Stereoscopic display electronic rearview mirror system based on eye tracking and working method thereof

Abstract

The invention relates to the technical field of streaming media, in particular to a stereoscopic display electronic rearview mirror system based on eye tracking and a working method of the stereoscopic display electronic rearview mirror system. The system comprises an eye position calculation unit used for calculating eye coordinate parameters of a driver according to a driver seat image; the left and right eye image cutting unit is used for cutting the left eye stereoscopic vision image and the right eye stereoscopic vision image according to the eye coordinate parameters; the left eye and right eye image synthesis unit is used for synthesizing the cut left eye stereoscopic vision image and the cut right eye stereoscopic vision image into a synthesized image; and the three-dimensional imaging calculation unit is used for processing the synthesized image into a three-dimensional display format and outputting a three-dimensional visual picture through a connected three-dimensional display. By recognizing the spatial position of the eyes of the driver, the content captured by the camera outside the vehicle is cut to the rearview mirror picture suitable for the driver to watch, meanwhile, the stereoscopic display effect is achieved, the picture of the rearview mirror has depth information, and the driver can observe the rearview mirror easily and judge the distance conveniently.

Description

Technical Field

[0001] The present invention relates to the technical field of streaming media, and particularly to a three-dimensional display electronic rearview mirror system and its working method. Background Art

[0002] Currently, the existing electronic rearview mirror system consists of a single camera outside the vehicle, a display screen inside the vehicle, and an image processing system. According to different driving habits of drivers, such as differences in driver height, sitting posture, and driving habits, the driver needs to manually adjust the field of view of the electronic rearview mirror. Due to the characteristics of the electronic rearview mirror, the content seen through the rearview mirror cannot change with the driver's perspective, and the displayed image does not have depth information (left-right parallax) either, resulting in inconvenient observation and difficulty in distance judgment. Summary of the Invention

[0003] The purpose of the present invention is to provide a three-dimensional display electronic rearview mirror system based on eye tracking to solve the above technical problems;

[0004] The purpose of the present invention is also to provide a working method of a three-dimensional display electronic rearview mirror system based on eye tracking to solve the above technical problems.

[0005] The technical problems solved by the present invention can be achieved by adopting the following technical solutions:

[0006] A three-dimensional display electronic rearview mirror system based on eye tracking includes a three-dimensional image processing module for adjusting the rearview mirror image according to the eye position of the driver. The three-dimensional image processing module includes,

[0007] An eye position calculation unit for calculating the eye coordinate parameters of the driver according to the driver's seat image;

[0008] A left and right eye image cropping unit connected to the eye position calculation unit for cropping the left-eye stereoscopic vision image and the right-eye stereoscopic vision image according to the eye coordinate parameters;

[0009] A left and right eye image synthesis unit connected to the left and right eye image cropping unit for synthesizing the cropped left-eye stereoscopic vision image and the cropped right-eye stereoscopic vision image into a synthesized image;

[0010] A three-dimensional imaging calculation unit connected to the left and right eye image synthesis unit for processing the synthesized image into a three-dimensional display format according to the eye coordinate parameters and outputting a three-dimensional vision image through the connected three-dimensional display.

[0011] Preferably, the eye position calculation unit includes,

[0012] An image feature point extraction subunit, configured to extract image feature points from the driver's seat image, and extract the coordinate positions of the driver's left eye and right eye;

[0013] A coordinate position establishment subunit, connected to the image feature point extraction subunit, configured to calculate the interpupillary distance according to the coordinate positions of the left eye and the right eye, identify the distance between the driver and the in-vehicle camera based on the interpupillary distance, and combine the binocular center point position to identify the orientation of the driver in the in-vehicle camera;

[0014] A coordinate conversion subunit, connected to the coordinate position establishment subunit, configured to calculate the spatial position of the driver according to the installation position of the in-vehicle camera, the distance between the driver and the in-vehicle camera, and the orientation of the driver in the in-vehicle camera, and obtain the eye coordinate parameters.

[0015] Preferably, it further includes the in-vehicle camera, connected to the stereoscopic image processing module, and the in-vehicle camera is configured to acquire the driver's seat image, and the in-vehicle camera is disposed at the A-pillar, the rearview mirror, or the instrument panel of the vehicle.

[0016] Preferably, it further includes an out-of-vehicle dual camera, connected to the stereoscopic image processing module, and the out-of-vehicle dual camera is configured to acquire a stereoscopic vision image, and the stereoscopic vision image includes the left-eye stereoscopic vision image and the right-eye stereoscopic vision image.

[0017] Preferably, the out-of-vehicle dual camera is disposed at the right rearview mirror, the left rearview mirror, or the rear window of the vehicle.

[0018] Preferably, the cropped left-eye stereoscopic vision image and the cropped right-eye stereoscopic vision image have a field of view angle and an optical relationship corresponding to the optical rearview mirror.

[0019] Preferably, the stereoscopic display includes a display screen and a micro prism disposed on the display screen, and the content of the display screen is directionally refracted in different directions through the micro prism in a fixed relationship, and according to the determined refraction relationship and the eye coordinate parameters, corresponding images are respectively displayed at the left-eye position and the right-eye position of the driver to form the stereoscopic vision image.

[0020] Preferably, the stereoscopic display is disposed at the driver-side A-pillar, the rearview mirror, or the passenger-side A-pillar of the vehicle.

[0021] A working method of a stereoscopic display electronic rearview mirror system based on eye tracking, applied to the stereoscopic display electronic rearview mirror system based on eye tracking, includes,

[0022] Step S1, acquiring the driver's seat image, and the eye position calculation unit calculates the eye coordinate parameters according to the driver's seat image;

[0023] Step S2. The left and right eye image cropping unit crops the left-eye stereoscopic vision image and the right-eye stereoscopic vision image according to the eye coordinate parameters;

[0024] Step S3. The left and right eye image synthesis unit synthesizes the cropped left-eye stereoscopic vision image and the cropped right-eye stereoscopic vision image into the synthesized image;

[0025] Step S4. The stereoscopic imaging calculation unit processes the synthesized image into a stereoscopic display format according to the eye coordinate parameters, and outputs a stereoscopic vision picture through the stereoscopic display.

[0026] Preferably, step S1 includes

[0027] Step S11. The image feature point extraction sub-unit extracts image feature points from the driver's seat image, and extracts the left-eye coordinate position and the right-eye coordinate position of the driver;

[0028] Step S12. The coordinate position establishment sub-unit calculates the interpupillary distance according to the left-eye coordinate position and the right-eye coordinate position, identifies the distance between the driver and the in-vehicle camera based on the interpupillary distance, and combines the binocular center point position to identify the driver's orientation in the in-vehicle camera;

[0029] Step S13. The coordinate conversion sub-unit calculates the driver's spatial position according to the installation position of the in-vehicle camera, the distance between the driver and the in-vehicle camera, and the driver's orientation in the in-vehicle camera, and obtains the eye coordinate parameters.

[0030] The beneficial effects of the present invention: Due to the above technical solutions, the present invention captures the content captured by the out-of-vehicle camera by identifying the spatial position of the driver's eyes, crops it into a rearview mirror picture suitable for the driver to view, and at the same time realizes the effect of stereoscopic display, so that the picture of the rearview mirror has depth information, making it easy for the driver to observe and convenient for judging distances. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 is the architecture diagram of the stereoscopic display electronic rearview mirror system based on eye tracking in the embodiment of the present invention;

[0032] Figure 2 is the architecture diagram of the eye position calculation unit in the embodiment of the present invention;

[0033] Figure 3 is the schematic diagram of the installation position of the in-vehicle camera in the embodiment of the present invention;

[0034] Figure 4 is the schematic diagram of the installation position of the out-of-vehicle dual cameras in the embodiment of the present invention;

[0035] Figure 5 Schematic diagram of the installation position of the stereoscopic display in the embodiment of the present invention;

[0036] Figure 6 Schematic diagram of the micro prism of the stereoscopic display in the embodiment of the present invention;

[0037] Figure 7 Schematic diagram of the steps of the working method of the stereoscopic display electronic rearview mirror system based on eye tracking in the embodiment of the present invention;

[0038] Figure 8 Schematic diagram of step S1 in the embodiment of the present invention. Detailed implementation manners

[0039] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0040] It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments may be combined with each other.

[0041] Next, the present invention will be further described in conjunction with the accompanying drawings and specific embodiments, but it is not a limitation of the present invention.

[0042] A stereoscopic display electronic rearview mirror system based on eye tracking, as Figure 1 shown, includes a stereoscopic image processing module 1 for adjusting the rearview mirror image according to the eye position of the driver. The stereoscopic image processing module 1 includes,

[0043] An eye position calculation unit 11 for calculating the eye coordinate parameters of the driver according to the driver's seat image;

[0044] A left and right eye image cropping unit 12 connected to the eye position calculation unit 11 for cropping the left-eye stereoscopic vision image and the right-eye stereoscopic vision image according to the eye coordinate parameters;

[0045] A left and right eye image synthesis unit 13 connected to the left and right eye image cropping unit 12 for synthesizing the cropped left-eye stereoscopic vision image and the cropped right-eye stereoscopic vision image into a synthesized image;

[0046] A stereoscopic imaging calculation unit 14 connected to the left and right eye image synthesis unit 13 for processing the synthesized image into a stereoscopic display format according to the eye coordinate parameters and outputting a stereoscopic vision image through the connected stereoscopic display 4.

[0047] Specifically, to address the deficiencies of the existing streaming media rearview mirror solutions, such as inconvenient observation and difficult distance judgment, the present invention provides a real-time transmission, processing, and stereoscopic display solution for stereoscopic images based on eye tracking. It can not only provide different rearview mirror images according to the driver's perspective but also display stereoscopic images with depth information on the screen.

[0048] In the present invention, the left and right eye images are respectively captured by a dual-camera system outside the vehicle, and the position of the user's eyes is recognized by a camera inside the vehicle to calculate the spatial position of the driver's eyes. According to the calculated spatial position of the driver's eyes, the content captured by the dual-camera system is cropped according to the position of the driver's eyes to obtain a rearview mirror image suitable for the driver to view, that is, to obtain a similar viewing angle (about 30°) and optical relationship as the optical rearview mirror.

[0049] At the same time, a stereoscopic display screen 41 is used to respectively display the cropped images of the dual-camera system in the driver's two eyes to achieve the effect of stereoscopic display, so that the rearview mirror image has depth information. This enables the driver using this system solution to easily observe and conveniently judge distances through the system.

[0050] Specifically, the stereoscopic image processing module 1 mainly includes four parts: an eye position calculation unit 11, a left and right eye image cropping unit 12, a left and right eye image synthesis unit 13, and a stereoscopic imaging calculation unit 14.

[0051] The real-time driver seat image input by the in-vehicle camera 2 will be processed by the eye position calculation unit 11, and the calculation result is the coordinate parameters of the driver's eyes;

[0052] The eye coordinate parameters will be used in the left and right eye image cropping unit 12. The two-way left and right eye images input in real time by the out-vehicle dual-camera 3 will be processed by the left and right eye image cropping unit 12. The left and right eye image cropping unit 12 will combine the driver's eye coordinate parameters for cropping to achieve the effect of simulating an optical rearview mirror.

[0053] The cropped two-way left and right eye images will be input into the left and right eye image synthesis unit 13 to synthesize the two images into one synthesized image, and the synthesized image will be input into the stereoscopic imaging calculation unit 14. The stereoscopic imaging calculation unit 14 will combine the coordinate parameters of the driver's eyes and cooperate with the stereoscopic display 4 to present the correct stereoscopic image on the display.

[0054] The present invention can automatically adjust the rearview mirror image according to the driver's eye position without manual adjustment, and at the same time achieve the stereoscopic display of the electronic rearview mirror. This is convenient for the driver to observe and judge.

[0055] In a preferred embodiment, as Figure 2 shown, the eye position calculation unit 11 includes

[0056] The image feature point extraction subunit 111 is used to extract image feature points from the driver's seat image and extract the coordinate positions of the driver's left eye and right eye;

[0057] The coordinate position establishment subunit 112 is connected to the image feature point extraction subunit 111 and is used to calculate the interpupillary distance based on the coordinate positions of the left eye and the right eye, identify the distance between the driver and the in-vehicle camera 2 based on the interpupillary distance, and combine the position of the binocular center point to identify the orientation of the driver with respect to the in-vehicle camera 2;

[0058] The coordinate conversion subunit 113 is connected to the coordinate position establishment subunit 112 and is used to calculate the spatial position of the driver based on the installation position of the in-vehicle camera 2, the distance between the driver and the in-vehicle camera 2, and the orientation of the driver with respect to the in-vehicle camera 2, and obtain the eye coordinate parameters.

[0059] Specifically, the main process of calculating the driver's eye position is executed by the eye position calculation unit 11, and the eye position calculation unit 11 mainly includes an image feature point extraction subunit 111, a coordinate position establishment subunit 112, and a coordinate conversion subunit 113;

[0060] The in-vehicle camera 2 part will provide two kinds of information: the installation position and the viewing angle of the camera and the driver's seat image.

[0061] The driver's seat image will be input into the image feature point extraction subunit 111. The image feature point extraction part will process the driver's seat image, extract the coordinate positions of the left and right eyes therein, and input them into the coordinate position establishment subunit 112.

[0062] The coordinate position establishment subunit 112 will combine the coordinates of the left and right eyes, calculate the interpupillary distance to judge the distance between the driver and the camera, and combine the position of the binocular center point to judge the orientation of the driver with respect to the camera.

[0063] Finally, the coordinate conversion subunit 113 combines the installation position of the in-vehicle camera 2, the distance of the user from the camera, and the orientation of the driver with respect to the camera, and the three are combined to calculate the spatial position where the driver is located.

[0064] In a preferred embodiment, as Figure 3 shown, it further includes an in-vehicle camera 2, which is connected to the stereoscopic image processing module 1. The in-vehicle camera 2 is used to obtain the driver's seat image, and the in-vehicle camera 2 is provided at the A-pillar, the rearview mirror, or the instrument panel of the vehicle.

[0065] Specifically, the in-vehicle camera 2 is used to identify the position of the driver's eyes and can be installed at the A-pillar, the rearview mirror, or the instrument panel of the vehicle.

[0066] In a preferred embodiment, asFigure 4 As shown, it further includes an external vehicle dual camera 3, which is connected to the stereoscopic image processing module 1. The external vehicle dual camera 3 is used to obtain stereoscopic vision images, and the stereoscopic vision images include a left-eye stereoscopic vision image and a right-eye stereoscopic vision image.

[0067] In a preferred embodiment, the external vehicle dual camera 3 is provided at the right rearview mirror, left rearview mirror or rear window of the vehicle.

[0068] Specifically, the external vehicle dual camera 3 is used to record stereoscopic vision images and can be installed at the right rearview mirror, left rearview mirror or rear window.

[0069] In a preferred embodiment, the cropped left-eye stereoscopic vision image and the cropped right-eye stereoscopic vision image have a field of view angle and an optical relationship corresponding to those of the optical rearview mirror.

[0070] Specifically, the images cropped by the left and right eye image cropping unit 12 have a field of view angle (about 30°) and an optical relationship corresponding to those of the optical rearview mirror, ensuring that the positions, sizes and angular relationships of the rear objects seen in the cropped left-eye stereoscopic vision image and the cropped right-eye stereoscopic vision image are similar to those observed through the optical rearview mirror.

[0071] In a preferred embodiment, as Figure 6 shown, the stereoscopic display 4 includes a display screen 41 and a micro prism 42 provided on the display screen 41. The content of the display screen 41 is directionally refracted in different directions in a fixed relationship through the micro prism 42, and corresponding images are respectively displayed at the left-eye position and right-eye position of the driver according to the determined refraction relationship and eye coordinate parameters, forming a stereoscopic vision image.

[0072] Specifically, Figure 6 briefly shows the operating principle of the stereoscopic display screen 41 in. The present invention uses an eye-tracking stereoscopic display screen 41, and through the micro prism 42 structure on the display screen 41, the content on the display screen 41 can be directionally refracted in different directions in a fixed relationship. According to the known refraction relationship and the user's position, different left-eye and right-eye images are respectively displayed at the left-eye and right-eye positions of the driver to achieve the effect of stereoscopic display.

[0073] More specifically, the stereoscopic imaging calculation unit 14 calculates how to correctly refract the content in the synthesized image to the left-eye and right-eye positions of the driver according to the known refraction relationship and eye coordinate parameters. Determine the parallax in the left and right eyes according to the eye coordinate parameters, and then accurately refract the corresponding images to the left-eye and right-eye positions of the driver through the micro prism 42 structure, so that the left-eye and right-eye of the driver see slightly different images, thus generating a sense of stereoscopy. Finally, a stereoscopic vision image is output through the stereoscopic display 4, enabling the driver to see the rear scene with depth information, which is convenient for observation and distance judgment.

[0074] In a specific embodiment, the present invention provides a schematic diagram of the steps for a stereoscopic imaging calculation unit 14 to correctly refract the synthetic image content to the positions of the driver's left and right eyes.

[0075] Step 1: Establish a spatial coordinate system and a model

[0076] First, establish a three-dimensional spatial coordinate system with the in-vehicle camera 2 as the origin to describe the spatial position relationship between the driver's eye positions and the objects in the synthetic image. Among them, the x-axis represents the lateral direction of the vehicle, the y-axis represents the longitudinal direction of the vehicle, and the z-axis represents the direction perpendicular to the vehicle plane. At the same time, according to the structure of the microprisms 42 of the stereoscopic display 4, establish a corresponding refraction model. This model is used to describe how the microprisms 42 refract the pixel points on the display screen 41 to specific positions in space in different directions, and determines the relationship between the refraction angle, refraction direction, and the pixel coordinates of the display screen 41. For a pixel at a certain coordinate point (x 0 , y 0 ) on the display screen 41, the light propagation direction vector in space after being refracted by the microprisms 42 can be calculated.

[0077] Step 2: Calculate the parallax information

[0078] According to the eye coordinate parameters (x e , y e , z e ) and the coordinate information of the objects in the synthetic image (x o , y o , z o ), calculate the parallax of the objects relative to the driver's binoculars. Parallax is used to reflect the difference in the imaging positions of the same object in the left and right eyes. The formula for calculating the parallax in the present invention is derived from the principles of geometric optics. The parallax Δx in the x direction = x o - x e , the parallax Δy in the y direction = y o - y e , and the parallax Δz in the z direction = z o - z e . Adjust the display positions of the synthetic image in the left and right eyes through the parallax information to generate a stereoscopic effect.

[0079] Step 3: Determine the adjustment amount of the left and right eye display positions

[0080] Based on the parallax information and the refraction model, calculate the position adjustment amount that the objects in the synthesized image should display in the left and right eyes. For the left eye, according to the position of the left eye, the parallax Δx, Δy, Δz, and the refraction model, calculate the pixel coordinates (Δxl, Δyl) that the object seen by the left eye should offset on the display screen 41. Similarly, for the right eye, calculate the pixel coordinates (Δxr, Δyr) that the object seen by the right eye should offset on the display screen 41.

[0081] Step 4, image content remapping

[0082] According to the calculated position adjustment amounts for the left and right eyes' displays, remap the content in the synthesized image. That is, reassign each pixel point in the synthesized image to the corresponding display positions of the left and right eyes according to the calculated offset amounts. For a feature point P in the synthesized image, assume its original coordinates are (x p , y p ). According to the offset amount (Δxl, Δyl) of the left eye, adjust its mapped coordinates in the left-eye display image to (x p + Δxl, y p + Δyl); for the right eye, according to the offset amount (Δx r , Δy r ), adjust its mapped coordinates in the right-eye display image to (x p + Δxr, y p + Δyr).

[0083] Finally, refract the remapped image content through the microprism 42 structure of the stereoscopic display 4. According to its fixed refraction relationship, the microprism 42 refracts the adjusted image pixels to the positions of the driver's left and right eyes in a predetermined direction.

[0084] In a preferred embodiment, as Figure 5 shown, the stereoscopic display 4 is provided at the A-pillar on the driver's side of the vehicle, at the rearview mirror, or at the A-pillar on the passenger's side.

[0085] Specifically, the stereoscopic display 4 is used to present a stereoscopic vision picture and can be installed at the A-pillar on the driver's side, at the rearview mirror, or at the A-pillar on the passenger's side.

[0086] A working method of a stereoscopic display electronic rearview mirror system based on eye tracking, applied to the stereoscopic display electronic rearview mirror system in any of the embodiments, as Figure 7 shown, includes,

[0087] Step S1, obtain the driver's seat image, and the eye position calculation unit 11 calculates the eye coordinate parameters according to the driver's seat image;

[0088] Step S2, the left and right eye image cropping unit 12 crops the left-eye stereoscopic vision image and the right-eye stereoscopic vision image according to the eye coordinate parameters;

[0089] Step S3, the left and right eye image synthesis unit 13 synthesizes the cropped left-eye stereoscopic vision image and the cropped right-eye stereoscopic vision image into a synthesized image;

[0090] Step S4, the stereoscopic imaging calculation unit 14 processes the synthesized image into a stereoscopic display format according to the eye coordinate parameters and outputs a stereoscopic vision picture through the stereoscopic display 4.

[0091] Specifically, the stereoscopic picture processing module 1 mainly includes four parts: an eye position calculation unit 11, a left and right eye image cropping unit 12, a left and right eye image synthesis unit 13, and a stereoscopic imaging calculation unit 14.

[0092] The real-time driver's seat image transmitted by the in-vehicle camera 2 will be processed by the eye position calculation unit 11, and the calculation result is the eye coordinate parameters of the driver;

[0093] The eye coordinate parameters will be used in the left and right eye image cropping unit 12. The two left and right eye images transmitted in real time by the out-of-vehicle dual cameras 3 will be processed by the left and right eye image cropping unit 12. The left and right eye image cropping unit 12 will combine the driver's eye coordinate parameters for cropping to achieve the effect of simulating an optical rearview mirror.

[0094] The two cropped left and right eye images will be transmitted to the left and right eye image synthesis unit 13 to synthesize the two images into one synthesized image, and the synthesized image will be transmitted to the stereoscopic imaging calculation unit 14. The stereoscopic imaging calculation unit 14 will combine the coordinate parameters of the driver's eyes and cooperate with the stereoscopic display 4 to present the correct stereoscopic image on the display.

[0095] In a preferred embodiment, as Figure 8 shown, step S1 includes,

[0096] Step S11, the image feature point extraction sub-unit 111 extracts image feature points from the driver's seat image and extracts the left-eye coordinate position and the right-eye coordinate position of the driver;

[0097] Step S12, the coordinate position establishment sub-unit 112 calculates the interpupillary distance according to the left-eye coordinate position and the right-eye coordinate position, identifies the distance between the driver and the in-vehicle camera 2 based on the interpupillary distance, and combines the binocular center point position to identify the driver's orientation in the in-vehicle camera 2;

[0098] Step S13, the coordinate conversion sub-unit 113 calculates the driver's spatial position according to the installation position of the in-vehicle camera 2, the distance between the driver and the in-vehicle camera 2, and the driver's orientation in the in-vehicle camera 2, and obtains the eye coordinate parameters.

[0099] Specifically, the main process of calculating the driver's eye position is executed by the eye position calculation unit 11, which mainly includes an image feature point extraction subunit 111, a coordinate position establishment subunit 112, and a coordinate conversion subunit 113;

[0100] The in-vehicle camera 2 will provide two kinds of information: the installation position and the viewing angle of the camera, and the driver's seat image.

[0101] The driver's seat image will be input into the image feature point extraction subunit 111. The image feature point extraction part will process the driver's seat image, extract the coordinate positions of the left and right eyes therein, and input them into the coordinate position establishment subunit 112.

[0102] The coordinate position establishment subunit 112 will combine the coordinates of the left and right eyes, calculate the interpupillary distance to judge the distance between the driver and the camera, and combine the position of the binocular center point to judge the orientation of the driver relative to the camera.

[0103] Finally, the coordinate conversion subunit 113 combines the installation position of the in-vehicle camera 2, the distance between the user and the camera, and the orientation of the driver relative to the camera, and calculates the spatial position where the driver is located by combining these three factors.

[0104] The above are only the preferred embodiments of the present invention, and do not limit the implementation manners and protection scope of the present invention. For those skilled in the art, it should be realized that all equivalent replacements and obvious changes made by using the description and illustrations of the present invention should be included in the protection scope of the present invention.

Claims

1. A stereoscopic display electronic rearview mirror system based on eye tracking, characterized in that: The invention comprises a stereoscopic image processing module (1) for adjusting the rearview mirror image according to the eye position of the driver, wherein the stereoscopic image processing module (1) comprises: An eye position calculation unit (11), used for calculating eye coordinate parameters of the driver based on the driver seat image; A left-eye and right-eye image cutting unit (12), connected to the eye position calculation unit (11), used for cutting the left-eye stereoscopic vision image and the right-eye stereoscopic vision image according to the eye coordinate parameters; A left-eye and right-eye image synthesis unit (13), connected to the left-eye and right-eye image cutting unit (12), used for synthesizing the cut left-eye stereoscopic vision image and the cut right-eye stereoscopic vision image into a synthetic image; A stereoscopic imaging calculation unit (14) is connected to the left and right eye image synthesis unit (13) and is used to process the synthesized image into a stereoscopic display format according to the eye coordinate parameters and output a stereoscopic visual image through a connected stereoscopic display (4).

2. The eye-tracking based stereoscopic display electronic rearview mirror system according to claim 1, characterized in that: The eye position calculation unit (11) comprises: An image feature point extraction subunit (111), used to extract image feature points from the driver's seat image, and to extract the left eye coordinate position and the right eye coordinate position of the driver; A coordinate position establishing subunit (112) is connected to the image feature point extracting subunit (111) and is used to calculate the pupil distance according to the left eye coordinate position and the right eye coordinate position, identify the distance between the driver and the in-vehicle camera (2) based on the pupil distance, and identify the position of the driver relative to the in-vehicle camera (2) in combination with the binocular center point position; A coordinate conversion subunit (113) is connected to the coordinate position establishment subunit (112) and is used to calculate the spatial position of the driver according to the installation position of the in-vehicle camera (2), the distance between the driver and the in-vehicle camera (2), and the orientation of the driver with respect to the in-vehicle camera (2), so as to obtain the eye coordinate parameters.

3. The eye-tracking-based stereoscopic display electronic rearview mirror system according to claim 2, characterized in that: It also includes an in-vehicle camera (2) connected to the stereoscopic image processing module (1), the in-vehicle camera (2) being used to obtain the driver's seat image, and the in-vehicle camera (2) being arranged at the A-pillar, the reflector or the instrument panel of the vehicle.

4. The eye-tracking-based stereoscopic display electronic rearview mirror system according to claim 1, characterized in that: It also includes an external dual camera (3) connected to the stereoscopic image processing module (1), and the external dual camera (3) is used to obtain a stereoscopic visual image, wherein the stereoscopic visual image includes the left-eye stereoscopic visual image and the right-eye stereoscopic visual image.

5. The eye-tracking-based stereoscopic display electronic rearview mirror system according to claim 4, characterized in that: The dual cameras (3) outside the vehicle are arranged at the right rearview mirror, the left rearview mirror or the rear window of the vehicle.

6. The eye-tracking based stereoscopic display electronic rearview mirror system according to claim 1, characterized in that: The cropped left-eye stereoscopic vision image and the cropped right-eye stereoscopic vision image have a viewing angle and an optical relationship corresponding to an optical rearview mirror.

7. The eye-tracking based stereoscopic display electronic rearview mirror system according to claim 1, characterized in that: The stereoscopic display (4) comprises a display screen (41) and a microprism (42) arranged on the display screen (41), wherein the content of the display screen (41) is directionally refracted in different directions in a fixed relationship through the microprism (42), and corresponding images are displayed to the left eye position and the right eye position of the driver respectively according to the determined refraction relationship and the eye coordinate parameters, thereby forming the stereoscopic visual image.

8. The eye-tracking based stereoscopic display electronic rearview mirror system according to claim 7, characterized in that: The three-dimensional display (4) is arranged at the driver's side A-pillar, the reflector or the assistant's side A-pillar of the vehicle.

9. A method for operating a stereoscopic display electronic rearview mirror system based on eye tracking, characterized in that: The eye-tracking-based stereoscopic display electronic rearview mirror system as claimed in any one of claims 1 to 8 comprises: Step S1, acquiring the driver's seat image, and the eye position calculation unit (11) calculates the eye coordinate parameters according to the driver's seat image; Step S2, the left-eye and right-eye image cutting unit (12) cuts the left-eye stereoscopic vision image and the right-eye stereoscopic vision image according to the eye coordinate parameters; Step S3, the left-eye and right-eye image synthesis unit (13) synthesizes the cropped left-eye stereoscopic vision image and the cropped right-eye stereoscopic vision image into the synthesized image; Step S4, the stereoscopic imaging calculation unit (14) processes the synthetic image into a stereoscopic display format according to the eye coordinate parameters, and outputs a stereoscopic visual picture through the stereoscopic display (4).

10. The working method of the stereoscopic display electronic rearview mirror system based on eye tracking according to claim 9, characterized in that: Step S1 comprises, Step S11, the image feature point extraction subunit (111) extracts image feature points from the driver's seat image to extract the left eye coordinate position and the right eye coordinate position of the driver; Step S12, the coordinate position establishing subunit (112) calculates the pupil distance according to the left eye coordinate position and the right eye coordinate position, identifies the distance between the driver and the in-vehicle camera (2) based on the pupil distance, and identifies the position of the driver relative to the in-vehicle camera (2) in combination with the binocular center point position; Step S13, the coordinate conversion subunit (113) calculates the spatial position of the driver according to the installation position of the in-vehicle camera (2), the distance between the driver and the in-vehicle camera (2), and the orientation of the driver with respect to the in-vehicle camera (2), and obtains the eye coordinate parameters.

Images