and correcting the initial face recognition result according to the snap scene to obtain a target face recognition result corresponding to the face snap image. According to the invention, the target face recognition result corresponding to the face snap image can be accurately obtained according to different snap scenes, so that the possibility that misjudgment or misjudgment of face recognition equipment in the vehicle inspection process affects the inspection efficiency is reduced.

Description

Vehicle inspection method, device, terminal and storage medium

Technical Field

The present invention relates to the field of intelligent security inspection technology, and in particular to a vehicle inspection method, device, terminal and storage medium.

Background technique

Vehicle inspection is of great significance to maintaining regional stability and social security. With the development of intelligent equipment, vehicle inspection is currently usually carried out through a variety of vehicle inspection equipment.

Generally speaking, the equipment involved in vehicle inspection includes vehicle identification equipment, face recognition equipment, document recognition equipment, etc. Although the use of multiple vehicle inspection equipment has improved the degree of automation of vehicle inspection, in some scenarios, the misjudgment or misjudgment of a certain inspection equipment may affect the inspection efficiency to a certain extent. For example, when using face recognition equipment to count the occupants in the car, there may be over-recording or omission, resulting in the vehicle being unable to pass smoothly.

Summary of the invention

Embodiments of the present invention provide a vehicle inspection method, device, terminal and storage medium to solve the problem that the wrong judgment or misjudgment of face recognition equipment during the current vehicle inspection process may affect the inspection efficiency.

In a first aspect, an embodiment of the present invention provides a vehicle inspection method, comprising:

When the vehicle is about to enter the inspection area, a facial capture image is obtained;

Performing scene detection on the captured face image according to a preset scene detection network to determine the captured scene corresponding to the captured face image;

Performing face recognition on the captured face image according to a preset face recognition network to obtain an initial face recognition result corresponding to the captured face image;

The initial face recognition result is corrected according to the captured scene to obtain a target face recognition result corresponding to the captured face image.

In a possible implementation, the initial face recognition result includes a face segmentation result and a face segmentation confidence;

The initial face recognition result is corrected according to the captured scene to obtain a target face recognition result corresponding to the captured face image, including:

Determine a suspicious face segmentation result and a credible face segmentation result in the face segmentation result according to the standard face segmentation result corresponding to the captured scene;

Correcting the suspicious face segmentation result according to the standard face segmentation result, and correcting the suspicious face segmentation confidence corresponding to the suspicious face segmentation result according to the standard segmentation confidence corresponding to the standard face segmentation result, to obtain a corrected face recognition result;

A target face recognition result corresponding to the captured face image is obtained according to the credible face segmentation result, the credible face segmentation confidence corresponding to the credible face segmentation result, and the corrected face recognition result.

In a possible implementation, determining suspicious face segmentation results and credible face segmentation results in the face segmentation results according to the standard face segmentation results corresponding to the captured scene includes:

Calculating the cosine similarity and Euclidean distance between the standard face segmentation result corresponding to the snapshot scene and each of the face segmentation results, and performing convolution extraction on the standard face segmentation result corresponding to the snapshot scene and each of the face segmentation results to obtain the local information similarity between the standard face segmentation result corresponding to the snapshot scene and each of the face segmentation results;

Performing weighted summation on the cosine similarity, the Euclidean distance and the local information similarity corresponding to each of the face segmentation results to obtain the similarity corresponding to each of the face segmentation results;

According to the similarities corresponding to the face segmentation results and a preset similarity threshold, suspicious face segmentation results and credible face segmentation results in the face segmentation results are determined.

In a possible implementation, the suspicious face segmentation confidence corresponding to the suspicious face segmentation result is corrected according to the standard segmentation confidence corresponding to the standard face segmentation result, including:

Determine whether the suspicious face segmentation confidence is greater than a confidence threshold;

If the suspicious face segmentation confidence is greater than the confidence threshold, the standard segmentation confidence corresponding to the standard face segmentation result is adjusted according to the target adjustment coefficient corresponding to the suspicious face segmentation confidence, and the adjusted standard segmentation confidence is used as the corrected face segmentation confidence corresponding to the suspicious face segmentation result.

In a possible implementation, after determining whether the suspicious face segmentation confidence is greater than a confidence threshold, the method further includes:

If the suspicious face segmentation confidence is less than or equal to the confidence threshold, the standard segmentation confidence corresponding to the standard face segmentation result is used as the revised face segmentation confidence corresponding to the suspicious face segmentation result.

In a possible implementation, the target control coefficient is determined by looking up a table of correspondences between the suspicious face segmentation confidence and preset control coefficients.

In a possible implementation, the training process of the preset scene detection network includes:

Obtain the corresponding facial capture images when the vehicle is empty, lightly loaded, fully loaded, and overloaded to form a training set;

The initial clustering network is trained according to the training set to obtain a preset scene detection network.

In a second aspect, an embodiment of the present invention provides a vehicle inspection device, comprising:

An acquisition module is used to acquire a captured face image when a vehicle is about to enter an inspection area;

A first processing module, configured to perform scene detection on the captured face image according to a preset scene detection network, and determine a capture scene corresponding to the captured face image;

A second processing module is used to perform face recognition on the captured face image according to a preset face recognition network to obtain an initial face recognition result corresponding to the captured face image;

The third processing module is used to correct the initial face recognition result according to the captured scene to obtain a target face recognition result corresponding to the captured face image.

In a third aspect, an embodiment of the present invention provides a terminal, comprising a memory and a processor, wherein the memory is used to store a computer program, and the processor is used to call and run the computer program stored in the memory to perform the steps of the method described in the first aspect or any possible implementation of the first aspect.

In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method described in the first aspect or any possible implementation method of the first aspect are implemented.

The embodiment of the present invention provides a vehicle inspection method, device, terminal and storage medium, which obtains a face capture image when the vehicle is about to enter the inspection area, and then performs scene detection on the face capture image according to a preset scene detection network to determine the capture scene corresponding to the face capture image; and performs face recognition on the face capture image according to a preset face recognition network to obtain an initial face recognition result corresponding to the face capture image; and then corrects the initial face recognition result according to the capture scene to obtain a target face recognition result corresponding to the face capture image. In this way, the target face recognition result corresponding to the face capture image is accurately obtained according to different capture scenes, thereby reducing the possibility that the face recognition device's misjudgment or misjudgment affects the inspection efficiency during the vehicle inspection process.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

FIG1 is a flow chart of a vehicle inspection method according to an embodiment of the present invention;

FIG2 is a schematic diagram of the structure of a vehicle inspection device provided by an embodiment of the present invention;

FIG. 3 is a schematic diagram of a terminal provided by an embodiment of the present invention.

Detailed ways

In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

In order to make the purpose, technical solutions and advantages of the present invention more clear, specific embodiments will be described below in conjunction with the accompanying drawings.

Referring to FIG. 1 , a flow chart of the vehicle inspection method according to an embodiment of the present invention is shown, which is described in detail as follows:

In step 101, when a vehicle is about to enter an inspection area, a facial capture image is acquired.

Among them, generally when a vehicle is about to enter the inspection area for inspection, the vehicle and the occupants can be inspected by combining vehicle identification equipment, face recognition equipment, document recognition equipment, etc. The face recognition equipment can also be used directly to inspect the occupants of the vehicle, such as overloading inspection of buses, school buses, mini vans, small cars, etc., illegal owner inspection, etc.

Optionally, in order to improve the applicability of the acquired facial capture image, when the vehicle is about to enter the inspection area, the vehicle type can be first identified by a vehicle identification device, and the height of the facial recognition device can be adjusted according to the identified vehicle type, and then the facial capture image can be acquired based on the facial recognition device with the adjusted height.

In step 102, scene detection is performed on the captured face image according to a preset scene detection network to determine the captured scene corresponding to the captured face image.

In this embodiment, considering that the face capture images obtained by the face recognition device may have overlapping, occlusion and other problems, which may cause the face recognition device to record or omit the occupants in the vehicle when counting the occupants, affecting the efficiency of vehicle inspection. Therefore, a preset scene detection network is obtained in advance, and scene detection is performed on the face capture images according to the preset scene detection network to determine the capture scene corresponding to the face capture images, so as to more accurately count the occupants in the vehicle according to the capture scene corresponding to the face capture images.

Optionally, the training process of the preset scene detection network may include:

The face capture images corresponding to when the vehicle is empty, lightly loaded, fully loaded, and overloaded are obtained to form a training set, and the initial clustering network is trained according to the training set to obtain a preset scene detection network.

In this embodiment, when performing scene detection on the facial capture image, since the facial capture images corresponding to when the vehicle is empty, lightly loaded, fully loaded, and overloaded are quite different, a preset scene detection network is trained based on the facial capture images corresponding to when the vehicle is empty, lightly loaded, fully loaded, and overloaded, so as to determine the capture scene corresponding to the currently obtained facial capture image according to the preset scene detection network.

In step 103, face recognition is performed on the captured face image according to a preset face recognition network to obtain an initial face recognition result corresponding to the captured face image.

In step 104, the initial face recognition result is corrected according to the captured scene to obtain a target face recognition result corresponding to the captured face image.

In this embodiment, in addition to performing scene detection on the captured face image according to the preset scene detection network to determine the capture scene corresponding to the captured face image, face recognition is also performed on the captured face image according to the preset face recognition network to obtain an initial face recognition result corresponding to the captured face image, and then the initial face recognition result can be corrected according to the capture scene to obtain a more accurate target face recognition result corresponding to the captured face image.

The embodiment of the present invention obtains a captured face image when the vehicle is about to enter the inspection area, then performs scene detection on the captured face image according to a preset scene detection network to determine the captured scene corresponding to the captured face image; performs face recognition on the captured face image according to a preset face recognition network to obtain an initial face recognition result corresponding to the captured face image; and then corrects the initial face recognition result according to the captured scene to obtain a target face recognition result corresponding to the captured face image. In this way, the target face recognition result corresponding to the captured face image is accurately obtained according to different captured scenes, thereby reducing the possibility that the face recognition device's misjudgment or misjudgment affects the inspection efficiency during the vehicle inspection process.

Optionally, the initial face recognition result may include a face segmentation result and a face segmentation confidence.

Correcting the initial face recognition result according to the capture scene to obtain the target face recognition result corresponding to the face capture image may include:

According to the standard face segmentation result corresponding to the captured scene, the suspicious face segmentation result and the credible face segmentation result in the face segmentation result are determined.

The suspicious face segmentation result is corrected according to the standard face segmentation result, and the suspicious face segmentation confidence corresponding to the suspicious face segmentation result is corrected according to the standard segmentation confidence corresponding to the standard face segmentation result to obtain a corrected face recognition result.

The target face recognition result corresponding to the face capture image is obtained according to the credible face segmentation result, the credible face segmentation confidence corresponding to the credible face segmentation result and the corrected face recognition result.

Among them, the process of performing face recognition on the face capture image according to the preset face recognition network can be understood as the process of performing face segmentation on the face capture image. After the preset face recognition network performs face segmentation on the face capture image, the confidence of each segmentation result, that is, the face segmentation confidence, can be obtained. Since the acquired face capture image is generally one of several capture scenes of the vehicle such as no load, light load, full load, overload, etc., the standard face segmentation result of the capture scene corresponding to the face capture image can be used to identify the suspicious face segmentation result and the credible face segmentation result, and the standard face segmentation result and the standard segmentation confidence are used to correct the suspicious face segmentation result and the suspicious face segmentation confidence, so as to obtain the final target face recognition result with higher credibility.

Optionally, determining the suspicious face segmentation result and the credible face segmentation result in the face segmentation result according to the standard face segmentation result corresponding to the captured scene may include:

The cosine similarity and Euclidean distance between the standard face segmentation result corresponding to the snapshot scene and each face segmentation result are calculated, and convolution extraction is performed on the standard face segmentation result corresponding to the snapshot scene and each face segmentation result to obtain the local information similarity between the standard face segmentation result corresponding to the snapshot scene and each face segmentation result.

The cosine similarity, Euclidean distance and local information similarity corresponding to each face segmentation result are weighted and summed to obtain the similarity corresponding to each face segmentation result.

According to the similarities corresponding to the various face segmentation results and a preset similarity threshold, suspicious face segmentation results and credible face segmentation results that do not match the standard face segmentation results in the face segmentation results are determined.

In this embodiment, in order to accurately determine the suspicious face segmentation results and the credible face segmentation results in the face segmentation results, multiple angles of the standard face segmentation results corresponding to the snapshot scene and each face segmentation result are considered, the cosine similarity and Euclidean distance between the standard face segmentation results corresponding to the snapshot scene and each face segmentation result are calculated, and convolution extraction is performed on the standard face segmentation results corresponding to the snapshot scene and each face segmentation result to obtain the local information similarity between the standard face segmentation results corresponding to the snapshot scene and each face segmentation result, thereby obtaining the similarity corresponding to each face segmentation result based on the cosine similarity, Euclidean distance and local information similarity corresponding to each face segmentation result, thereby determining the suspicious face segmentation results and the credible face segmentation results in the face segmentation results based on the similarity corresponding to each face segmentation result and a preset similarity threshold.

Exemplarily, if the similarity corresponding to the face segmentation result is greater than a preset similarity threshold, the face segmentation result can be determined to be a credible face segmentation result. If the similarity corresponding to the face segmentation result is less than or equal to the preset similarity threshold, the face segmentation result can be determined to be a suspicious face segmentation result.

Optionally, correcting the suspicious face segmentation confidence corresponding to the suspicious face segmentation result according to the standard segmentation confidence corresponding to the standard face segmentation result may include:

Determine whether the suspicious face segmentation confidence is greater than the confidence threshold.

If the suspicious face segmentation confidence is greater than the confidence threshold, the standard segmentation confidence corresponding to the standard face segmentation result is adjusted according to the target adjustment coefficient corresponding to the suspicious face segmentation confidence, and the adjusted standard segmentation confidence is used as the corrected face segmentation confidence corresponding to the suspicious face segmentation result.

In this embodiment, when correcting the suspicious face segmentation confidence corresponding to the suspicious face segmentation result according to the standard segmentation confidence corresponding to the standard face segmentation result, it is possible to first determine whether the suspicious face segmentation confidence is greater than the confidence threshold. If the suspicious face segmentation confidence is greater than the confidence threshold, it means that the current face capture image is not very messy, and the suspicious face segmentation result in the obtained face segmentation result still has a certain degree of credibility. Therefore, the standard segmentation confidence corresponding to the standard face segmentation result can be adjusted according to the target control coefficient corresponding to the suspicious face segmentation confidence, and the adjusted standard segmentation confidence is used as the corrected face segmentation confidence corresponding to the suspicious face segmentation result.

Optionally, the target control coefficient can be determined by looking up a table of correspondences between suspicious face segmentation confidence levels and preset control coefficients.

Optionally, after determining whether the suspicious face segmentation confidence is greater than the confidence threshold, it may also include: if the suspicious face segmentation confidence is less than or equal to the confidence threshold, then using the standard segmentation confidence corresponding to the standard face segmentation result as the revised face segmentation confidence corresponding to the suspicious face segmentation result.

In this embodiment, after determining whether the suspicious face segmentation confidence is greater than the confidence threshold, if the suspicious face segmentation confidence is less than or equal to the confidence threshold, it means that the current face capture image is highly chaotic, and the credibility of the suspicious face segmentation result in the obtained face segmentation result is relatively low. In this case, the standard segmentation confidence corresponding to the standard face segmentation result can be directly used as the corrected face segmentation confidence corresponding to the suspicious face segmentation result.

It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

The following is an embodiment of the device of the present invention. For details not described in detail, reference may be made to the corresponding method embodiment described above.

FIG2 shows a schematic diagram of the structure of a vehicle inspection device provided by an embodiment of the present invention. For ease of description, only the parts related to the embodiment of the present invention are shown, which are described in detail as follows:

As shown in FIG. 2 , the vehicle inspection device includes: an acquisition module 21 , a first processing module 22 , a second processing module 23 and a third processing module 24 .

The acquisition module 21 is used to acquire a captured face image when the vehicle is about to enter the inspection area;

A first processing module 22, configured to perform scene detection on the captured face image according to a preset scene detection network, and determine a capture scene corresponding to the captured face image;

The second processing module 23 is used to perform face recognition on the captured face image according to a preset face recognition network to obtain an initial face recognition result corresponding to the captured face image;

The third processing module 24 is used to correct the initial face recognition result according to the captured scene to obtain a target face recognition result corresponding to the captured face image.

The embodiment of the present invention obtains a captured face image when the vehicle is about to enter the inspection area, then performs scene detection on the captured face image according to a preset scene detection network to determine the captured scene corresponding to the captured face image; performs face recognition on the captured face image according to a preset face recognition network to obtain an initial face recognition result corresponding to the captured face image; and then corrects the initial face recognition result according to the captured scene to obtain a target face recognition result corresponding to the captured face image. In this way, the target face recognition result corresponding to the captured face image is accurately obtained according to different captured scenes, thereby reducing the possibility that the face recognition device's misjudgment or misjudgment affects the inspection efficiency during the vehicle inspection process.

In a possible implementation, the initial face recognition result includes a face segmentation result and a face segmentation confidence;

The third processing module 24 may be used to determine a suspicious face segmentation result and a credible face segmentation result in the face segmentation result according to the standard face segmentation result corresponding to the captured scene;

Correcting the suspicious face segmentation result according to the standard face segmentation result, and correcting the suspicious face segmentation confidence corresponding to the suspicious face segmentation result according to the standard segmentation confidence corresponding to the standard face segmentation result, to obtain a corrected face recognition result;

A target face recognition result corresponding to the captured face image is obtained according to the credible face segmentation result, the credible face segmentation confidence corresponding to the credible face segmentation result, and the corrected face recognition result.

In a possible implementation, the third processing module 24 may be used to calculate the cosine similarity and Euclidean distance between the standard face segmentation result corresponding to the snapshot scene and each of the face segmentation results, and perform convolution extraction on the standard face segmentation result corresponding to the snapshot scene and each of the face segmentation results to obtain the local information similarity between the standard face segmentation result corresponding to the snapshot scene and each of the face segmentation results;

Performing weighted summation on the cosine similarity, the Euclidean distance and the local information similarity corresponding to each of the face segmentation results to obtain the similarity corresponding to each of the face segmentation results;

According to the similarities corresponding to the face segmentation results and a preset similarity threshold, suspicious face segmentation results and credible face segmentation results in the face segmentation results are determined.

In a possible implementation, the third processing module 24 may be used to determine whether the suspicious face segmentation confidence is greater than a confidence threshold;

If the suspicious face segmentation confidence is greater than the confidence threshold, the standard segmentation confidence corresponding to the standard face segmentation result is adjusted according to the target adjustment coefficient corresponding to the suspicious face segmentation confidence, and the adjusted standard segmentation confidence is used as the corrected face segmentation confidence corresponding to the suspicious face segmentation result.

In a possible implementation, the third processing module 24 may also be used to use the standard segmentation confidence corresponding to the standard face segmentation result as the revised face segmentation confidence corresponding to the suspicious face segmentation result if the suspicious face segmentation confidence is less than or equal to the confidence threshold.

In a possible implementation, the target control coefficient is determined by looking up a table of correspondences between the suspicious face segmentation confidence and preset control coefficients.

In a possible implementation, the training process of the preset scene detection network includes:

Obtain the corresponding facial capture images when the vehicle is empty, lightly loaded, fully loaded, and overloaded to form a training set;

The initial clustering network is trained according to the training set to obtain a preset scene detection network.

FIG3 is a schematic diagram of a terminal provided by an embodiment of the present invention. As shown in FIG3 , the terminal 3 of this embodiment includes: a processor 30, a memory 31, and a computer program 32 stored in the memory 31 and executable on the processor 30. When the processor 30 executes the computer program 32, the steps in the above-mentioned vehicle inspection method embodiments are implemented, such as steps 101 to 104 shown in FIG1 . Alternatively, when the processor 30 executes the computer program 32, the functions of the modules/units in the above-mentioned device embodiments are implemented, such as the functions of the modules/units 21 to 24 shown in FIG2 .

Exemplarily, the computer program 32 may be divided into one or more modules/units, one or more modules/units are stored in the memory 31, and are executed by the processor 30 to complete the present invention. One or more modules/units may be a series of computer program instruction segments capable of completing specific functions, and the instruction segments are used to describe the execution process of the computer program 32 in the terminal 3. For example, the computer program 32 may be divided into modules/units 21 to 24 shown in FIG. 2 .

The terminal 3 may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal 3 may include, but is not limited to, a processor 30 and a memory 31. Those skilled in the art may understand that FIG. 3 is only an example of the terminal 3 and does not constitute a limitation on the terminal 3. The terminal 3 may include more or fewer components than shown in the figure, or may combine certain components, or different components. For example, the terminal may also include input and output devices, network access devices, buses, etc.

The processor 30 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

The memory 31 may be an internal storage unit of the terminal 3, such as a hard disk or memory of the terminal 3. The memory 31 may also be an external storage device of the terminal 3, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the terminal 3. Further, the memory 31 may also include both an internal storage unit of the terminal 3 and an external storage device. The memory 31 is used to store computer programs and other programs and data required by the terminal. The memory 31 may also be used to temporarily store data that has been output or is to be output.

The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

In the embodiments provided by the present invention, it should be understood that the disclosed devices/terminals and methods can be implemented in other ways. For example, the device/terminal embodiments described above are only schematic, for example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units 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 interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components shown as units 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.

In addition, each functional unit in each embodiment of the present invention 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. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

If the integrated module/unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, it can implement the steps of the above-mentioned vehicle inspection method embodiments. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims (10)

1. A vehicle inspection method, comprising: When the vehicle is about to enter the inspection area, a facial capture image is obtained; Performing scene detection on the captured face image according to a preset scene detection network to determine the captured scene corresponding to the captured face image; Performing face recognition on the captured face image according to a preset face recognition network to obtain an initial face recognition result corresponding to the captured face image; The initial face recognition result is corrected according to the captured scene to obtain a target face recognition result corresponding to the captured face image. 2. The vehicle inspection method according to claim 1, characterized in that the initial face recognition result includes a face segmentation result and a face segmentation confidence; The initial face recognition result is corrected according to the captured scene to obtain a target face recognition result corresponding to the captured face image, including: Determine a suspicious face segmentation result and a credible face segmentation result in the face segmentation result according to the standard face segmentation result corresponding to the captured scene; Correcting the suspicious face segmentation result according to the standard face segmentation result, and correcting the suspicious face segmentation confidence corresponding to the suspicious face segmentation result according to the standard segmentation confidence corresponding to the standard face segmentation result, to obtain a corrected face recognition result; A target face recognition result corresponding to the captured face image is obtained according to the credible face segmentation result, the credible face segmentation confidence corresponding to the credible face segmentation result, and the corrected face recognition result. 3. The vehicle inspection method according to claim 2, characterized in that the step of determining the suspicious face segmentation results and the credible face segmentation results in the face segmentation results according to the standard face segmentation results corresponding to the captured scene comprises: Calculating the cosine similarity and Euclidean distance between the standard face segmentation result corresponding to the snapshot scene and each of the face segmentation results, and performing convolution extraction on the standard face segmentation result corresponding to the snapshot scene and each of the face segmentation results to obtain the local information similarity between the standard face segmentation result corresponding to the snapshot scene and each of the face segmentation results; Performing weighted summation on the cosine similarity, the Euclidean distance and the local information similarity corresponding to each of the face segmentation results to obtain the similarity corresponding to each of the face segmentation results; According to the similarities corresponding to the face segmentation results and a preset similarity threshold, suspicious face segmentation results and credible face segmentation results in the face segmentation results are determined. 4. The vehicle inspection method according to claim 2, characterized in that the suspicious face segmentation confidence corresponding to the suspicious face segmentation result is corrected according to the standard segmentation confidence corresponding to the standard face segmentation result, comprising: Determine whether the suspicious face segmentation confidence is greater than a confidence threshold; If the suspicious face segmentation confidence is greater than the confidence threshold, the standard segmentation confidence corresponding to the standard face segmentation result is adjusted according to the target adjustment coefficient corresponding to the suspicious face segmentation confidence, and the adjusted standard segmentation confidence is used as the corrected face segmentation confidence corresponding to the suspicious face segmentation result. 5. The vehicle inspection method according to claim 4, characterized in that after determining whether the suspicious face segmentation confidence is greater than the confidence threshold, it also includes: If the suspicious face segmentation confidence is less than or equal to the confidence threshold, the standard segmentation confidence corresponding to the standard face segmentation result is used as the revised face segmentation confidence corresponding to the suspicious face segmentation result. 6. The vehicle inspection method according to claim 4 is characterized in that the target control coefficient is determined by looking up a table of correspondence between the suspicious face segmentation confidence and the preset control coefficient. 7. The vehicle inspection method according to any one of claims 1 to 6, characterized in that the training process of the preset scene detection network includes: Obtain the corresponding facial capture images when the vehicle is empty, lightly loaded, fully loaded, and overloaded to form a training set; The initial clustering network is trained according to the training set to obtain a preset scene detection network. 8. A vehicle inspection device, comprising: An acquisition module is used to acquire a captured face image when a vehicle is about to enter an inspection area; A first processing module, configured to perform scene detection on the captured face image according to a preset scene detection network, and determine a capture scene corresponding to the captured face image; A second processing module is used to perform face recognition on the captured face image according to a preset face recognition network to obtain an initial face recognition result corresponding to the captured face image; The third processing module is used to correct the initial face recognition result according to the captured scene to obtain a target face recognition result corresponding to the captured face image. 9. A terminal, characterized in that it comprises a memory and a processor, wherein the memory is used to store a computer program, and the processor is used to call and run the computer program stored in the memory to execute the method according to any one of claims 1 to 7. 10. A computer-readable storage medium storing a computer program, wherein the computer program implements the steps of the method according to any one of claims 1 to 7 when executed by a processor.