CN116061878A - Vehicle keyless entry and starting method, device and storage medium - Google Patents

Abstract

The invention discloses a keyless entry and starting method of a vehicle, which comprises the following steps: acquiring first depth image information of a person approaching a vehicle door, and performing face detection processing on the first depth image information; performing living body detection processing on the first depth image information according to a face detection processing result, and performing identity verification processing on the first depth image information according to the living body detection processing result; controlling whether the vehicle door is unlocked or not according to the authentication processing result; after the vehicle door is unlocked, second depth image information of a driver of the vehicle is obtained, and face recognition verification processing is carried out on the second depth image information; and controlling whether the vehicle is started or not according to the face recognition verification processing result. A device and a storage medium for implementing the vehicle keyless entry and start method are also disclosed. The invention improves the convenience and safety of unlocking and starting the vehicle.

Description

Vehicle keyless entry and start method, device and storage medium

technical field

The present invention relates to the technical field of vehicle identification methods, in particular to a face recognition-based keyless vehicle entry and start method, device and storage medium.

Background technique

With the continuous development of automobile technology and the continuous increase of car ownership, people's demand for automobile intelligent programs is also increasing. When users buy a car, in addition to the car's driving and riding comfort requirements, whether the car is equipped with intelligent equipment is also one of the key considerations.

The keyless entry system has become a common configuration of modern cars, which provides convenience to car owners to a certain extent, but also brings certain safety issues. Most of the existing vehicle keyless entry systems mainly use the dual radio frequency system matching between the car key and the vehicle, and still require the car owner to carry the car key, so the action of taking out the car key to unlock can only be omitted. And this kind of vehicle keyless entry system generally only verifies the legitimacy of the car key, and does not have the function of identifying and verifying the identity of the car owner. When the car key is lost, if it is picked up by others, others can unlock it and enter the car. However, in real life, it is inevitable that there will be scenes where the car key is forgotten or lost, or it is not convenient to carry the car key and it is necessary to enter the vehicle and start the vehicle. Therefore, the disadvantages of the existing keyless entry system for vehicles appear.

For this reason, the applicant has found a solution to the above-mentioned problems through beneficial exploration and research, and the technical solutions to be introduced below are generated under this background.

Contents of the invention

One of the technical problems to be solved by the present invention is to provide a keyless vehicle entry and start method based on face recognition that improves the convenience and safety of unlocking and starting the vehicle in view of the deficiencies in the prior art.

The second technical problem to be solved by the present invention is to provide a device for realizing the above method for keyless entry and start of a vehicle.

The third technical problem to be solved by the present invention is to provide a storage medium for realizing the above method for keyless entry and start of a vehicle.

As a first aspect of the present invention, a keyless entry and start method for a vehicle includes:

Acquiring the first depth image information of the person close to the car door, and performing face detection processing on the first depth image information;

performing live body detection processing on the first depth image information according to the face detection processing result, and performing identity verification processing on the first depth image information according to the live body detection processing result;

Control whether the door is unlocked according to the identity verification processing result;

After the door is unlocked, acquiring the second depth image information of the driver seat of the vehicle, and performing face recognition verification processing on the second depth image information; and

Control whether the vehicle starts according to the face recognition verification processing result.

In a preferred embodiment of the present invention, the acquisition of the first depth image information of the person close to the car door includes:

When the vehicle is in the state of parking and locking the door, the depth image information outside the door is collected by the first TOF camera installed at the door at a standby acquisition frame rate;

Detecting and processing the collected depth image information to determine whether there is a person approaching the car door; and

If it is detected that there is a person approaching the door, the first TOF camera will perform image acquisition processing on the person approaching the door at a working acquisition frame rate, so as to obtain the first depth image information of the person approaching the door.

In a preferred embodiment of the present invention, the first depth image information includes a depth image and a near-infrared image.

In a preferred embodiment of the present invention, performing face detection processing on the first depth image information includes:

Obtain a near-infrared image in the first depth image information, and send the near-infrared image to a face detection module for face detection processing; and

Judging whether there is a human face in the near-infrared image according to the result of human face detection processing, if a human face is detected, then perform living body detection processing and identity verification processing, if no human face is detected, continue to pass through the first TOF camera At the working acquisition frame rate, image acquisition is performed on the person approaching the car door and the face detection process is performed again. If it is detected that the person leaves, the acquisition frame rate of the first TOF camera is adjusted to the standby acquisition frame rate.

In a preferred embodiment of the present invention, performing the living body detection processing on the first depth image information according to the face detection processing result includes:

If the face detection processing result is that a human face is detected, image adjustment processing is performed on the depth image and the near-infrared image in the first depth image information respectively;

Gradient image processing is performed on the depth image after image adjustment processing to obtain gradient images in the horizontal direction, vertical direction, and center-to-edge direction of the depth image, and merge the three gradient images into a three-channel depth image;

Combine the depth image and near-infrared image after image adjustment to obtain a three-channel grayscale image;

The three-channel depth image is sent into a depth image-based living body detection model to calculate the depth image model score;

If the depth image model score is higher than the first living body threshold, the three-channel grayscale image is sent to a living body detection model based on near-infrared images to calculate the grayscale image model score;

If the grayscale image model score is higher than the second in vivo threshold, the three-channel depth image and the three-channel grayscale image are correspondingly put into a depth image cache queue and a grayscale image cache queue;

Repeat the calculation of the model score until the queue lengths of the depth image cache queue and the grayscale image cache queue are equal to the preset values, and the three-channel depth image and the three-channel grayscale image respectively taken out from the depth image cache queue and the grayscale image cache queue are degree image, and empty the depth image cache queue and the grayscale image cache queue respectively;

Calculate the LBP-TOP features of the extracted three-channel depth image and three-channel grayscale image respectively, and splicing the calculated LBP-TOP features of the three-channel depth image and three-channel grayscale image into the time-series living body based on comprehensive features Computing time-series model scores within the detection model; and

If the score of the time series model is greater than the third living body threshold, it is determined as a living body.

In a preferred embodiment of the present invention, the image adjustment processing of the depth image and the near-infrared image in the first depth image information includes:

Performing correction and adjustment processing on the depth image and the near-infrared image respectively;

Perform filtering processing on the depth image and the near-infrared image processed by the orthodontic adjustment respectively;

performing dynamic compression processing on the filtered depth image and the near-infrared image respectively; and

Normalize the gray value of the depth image and the near-infrared image after dynamic compression processing respectively.

In a preferred embodiment of the present invention, performing identity verification processing on the first depth image information according to the living body detection processing result includes:

If the living body detection processing result is identified as a living body, acquiring a near-infrared image in the first depth image information; and

The near-infrared image is compared with the local identity library authorized to unlock the vehicle, and if the comparison is passed, the door is unlocked.

In a preferred embodiment of the present invention, the acquisition of the second depth image information of the driver's seat of the vehicle, and performing face recognition and verification processing on the second depth image information includes:

When the door is unlocked, the second TOF camera installed at the driving position of the vehicle is used to collect the depth image information outside the driving position of the vehicle at a standby acquisition frame rate;

Detect and process the collected depth image information to determine whether there is a person sitting in the driver's seat of the vehicle;

If it is detected that there is a person sitting on the driver's seat of the vehicle, the second TOF camera is used to collect images of the person in the driver's seat of the vehicle at a work acquisition frame rate, so as to obtain the second depth image information of the person in the driver's seat of the vehicle;

Sending the near-infrared image in the second depth image information to the face detection module for face detection processing; and

Whether there is a human face is judged according to the result of the face detection process. If a human face is detected, the near-infrared image is compared with the local identity database authorized to start the vehicle. If the comparison is passed, the vehicle is authorized to start.

A vehicle keyless entry and start device as the second aspect of the present invention includes:

The first TOF camera, the first TOF camera is installed at the car door, and is used to collect the first depth image information of the person close to the car door;

A second TOF camera, the second TOF camera is installed at the driver's seat of the vehicle, and is used to collect the second depth image information of the person in the driver's seat of the vehicle;

A first depth image information processing module, the first depth image information processing module is used to obtain the first depth image information of the person close to the car door, and perform face detection processing on the first depth image information;

A live body detection and identity verification module, the live body detection and identity verification module is used to perform live body detection processing on the first depth image information according to the face detection processing result, and to perform live body detection processing on the first depth image information according to the live body detection processing result. Image information for identity verification processing;

A door lock control module, the door lock control module is used to control whether the door is unlocked according to the identity verification processing result;

The second depth image information processing module, the second depth image information processing module is used to obtain the second depth image information of the driver's seat of the vehicle after the door is unlocked, and perform face recognition verification on the second depth image information processing; and

A vehicle start control module, the vehicle start control module is used to control whether the vehicle starts according to the face recognition verification processing result.

As a third invention of the present invention, a storage medium for implementing the above keyless vehicle entry and starting method has a program stored thereon, and the following steps are implemented when the program is executed by a processor:

Acquiring the first depth image information of the person close to the car door, and performing face detection processing on the first depth image information;

performing live body detection processing on the first depth image information according to the face detection processing result, and performing identity verification processing on the first depth image information according to the live body detection processing result;

Control whether the door is unlocked according to the identity verification processing result;

After the door is unlocked, acquiring the second depth image information of the driver seat of the vehicle, and performing face recognition verification processing on the second depth image information; and

Control whether the vehicle starts according to the face recognition verification processing result.

Owing to adopting above technical scheme, the beneficial effect of the present invention is:

1. The present invention collects the user's biological information, verifies the user's biological characteristics and authorizes the user to operate the vehicle, so as to realize keyless entry and start the car.

2. Compared with the traditional use of monocular or binocular sensors to collect faces for liveness detection, the present invention can obtain depth maps and IR maps through a single frame based on TOF sensing technology, with a small amount of calculation and high real-time performance; at the same time, this The invention adopts an active light source, which is less disturbed by ambient light and can be used under complex ambient light conditions, has lower requirements on hardware, and is stable and reliable.

3. Compared with the current keyless entry system scheme, the present invention does not require the user to carry a physical key, and has a higher degree of intelligence, and can effectively prevent the loss of the key or the inability to enter the vehicle without the key and the property when the key is stolen loss.

4. Compared with the keyless entry system solutions based on fingerprint and other touch sensing technologies, the present invention has dual authentication of living body identification and face ID, which is more secure and does not require cooperation and waiting of the user.

5. Compared with the keyless entry system scheme of cloud identity verification, the present invention provides a local offline identity information database, which can be set up offline, and all calculations are completed locally without uploading to the cloud, ensuring user privacy and security, and more real-time good.

6. Compared with the traditional keyless entry system, the present invention is based on the face ID method, which requires less hardware, low power consumption, low cost, and low system complexity, which brings a lower failure rate , which can bring a better experience to users.

7. Compared with the traditional vehicle start logic, the present invention proposes a new vehicle start authorization verification logic. Unauthorized personnel cannot start the vehicle engine even if the vehicle is unlocked, which can effectively prevent children and other personnel from starting the vehicle by mistake, and further Improved safety performance.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present invention. Those skilled in the art can also obtain other drawings based on these drawings without creative work.

Fig. 1 is a flow chart of the vehicle keyless entry and starting method of the present invention.

Fig. 2 is a flowchart of an application embodiment of the vehicle keyless entry and start method of the present invention.

Fig. 3 is a flow chart of an application embodiment of living body detection in the vehicle keyless entry and starting method of the present invention.

Fig. 4 is a structural schematic diagram of the vehicle keyless entry and starting device of the present invention.

Detailed ways

In order to make the technical means, creative features, goals and effects achieved by the present invention easy to understand, the present invention will be further described below in conjunction with specific illustrations.

Referring to Figure 1, the figure shows a method for keyless entry and start of a vehicle, including the following steps:

Step S10, acquiring the first depth image information of the person approaching the car door, and performing face detection processing on the first depth image information;

Step S20, performing live body detection processing on the first depth image information according to the face detection processing result, and performing identity verification processing on the first depth image information according to the live body detection processing result;

Step S30, controlling whether the door is unlocked according to the identity verification processing result;

Step S40, after the door is unlocked, obtain the second depth image information of the driver's seat of the vehicle, and perform face recognition verification processing on the second depth image information;

Step S50, control whether the vehicle starts according to the result of the face recognition verification process.

Referring to Fig. 1 and in conjunction with Fig. 2, in step S10, obtaining the first depth image information of the person close to the car door includes the following steps:

Step S11, when the vehicle is in the parking lock state, the first TOF camera (image acquisition unit shown in Figure 2) installed at the door is used to collect depth image information outside the door at a standby acquisition frame rate. Among them, the standby acquisition frame rate can be adjusted adaptively according to the acquisition needs;

Step S12, detecting and processing the collected depth image information to determine whether there is a person approaching the car door;

Step S13, if it is detected that there is a person approaching the door, the first TOF camera will use the working acquisition frame rate to perform image acquisition processing on the person approaching the door, so as to obtain the first depth image information of the person approaching the door, wherein the first depth image information Including depth images and near-infrared images. Among them, the work collection frame rate can be adaptively adjusted according to the collection needs. The imaging principle of the near-infrared image is that the first TOF camera actively emits light with a wavelength of 850nm or 940nm, and processes it into a near-infrared image according to the intensity of the reflected light.

Referring to Fig. 1 and in conjunction with Fig. 2, in step S10, face detection processing is carried out to the first depth image information, comprises the following steps:

Step S14, acquiring the near-infrared image in the first depth image information, and sending the near-infrared image to the face detection module for face detection processing;

Step S15, judging whether there is a human face in the near-infrared image according to the result of the human face detection processing, if a human face is detected, then proceed to step S20 for living body detection processing and identity verification processing, if no human face is detected, then proceed to step S16;

Step S16, return to step S13, continue to use the first TOF camera to collect images of people close to the car door at the work collection frame rate and perform face detection processing again, if it is detected that the person leaves, adjust the collection frame rate of the first TOF camera Capture frame rate for standby.

Referring to Fig. 3 and in conjunction with Fig. 1 and Fig. 2, in step S20, according to face detection processing result, carry out living body detection processing to first depth image information, comprise the following steps:

Step S211, if the result of the face detection process is that a face is detected, perform image adjustment processing on the depth image and the near-infrared image in the first depth image information;

Step S212, performing gradient image processing on the depth image after image adjustment processing, obtaining gradient images in the horizontal direction, vertical direction, and center-to-edge direction of the depth image, and merging the three gradient images into a three-channel depth image D_IMG;

Step S213, combining the depth image and the near-infrared image after image adjustment processing to obtain a three-channel grayscale image I_IMG;

Step S214, sending the three-channel depth image D_IMG into the depth image-based living body detection model D_MODEL to calculate the depth image model score;

Step S215, if the score of the depth image model is higher than the first living body threshold S_D, which is preset according to the model requirements, then send the three-channel grayscale image I_IMG to the living body detection model I_MODEL based on near-infrared images to calculate the grayscale image model score;

Step S216, if the grayscale image model score is higher than the second in vivo threshold S_I, which is preset according to the model requirements, then the three-channel depth image D_IMG and the three-channel grayscale image I_IMG are correspondingly put into the depth image buffer queue In Q_D, the grayscale image cache queue Q_I;

Step S217, repeatedly calculate the model score (repeatedly execute step S24 to step S26) until the queue length of the depth image buffer queue Q_D and the grayscale image buffer queue Q_I is equal to the preset value, in this embodiment, the preset value is 3, of course It can also be set to other values according to the model requirements; then the three-channel depth image D_IMG and the three-channel grayscale image I_IMG are taken out from the depth image cache queue Q_D and the grayscale image cache queue Q_I respectively, and the depth image cache queue Q_D, Grayscale image cache queue Q_I;

Step S218, respectively calculate the LBP-TOP feature (Local Binary Patterns from Three Orthogonal Planes) of the three-channel depth image D_IMG and three-channel grayscale image I_IMG taken out, and calculate the three-channel depth image D_IMG and three-channel grayscale image The LBP-TOP features of I_IMG are spliced and put into T_MODEL, a time-series living body detection model based on comprehensive features, to calculate the time-series model score;

In step S219, if the score of the time series model is greater than the third living body threshold, it is determined as a living body; otherwise, it fails the living body detection, and returns to step S217.

In step S211, image adjustment processing is performed on the depth image and the near-infrared image in the first depth image information, including the following steps:

Step S2111, respectively performing correction and adjustment processing on the depth image and the near-infrared image;

Step S2112, performing filtering processing on the depth image and the near-infrared image after the correction adjustment processing;

Step S2113, perform dynamic compression processing on the filtered depth image and near-infrared image respectively according to pixel values, and in this embodiment, dynamically compress to 8 bits. Set the working condition distance to the depth range of interest according to the requirements, increase the depth value within this distance range, improve the contrast, and compress the depth value outside this distance. Find the maximum and minimum value coordinates of the pixel value corresponding to the near-infrared image according to the depth value pixel coordinates within the working condition range, and dynamically adjust the near-infrared image according to the coordinates to find the maximum and minimum values, and compress it to 8 bits.

Step S2114, performing gray value normalization processing on the depth image and the near-infrared image after the dynamic compression processing respectively.

In step S20, the identity verification process is performed on the first depth image information according to the living body detection processing result, including the following steps:

Step S221, if the living body detection processing result is determined to be a living body, then acquire the near-infrared image in the first depth image information;

In step S221, compare the near-infrared image with the local identity database authorized to unlock the vehicle. If the comparison passes, the door is unlocked. If the comparison fails, return to step S10.

In step S40, the second depth image information of the vehicle driver's seat is acquired, and the face recognition verification process is performed on the second depth image information, including:

Step S41, when the door is unlocked, the second TOF camera installed at the driving position of the vehicle is used to collect depth image information outside the driving position of the vehicle at a standby acquisition frame rate;

Step S42, detecting and processing the collected depth image information to determine whether there is a person sitting on the driver's seat of the vehicle;

Step S43, if it is detected that there is a person sitting on the driver's seat of the vehicle, the second TOF camera is used to collect images of the person in the driver's seat of the vehicle at the working acquisition frame rate, so as to obtain the second depth image information of the person in the driver's seat of the vehicle; If it is not detected that there is a person sitting on the driver's seat of the vehicle, then return to step S41;

Step S44, sending the near-infrared image in the second depth image information to the face detection module for face detection processing;

Step S45, judging whether there is a human face according to the human face detection processing result, if a human face is detected, then enter step S46, if no human face is detected, then return to step S41;

Step S46, compare the near-infrared image with the local identity database authorized to start the vehicle; if the comparison is passed, the vehicle is authorized to start; if the comparison is not passed, return to step S41.

In view of the inconvenience and potential safety hazards of the traditional keyless entry system, the present invention collects the biological characteristics of the user by using the depth sensor of TOF technology, performs liveness detection and ID identification verification on the user, and can enter the vehicle after passing the verification. If you pass the ID verification of the driver's seat, you can be authorized to start the vehicle. In addition, the present invention provides users with a local identity information database, and the owner can selectively enter the identity database authorized to enter the vehicle and the identity database authorized to start the vehicle, which provides convenience for family members to enter the vehicle while ensuring that family members who are not authorized to start the vehicle such as The child started the vehicle by mistake and caused a safety accident.

Referring to FIG. 2 , the figure shows a vehicle keyless entry and starting device, including a first TOF camera 100, a second TOF camera 200, a first depth image information processing module 300, a living body detection and identity verification module 400, The door lock control module 500 , the second depth image information processing module 600 and the vehicle start control module 700 .

The first TOF camera 100 is installed at the car door, and is used to collect first depth image information of people approaching the car door.

The second TOF camera 200 is installed at the driving seat of the vehicle, and is used for collecting second depth image information of the person in the driving seat of the vehicle.

The first depth image information processing module 300 is used to acquire the first depth image information of the person approaching the car door, and perform face detection processing on the first depth image information.

The living body detection and identity verification module 400 is configured to perform life detection processing on the first depth image information according to the face detection processing result, and perform identity verification processing on the first depth image information according to the living body detection processing result.

The car door lock control module 500 is used for controlling whether the car door is unlocked according to the identity verification processing result.

The second depth image information processing module 600 is used to obtain the second depth image information of the driver seat of the vehicle after the door is unlocked, and perform face recognition verification processing on the second depth image information.

The vehicle start control module 700 is used to control whether the vehicle starts according to the face recognition verification processing result.

Each module in the vehicle keyless entry and starting device of the present invention can be fully or partially realized by software, hardware and combinations thereof. The above modules can be embedded in or independent of the processor in the vehicle control system in the form of hardware, and can also be stored in the memory of the vehicle control system in the form of software, so that the processor can call and execute the corresponding operations of the above modules.

The present invention also provides a storage medium for realizing the above keyless vehicle entry and starting method, on which a program is stored, and when the program is executed by a processor, the following steps are implemented:

Step S10, acquiring the first depth image information of the person approaching the car door, and performing face detection processing on the first depth image information;

Step S20, performing live body detection processing on the first depth image information according to the face detection processing result, and performing identity verification processing on the first depth image information according to the live body detection processing result;

Step S30, controlling whether the door is unlocked according to the identity verification processing result;

Step S40, after the door is unlocked, obtain the second depth image information of the driver's seat of the vehicle, and perform face recognition verification processing on the second depth image information;

Step S50, control whether the vehicle starts according to the result of the face recognition verification process.

Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above-mentioned embodiments can be realized by instructing related hardware through a program, and the program can be stored in a non-volatile computer-readable storage medium , when the program is executed, it may include the procedures of the embodiments of the above-mentioned methods. Wherein, any references to memory, storage, database or other media used in the various embodiments provided in the present application may include non-volatile and/or volatile memory. Nonvolatile memory can include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Chain Synchlink DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

The basic principles and main features of the present invention and the advantages of the present invention have been shown and described above. Those skilled in the industry should understand that the present invention is not limited by the above-mentioned embodiments. What are described in the above-mentioned embodiments and the description only illustrate the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will also have Variations and improvements are possible, which fall within the scope of the claimed invention. The protection scope of the present invention is defined by the appended claims and their equivalents.

Claims (10)

1. A vehicle keyless entry and start method, characterized in that, comprising: Acquiring the first depth image information of the person close to the car door, and performing face detection processing on the first depth image information; performing live body detection processing on the first depth image information according to the face detection processing result, and performing identity verification processing on the first depth image information according to the live body detection processing result; Control whether the door is unlocked according to the identity verification processing result; After the door is unlocked, acquiring the second depth image information of the driver seat of the vehicle, and performing face recognition verification processing on the second depth image information; and Control whether the vehicle starts according to the face recognition verification processing result. 2. The method for keyless entry and start of a vehicle according to claim 1, wherein said acquiring the first depth image information of a person close to the vehicle door comprises: When the vehicle is in the state of parking and locking the door, the depth image information outside the door is collected by the first TOF camera installed at the door at a standby acquisition frame rate; Detecting and processing the collected depth image information to determine whether there is a person approaching the car door; and If it is detected that there is a person approaching the door, the first TOF camera will perform image acquisition processing on the person approaching the door at the work acquisition frame rate, so as to obtain the first depth image information of the person approaching the door, and the first depth image information includes Depth images and near-infrared images. 3. The method for keyless entry and start of a vehicle according to claim 2, wherein the first depth image information includes a depth image and a near-infrared image. 4. The method for keyless entry and start of a vehicle according to claim 2, wherein said performing face detection processing on said first depth image information comprises: Obtain a near-infrared image in the first depth image information, and send the near-infrared image to a face detection module for face detection processing; and Judging whether there is a human face in the near-infrared image according to the result of human face detection processing, if a human face is detected, then perform living body detection processing and identity verification processing, if no human face is detected, continue to pass through the first TOF camera At the working acquisition frame rate, image acquisition is performed on the person approaching the car door and the face detection process is performed again. If it is detected that the person leaves, the acquisition frame rate of the first TOF camera is adjusted to the standby acquisition frame rate. 5. The vehicle keyless entry and start method according to any one of claims 1 to 4, characterized in that, performing the living body detection processing on the first depth image information according to the face detection processing result, comprising: If the face detection processing result is that a human face is detected, image adjustment processing is performed on the depth image and the near-infrared image in the first depth image information respectively; Gradient image processing is performed on the depth image after image adjustment processing to obtain gradient images in the horizontal direction, vertical direction, and center-to-edge direction of the depth image, and merge the three gradient images into a three-channel depth image; Combine the depth image and near-infrared image after image adjustment to obtain a three-channel grayscale image; The three-channel depth image is sent into a depth image-based living body detection model to calculate the depth image model score; If the depth image model score is higher than the first living body threshold, the three-channel grayscale image is sent to a living body detection model based on near-infrared images to calculate the grayscale image model score; If the grayscale image model score is higher than the second in vivo threshold, the three-channel depth image and the three-channel grayscale image are correspondingly put into a depth image cache queue and a grayscale image cache queue; Repeat the calculation of the model score until the queue lengths of the depth image cache queue and the grayscale image cache queue are equal to the preset values, and the three-channel depth image and the three-channel grayscale image respectively taken out from the depth image cache queue and the grayscale image cache queue are degree image, and empty the depth image cache queue and the grayscale image cache queue respectively; Calculate the LBP-TOP features of the extracted three-channel depth image and three-channel grayscale image respectively, and splicing the calculated LBP-TOP features of the three-channel depth image and three-channel grayscale image into the time-series living body based on comprehensive features Computing time-series model scores within the detection model; and If the score of the time series model is greater than the third living body threshold, it is determined as a living body. 6. The method for keyless entry and start of a vehicle according to claim 5, wherein the image adjustment processing of the depth image and the near-infrared image in the first depth image information includes: Performing correction and adjustment processing on the depth image and the near-infrared image respectively; Perform filtering processing on the depth image and the near-infrared image processed by the orthodontic adjustment respectively; performing dynamic compression processing on the filtered depth image and the near-infrared image respectively; and Normalize the gray value of the depth image and the near-infrared image after dynamic compression processing respectively. 7. The method for keyless entry and start of a vehicle according to claim 5, wherein said performing identity verification processing on said first depth image information according to said living body detection processing result comprises: If the living body detection processing result is identified as a living body, acquiring a near-infrared image in the first depth image information; and The near-infrared image is compared with the local identity library authorized to unlock the vehicle, and if the comparison is passed, the door is unlocked. 8. The method for keyless entry and start of a vehicle according to claim 1, wherein the second depth image information of the vehicle driver is acquired, and the face recognition verification process is performed on the second depth image information ,include: When the door is unlocked, the second TOF camera installed at the driving position of the vehicle is used to collect the depth image information outside the driving position of the vehicle at a standby acquisition frame rate; Detect and process the collected depth image information to determine whether there is a person sitting in the driver's seat of the vehicle; If it is detected that there is a person sitting on the driver's seat of the vehicle, the second TOF camera is used to collect images of the person in the driver's seat of the vehicle at a work acquisition frame rate, so as to obtain the second depth image information of the person in the driver's seat of the vehicle; Sending the near-infrared image in the second depth image information to the face detection module for face detection processing; and Whether there is a human face is judged according to the result of the face detection process. If a human face is detected, the near-infrared image is compared with the local identity database authorized to start the vehicle. If the comparison is passed, the vehicle is authorized to start. 9. A vehicle keyless entry and start device, characterized in that it comprises: The first TOF camera, the first TOF camera is installed at the car door, and is used to collect the first depth image information of the person close to the car door; A second TOF camera, the second TOF camera is installed at the driver's seat of the vehicle, and is used to collect the second depth image information of the person in the driver's seat of the vehicle; A first depth image information processing module, the first depth image information processing module is used to obtain the first depth image information of the person close to the car door, and perform face detection processing on the first depth image information; A live body detection and identity verification module, the live body detection and identity verification module is used to perform live body detection processing on the first depth image information according to the face detection processing result, and to perform live body detection processing on the first depth image information according to the live body detection processing result. Image information for identity verification processing; A door lock control module, the door lock control module is used to control whether the door is unlocked according to the identity verification processing result; The second depth image information processing module, the second depth image information processing module is used to obtain the second depth image information of the driver's seat of the vehicle after the door is unlocked, and perform face recognition verification on the second depth image information processing; and A vehicle start control module, the vehicle start control module is used to control whether the vehicle starts according to the face recognition verification processing result. 10. A storage medium, on which a program is stored, wherein when the program is executed by a processor, the steps in the method for keyless entry and start of a vehicle according to any one of claims 1 to 8 are realized.

Images