Method of controlling a vehicle based on iris recognition and vehicle system

Abstract

The application discloses a method for controlling a vehicle based on iris recognition and a vehicle system, and belongs to the fields of vehicle technology, intelligent vehicles and vehicle networks. The application provides a method for unlocking a vehicle door outside a vehicle and starting the vehicle inside the vehicle based on iris recognition. The method performs identity authentication on a user who wants to perform a driving operation through double cloud verification, guarantees data security and vehicle safety, solves the vehicle use demand when the user does not carry a key or a mobile phone, and guarantees the driving experience of the user.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicles, in particular to a method for controlling a vehicle based on iris recognition and a vehicle system. BACKGROUND

[0002] With the development of the times and the development of intelligent vehicles, more and more diverse vehicles appear in front of us. At present, most vehicles on the market enter by using a remote key, an NFC card key or a mobile phone Bluetooth key to open the door, and the vehicle is started by detecting the key in the vehicle. These methods are not suitable for scenarios where the key or mobile phone is not carried. In order to provide customers with a good riding experience, some technologies also propose convenient methods such as fingerprint recognition and face recognition to authenticate fingerprint or face information. The vehicle owner only needs to authenticate the fingerprint or bring the face close to the mobile phone camera and pass the authentication to realize the opening of the vehicle door, and other people cannot open the door. These technologies improve the convenience of vehicle use, but there are certain safety risks. Fingerprint is easy to copy, and face recognition has the risk of being stolen by others. SUMMARY

[0003] The embodiments of the present application provide a method for controlling a vehicle based on iris recognition and a vehicle system, which can ensure data security and vehicle safety while solving the problem that the user does not carry a key or a mobile phone and still needs to use the vehicle. The technical solution is as follows:

[0004] In one aspect, a method for controlling a vehicle based on iris recognition is provided, the vehicle comprising an outside recognition module, an inside recognition module, an iris processing module, and a vehicle body controller, the method comprising: a high-precision camera in the outside recognition module capturing an eye image of an outside object, and transmitting the eye image to the iris processing module, the iris processing module extracting a first iris image from the eye image of the outside object, performing feature extraction on the first iris image to obtain a first iris feature, and uploading the first iris feature to a cloud through a vehicle local area network; the vehicle body controller receiving a first comparison result fed back by the cloud, the first comparison result being used to indicate whether the first iris feature is consistent with an iris feature of a driver of the vehicle in a database of the cloud, and if the first comparison result indicates that the outside object is consistent with the iris feature of the driver, unlocking a door of the vehicle; in a case where the door is unlocked, a high-precision camera in the inside recognition module capturing an eye image of an inside object, and transmitting the eye image to the iris processing module, the iris processing module extracting a second iris image from the eye image of the inside object, performing feature extraction on the second iris image to obtain a second iris feature, and uploading the second iris feature to the cloud through the vehicle local area network; and the vehicle body controller receiving a second comparison result fed back by the cloud, the second comparison result being used to indicate whether the second iris feature is consistent with the iris feature of the driver of the vehicle in the database of the cloud, and if the second comparison result indicates that the inside object is consistent with the iris feature of the driver, starting the vehicle.

[0005] Optionally, before the feature extraction on the iris image, the method further comprises: the iris processing module performing iris positioning, iris image normalization, and image enhancement processing on the iris image.

[0006] Optionally, the method further comprises: the vehicle body controller, upon receiving a remote start instruction, adjusting a rotation of the high-precision camera in the outside recognition module to perform pedestrian detection, and if a pedestrian is detected at any angle of rotation and the pedestrian performs a detection trigger action, performing image capture on a head of the pedestrian.

[0007] Optionally, the method further comprises: the vehicle body controller, upon detecting that a driver-side door is opened, adjusting a rotation of the high-precision camera in the inside recognition module to perform image capture towards a target position of a driver seat to obtain the eye image of the inside object.

[0008] Optionally, the adjusting of the rotation of the high-precision camera in the outside recognition module to perform pedestrian detection comprises: the vehicle body controller adjusting the rotation of the high-precision camera in the outside recognition module to a direction of a source of the remote start instruction to perform pedestrian detection.

[0009] In one aspect, a vehicle system is provided, the vehicle system comprising an outside recognition module, an inside recognition module, an iris processing module, and a vehicle body controller:

[0010] The high-precision camera in the outside-vehicle recognition module is configured to collect an eye image of an outside-vehicle object and transmit the eye image to the iris processing module.

[0011] The iris processing module is configured to extract a first iris image from the eye image of the outside-vehicle object, perform feature extraction on the first iris image to obtain a first iris feature, and upload the first iris feature to the cloud via the vehicle local area network.

[0012] The vehicle body controller is configured to receive a first comparison result fed back by the cloud, the first comparison result being configured to indicate whether the first iris feature is consistent with an iris feature of a driver of the vehicle in a cloud database, and if the first comparison result indicates that the outside-vehicle object is consistent with the iris feature of the driver, unlock a door of the vehicle.

[0013] In the case that the door is unlocked, the high-precision camera in the in-vehicle recognition module is configured to collect an eye image of an in-vehicle object and transmit the eye image to the iris processing module.

[0014] The iris processing module is further configured to extract a second iris image from the eye image of the in-vehicle object, perform feature extraction on the second iris image to obtain a second iris feature, and upload the second iris feature to the cloud via the vehicle local area network.

[0015] The vehicle body controller is further configured to receive a second comparison result fed back by the cloud, the second comparison result being configured to indicate whether the second iris feature is consistent with the iris feature of the driver of the vehicle in the cloud database, and if the second comparison result indicates that the in-vehicle object is consistent with the iris feature of the driver, start the vehicle.

[0016] Optionally, the iris processing module is further configured to perform iris positioning, iris image normalization, and image enhancement processing on the iris image.

[0017] Optionally, the vehicle body controller is further configured to, when detecting that a driver-side door is opened, adjust the high-precision camera in the in-vehicle recognition module to rotate to a target position of a driver seat for image collection.

[0018] Optionally, the vehicle body controller is further configured to, when receiving a remote start instruction, adjust the high-precision camera in the outside-vehicle recognition module to rotate to perform pedestrian detection, and if a pedestrian is detected at any angle of rotation and the pedestrian performs a detection trigger action, collect an image of a head of the pedestrian.

[0019] Optionally, the vehicle body controller is configured to, based on a source direction of the remote start instruction, adjust the high-precision camera in the outside-vehicle recognition module to rotate to the source direction for pedestrian detection.

[0020] It should be understood that the foregoing general description and the following detailed description are only exemplary and do not limit the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 is a basic structure schematic diagram of a vehicle system provided by an embodiment of the present application;

[0022] Figure 2 is a flowchart of a method for controlling a vehicle based on iris recognition provided by an embodiment of the present application;

[0023] Figure 3 is a structure schematic diagram of a vehicle system provided by an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to make the technical solutions and advantages of the present application clearer, the embodiments of the present application are described in further detail below.

[0025] The terms "first", "second", "third", and "fourth" and the like in the specification and claims of the present application and the drawings are used to distinguish different objects, and are not used to describe a particular order. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but can optionally include steps or units not listed, or can optionally include other steps or units inherent to the process, method, product or device.

[0026] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data for analysis, stored data, displayed data, etc.) and signals involved in the present application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of related data need to comply with relevant laws, regulations and standards of relevant countries and regions. For example, the data and the like involved in the present application are obtained under sufficient authorization.

[0027] Figure 1is a basic structure schematic diagram of a vehicle system provided by an embodiment of the present application. The vehicle system comprises: an outside identification module 101, an inside identification module 102, an iris processing module 103, and a vehicle body controller 104. The outside identification module 101 comprises a high-precision camera arranged outside the vehicle, the inside identification module 102 comprises a high-precision camera arranged inside the vehicle, the outside identification module 101 and the inside identification module 102 are connected with the iris processing module 103 respectively, and the vehicle body controller 104 can control the rotation angle of the outside identification module 101 and the inside identification module 102, so as to realize the adjustment of the shooting direction of the camera. In addition, the vehicle system further comprises a processor and a communication component, the communication component is used to upload the data of the vehicle system to the cloud for corresponding data processing, and the vehicle controller is instructed to be issued to achieve the purpose of controlling the vehicle. The high-precision camera can be a DMS (Driver Monitor System) camera.

[0028] Figure 2 is a flowchart of a method for controlling a vehicle based on iris recognition provided by an embodiment of the present application, which can be implemented based on the vehicle system of the above Figure 1 , and the method comprises the following steps.

[0029] 201. The high-precision camera in the outside identification module in the vehicle system collects the eye image of the outside object and transmits it to the iris processing module.

[0030] In some embodiments, when the high-precision camera in the outside identification module collects the eye image of the outside object, it can first collect the face image of the outside object, and then extract the eye image from the face image. In other embodiments, the high-precision camera in the outside identification module first locates the eye region from the head of the outside object, and then collects the eye region to obtain the eye image. The present application does not make specific limitation on how to realize the above collection process.

[0031] In some embodiments, the method further comprises: the vehicle body controller adjusts the rotation of the high-precision camera in the outside identification module to perform pedestrian detection after receiving the remote start instruction, and if a pedestrian is detected at any angle and the pedestrian performs a detection trigger action, the head of the pedestrian is collected.

[0032] In remote start scenarios, users can trigger vehicle start by operating a remote control. The remote control responds to the user's start operation by sending a remote start command to the vehicle. Upon receiving the command, the vehicle controller automatically adjusts the camera to actively detect and determine if the person approaching is the driver, significantly improving detection timeliness. Of course, to avoid collecting unauthorized data, it's necessary to detect whether the pedestrian has performed a trigger action before image acquisition, such as a specific gesture or body movement.

[0033] In some embodiments, adjusting the rotation of the high-precision camera in the vehicle exterior recognition module for pedestrian detection includes: the vehicle body controller adjusting the rotation of the high-precision camera in the vehicle exterior recognition module towards the direction of the remote start command to perform pedestrian detection. For the remote start command, the vehicle body controller can adjust the camera rotation based on the direction of the signal's origin, thereby achieving accurate location of approaching pedestrians and greatly improving detection timeliness.

[0034] 202. The iris processing module extracts the first iris image from the eye image of the object outside the vehicle, performs feature extraction on the first iris image to obtain the first iris feature, and uploads it to the cloud via the vehicle local area network.

[0035] Optionally, before feature extraction of the iris image, the method further includes: the iris processing module performing iris localization, iris image normalization, and image enhancement processing on the iris image. Through the above image processing, the recognition rate of iris information in the image can be improved.

[0036] The aforementioned iris processing module employs a two-dimensional Gabor wavelet algorithm to extract the feature points required for iris recognition from the iris image. During processing, the iris processing module describes the iris using the real and imaginary parts of a two-dimensional Gabor filter; that is, the feature extraction process includes the following steps:

[0037] 202A. The iris processing module estimates the center and radius of the iris and pupil, which is to say, iris localization is achieved.

[0038] This iris localization process uses a coarse-to-fine strategy, ultimately achieving single-pixel accuracy. Generally, the center of the pupil and the iris are not concentric; the pupil center is slightly less important than the iris center, and its radius is 0.1 to 0.8 times the iris radius. Therefore, the three parameters determining the pupil's circumference are estimated separately from the iris's circumference.

[0039] 202B. The iris processing module normalizes the iris localization results, and then filters the normalized iris texture using a two-dimensional Gabor filter. The filtering formula is as follows:

[0040]

[0041] where (x0, y0) represents the position in the image, (a, b) is the effective width and length of the filter, (u0, v0) determines the frequency of the modulation term and the direction (q0 = arctan(v0 / u0)).

[0042] It should be noted that the polarity of the filtered result is quantized to 2-bit binary, when the real part and the imaginary part are both positive, the quantization value is 11; when the real part is positive and the imaginary part is negative, the quantization value is 10; when the real part is negative and the imaginary part is positive, the quantization value is 01; and when the real part and the imaginary part are both negative, the quantization value is 00, thereby obtaining the iris feature.

[0043] In any of the above stages, the stage result can be subjected to image enhancement processing to obtain image data containing more accurate information, so as to achieve the purpose of accurately extracting the iris feature.

[0044] 203、The cloud server compares the first iris feature with the iris feature of the driver of the vehicle in the cloud database after receiving the first iris feature, and returns a first comparison result to the vehicle, which indicates whether the first iris feature is consistent with the iris feature of the driver of the vehicle in the cloud database.

[0045] For the cloud server, it maintains a cloud database, which is used to store the iris features and other information of authorized drivers, so as to provide cloud verification services.

[0046] Among them, the vehicle will also carry a vehicle identification, such as a license plate number, when sending the first iris feature, so that the cloud server can obtain the iris feature corresponding to the vehicle identification based on the correspondence between the vehicle identification and the iris feature stored in the cloud database, and compare it to verify whether the received iris feature is the iris feature of the registered or legal driver of the vehicle.

[0047] If it is determined through comparison that the first iris feature is consistent with the iris feature of the driver of the vehicle in the cloud database, a first comparison result indicating that the two are consistent is returned, and if it is determined through comparison that the first iris feature is inconsistent with the iris feature of the driver of the vehicle in the cloud database, a first comparison result indicating that the two are inconsistent is returned.

[0048] 204、The vehicle body controller receives the first comparison result fed back by the cloud, and if the first comparison result indicates that the extravehicular object is consistent with the iris feature of the driver, the vehicle door is unlocked.

[0049] By verifying whether the object outside the vehicle is a registered or legal driver of the vehicle to determine whether to unlock, the unlocking operation for the driver can be performed in advance to the greatest extent while ensuring the safety of the vehicle, thereby ensuring the driving experience of the user.

[0050] 205、In the case that the vehicle door is unlocked, the high-precision camera in the in-vehicle identification module collects an eye image of the in-vehicle object and transmits the eye image to the iris processing module.

[0051] In some embodiments, when the high-precision camera in the in-vehicle identification module collects the eye image of the in-vehicle object, the high-precision camera can first collect a face image of the in-vehicle object and then extract the eye image from the face image. In other embodiments, the high-precision camera in the in-vehicle identification module first locates an eye region from the head of the in-vehicle object and then collects the eye region to obtain the eye image. The present application does not make a specific limitation on how to implement the above collection process.

[0052] Optionally, the method further includes: the vehicle body controller adjusts the rotation of the high-precision camera in the in-vehicle identification module to perform image collection toward the target position of the driving seat to obtain the eye image of the in-vehicle object when detecting that the driving side door is opened.

[0053] For the remote start scenario, the user can trigger the vehicle start by operating the remote controller. The remote controller sends a remote start instruction to the vehicle in response to the start operation of the user. The vehicle door is unlocked only after the first verification is passed. After detecting that the vehicle door is opened, the image on the driving seat is collected again by the high-precision camera arranged in the vehicle to determine whether the person is the driver, thereby greatly improving the detection timeliness. Since this time is the identity verification of the person on the driving seat, the safety can also be improved. Of course, in order to avoid collecting unauthorized data, the personnel on the driving seat can be detected to perform a detection trigger action, such as performing a specific gesture or a specific body movement, etc. before image collection.

[0054] In some embodiments, the adjustment of the rotation of the high-precision camera in the in-vehicle identification module to perform image collection includes: the vehicle body controller adjusts the rotation of the high-precision camera in the in-vehicle identification module based on the setting state of the seat of the driving seat to enable the camera to accurately collect the face image of the person on the driving seat. The setting state of the seat includes the front and back positions of the seat and the angle of the seat backrest, etc. which enables the driver to drive the vehicle at the most suitable distance and sitting posture. It should be noted that the rotation angle can be obtained based on a table, that is, a table for maintaining the mapping relationship between the setting state of the seat and the camera angle is maintained, so that after obtaining the setting state of the seat of the driving seat, the camera angle is obtained from the table, the high-precision camera is controlled to rotate to the obtained camera angle, thereby achieving the purpose of accurately collecting the image.

[0055] 206、The iris processing module extracts a second iris image from the eye image of the in-vehicle object, performs feature extraction on the second iris image to obtain a second iris feature, and uploads the second iris feature to the cloud through the vehicle local area network.

[0056] 207、After receiving the second iris feature, the cloud server compares the second iris feature with the iris feature of the driver of the vehicle in the cloud database, and returns a second comparison result to the vehicle, the second comparison result indicating whether the second iris feature is consistent with the iris feature of the driver of the vehicle in the cloud database.

[0057] The processes of steps 206 to 207 are the same as steps 202 to 203, and will not be repeated here.

[0058] 208、The vehicle body controller receives the second comparison result fed back by the cloud, and if the second comparison result indicates that the iris feature of the in-vehicle object is consistent with that of the driver, the vehicle is started.

[0059] The application provides a method for unlocking the vehicle door from outside and starting the vehicle from inside based on iris recognition, which verifies the identity of the user who wants to perform the driving operation through double cloud verification, ensures the data security and vehicle safety, solves the vehicle demand when the user does not carry the key or mobile phone, and ensures the driving experience of the user.

[0060] Figure 3 is a structural schematic diagram of a vehicle system provided by an embodiment of the application. The vehicle system includes an external identification module 301, an internal identification module 302, an iris processing module 303, and a vehicle body controller 304.

[0061] The high-precision camera in the external identification module is used to collect the eye image of the external object and transmit it to the iris processing module.

[0062] The iris processing module is used to extract a first iris image from the eye image of the external object, perform feature extraction on the first iris image to obtain a first iris feature, and upload the first iris feature to the cloud through the vehicle local area network.

[0063] The vehicle body controller is used to receive the first comparison result fed back by the cloud, the first comparison result indicating whether the first iris feature is consistent with the iris feature of the driver of the vehicle in the cloud database, and if the first comparison result indicates that the iris feature of the external object is consistent with that of the driver, the vehicle door is unlocked.

[0064] In the case of unlocking the vehicle door, the high-precision camera in the internal identification module is used to collect the eye image of the in-vehicle object and transmit it to the iris processing module.

[0065] The iris processing module is further configured to extract a second iris image from the eye image of the in-vehicle object, perform feature extraction on the second iris image to obtain a second iris feature, and upload the second iris feature to the cloud through the vehicle local area network;

[0066] The vehicle body controller is further configured to receive a second comparison result fed back by the cloud, the second comparison result being used to indicate whether the second iris feature is consistent with an iris feature of a driver of the vehicle in a database of the cloud, and start the vehicle if the second comparison result indicates that the in-vehicle object is consistent with the iris feature of the driver.

[0067] In some embodiments, the iris processing module is further configured to perform iris positioning, iris image normalization, and image enhancement processing on the iris image.

[0068] In some embodiments, the vehicle body controller is further configured to, when detecting that the driver side door is opened, adjust the rotation of the high-precision camera in the in-vehicle identification module to image collection towards the target position of the driver's seat.

[0069] In some embodiments, the vehicle body controller is further configured to, when receiving a remote start instruction, adjust the rotation of the high-precision camera in the out-of-vehicle identification module to perform pedestrian detection, and if a pedestrian is detected at any angle of rotation and the pedestrian performs a detection trigger action, image collection is performed on the head of the pedestrian.

[0070] In some embodiments, the vehicle body controller is configured to, based on a source direction of the remote start instruction, adjust the rotation of the high-precision camera in the out-of-vehicle identification module to the source direction to perform pedestrian detection.

[0071] The application provides a control system for unlocking a vehicle door from outside and starting a vehicle from inside based on iris recognition, which verifies the identity of a user who wants to perform a driving operation through double cloud verification, ensures data security and vehicle safety, solves the vehicle use demand when the user does not carry a key or a mobile phone, and ensures the driving experience of the user.

[0072] The application also provides a vehicle control method, which includes:

[0073] controlling a high-precision camera in an out-of-vehicle identification module to collect an eye image of an out-of-vehicle object and transmit the eye image to an iris processing module;

[0074] extracting a first iris image from the eye image of the out-of-vehicle object through the iris processing module, performing feature extraction on the first iris image to obtain a first iris feature, and uploading the first iris feature to the cloud through a vehicle local area network;

[0075] if a first comparison result of the cloud feedback is received, the first comparison result is used to indicate whether the first iris feature is consistent with the iris feature of the driver of the vehicle in the cloud database, if the first comparison result indicates that the out-of-vehicle object is consistent with the iris feature of the driver, the door of the vehicle is unlocked;

[0076] In the case of unlocking the door, the high-precision camera in the in-vehicle identification module is controlled to collect the eye image of the in-vehicle object and transmit it to the iris processing module;

[0077] The second iris image is extracted from the eye image of the in-vehicle object by the iris processing module, the feature extraction is performed on the second iris image to obtain the second iris feature, and the second iris feature is uploaded to the cloud through the vehicle local area network;

[0078] The second comparison result of the cloud feedback is received, the second comparison result is used to indicate whether the second iris feature is consistent with the iris feature of the driver of the vehicle in the cloud database, if the second comparison result indicates that the in-vehicle object is consistent with the iris feature of the driver, the vehicle is started.

[0079] In some embodiments, the vehicle control method further comprises: performing iris positioning, iris image normalization and image enhancement processing on the iris image.

[0080] In some embodiments, the vehicle control method further comprises: when it is detected that the driver side door is opened, the high-precision camera in the in-vehicle identification module is adjusted to rotate to the target position of the driver's seat for image collection.

[0081] In some embodiments, the vehicle control method further comprises: after receiving the remote start instruction, the high-precision camera in the out-of-vehicle identification module is adjusted to rotate to perform pedestrian detection, if a pedestrian is detected at any angle and the pedestrian performs a detection trigger action, the head of the pedestrian is collected for image collection.

[0082] In some embodiments, the vehicle control method further comprises: based on the source direction of the remote start instruction, the high-precision camera in the out-of-vehicle identification module is adjusted to rotate to the source direction to perform pedestrian detection.

[0083] The application also provides a vehicle control device for executing the above-mentioned vehicle control method.

[0084] Those skilled in the art can understand that all or part of the steps of the above-mentioned embodiments can be completed by hardware, or by program to instruct related hardware, the program can be stored in a computer readable storage medium, the storage medium mentioned above can be read-only memory, disk or optical disk, etc.

[0085] The above merely describes the technical solutions of the present application for the purpose of enabling those skilled in the art to understand the technical solutions of the present application, and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for controlling a vehicle based on iris recognition, characterized in that, The vehicle system includes an external recognition module, an internal recognition module, an iris processing module, and a body controller; the method includes: The high-precision camera in the vehicle exterior recognition module captures eye images of objects outside the vehicle and transmits them to the iris processing module. The iris processing module extracts a first iris image from the eye images of the objects outside the vehicle, performs feature extraction on the first iris image to obtain the first iris feature, and uploads it to the cloud via the vehicle local area network. The vehicle body controller receives a first comparison result from the cloud. The first comparison result is used to indicate whether the first iris feature is consistent with the iris feature of the driver of the vehicle in the cloud database. If the first comparison result indicates that the external object is consistent with the iris feature of the driver, then the vehicle door is unlocked. When the car door is unlocked, the high-precision camera in the in-vehicle recognition module captures an eye image of an object inside the vehicle and transmits it to the iris processing module. The iris processing module extracts a second iris image from the eye image of the object inside the vehicle, performs feature extraction on the second iris image to obtain a second iris feature, and uploads it to the cloud via the vehicle local area network. The vehicle controller receives a second comparison result from the cloud. The second comparison result is used to indicate whether the second iris feature matches the iris feature of the driver of the vehicle in the cloud database. If the second comparison result indicates that the iris feature of the object inside the vehicle matches the iris feature of the driver, then the vehicle is started. The method further includes: Upon receiving a remote start command, the vehicle body controller adjusts the rotation of the high-precision camera in the vehicle exterior recognition module to detect pedestrians. If a pedestrian is detected at any angle and the pedestrian performs a detection trigger action, an image of the pedestrian's head is captured.

2. The method according to claim 1, characterized in that, Before performing feature extraction on the first iris image and the second iris image, the method further includes: The iris processing module performs iris localization, iris image normalization, and image enhancement processing on the first iris image and the second iris image.

3. The method according to claim 1, characterized in that, The method further includes: When the vehicle body controller detects that the driver's side door is opened, it adjusts the rotation of the high-precision camera in the in-vehicle recognition module to capture images of the target position in the driver's seat, thereby obtaining an eye image of the object inside the vehicle.

4. The method according to claim 1, characterized in that, The adjustment of the high-precision camera in the vehicle exterior recognition module for pedestrian detection includes: The vehicle body controller adjusts the high-precision camera in the external recognition module to rotate in the direction of the source of the remote start command, so as to perform pedestrian detection.

5. A vehicle system, characterized in that, The vehicle system includes an external recognition module, an internal recognition module, an iris processing module, and a body controller. The high-precision camera in the vehicle exterior recognition module is used to capture eye images of objects outside the vehicle and transmit them to the iris processing module. The iris processing module is used to extract a first iris image from the eye image of the object outside the vehicle, perform feature extraction on the first iris image to obtain the first iris feature, and upload it to the cloud via the vehicle local area network; The vehicle body controller is used to receive a first comparison result fed back from the cloud. The first comparison result is used to indicate whether the first iris feature is consistent with the iris feature of the driver of the vehicle in the cloud database. If the first comparison result indicates that the external object is consistent with the iris feature of the driver, then the vehicle door is unlocked. When the car door is unlocked, the high-precision camera in the in-vehicle recognition module is used to capture eye images of objects inside the vehicle and transmit them to the iris processing module. The iris processing module is also used to extract a second iris image from the eye image of the object inside the vehicle, extract features from the second iris image to obtain second iris features, and upload them to the cloud via the vehicle local area network; The vehicle controller is also used to receive a second comparison result fed back from the cloud. The second comparison result is used to indicate whether the second iris feature is consistent with the iris feature of the driver of the vehicle in the cloud database. If the second comparison result indicates that the object inside the vehicle is consistent with the iris feature of the driver, then the vehicle is started. The vehicle body controller is also used to adjust the rotation of the high-precision camera in the vehicle exterior recognition module to perform pedestrian detection when a remote start command is received. If a pedestrian is detected when the camera is rotated to any angle and the pedestrian performs a detection trigger action, the image of the pedestrian's head is captured.

6. The vehicle system according to claim 5, characterized in that, The iris processing module is also used to perform iris localization, iris image normalization, and image enhancement processing on the first iris image and the second iris image.

7. The vehicle system according to claim 5, characterized in that, The vehicle body controller is also used to adjust the rotation of the high-precision camera in the in-vehicle recognition module to capture images toward the target position of the driver's seat when the driver's side door is detected to be opened.

8. The vehicle system according to claim 5, characterized in that, The vehicle body controller is also used to adjust the high-precision camera in the vehicle exterior recognition module to rotate in the direction of the source of the remote start command, so as to perform pedestrian detection.

Images