Intelligent automobile rearview mirror

Abstract

The invention discloses an intelligent automobile rearview mirror which comprises a rearview mirror frame body and a mirror surface assembly, the mirror surface assembly is installed on the rearview mirror frame body and comprises a liquid crystal mirror surface and a conductive layer which are installed on the rearview mirror frame body, a plurality of rearview light sensors are installed on the liquid crystal mirror surface, and the liquid crystal mirror surface is coated with an electrochemical layer. An electronic controller and a control circuit are integrated on the conducting layer, and an angle control module and a voltage control module are integrated in the control circuit; the angle control module and the voltage control module integrated in the control circuit are matched with the rearview light sensor, the angle of the mirror surface assembly is adjusted according to the incident angle and illumination intensity of strong light from the back, meanwhile, the surface of the electrochemical layer is darkened, reflected light reflected from the rearview mirror is weakened, and the brightness of the rearview mirror is improved. Therefore, transient visual dizziness cannot be caused to a driver, and the driving risk is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent vehicle devices, and particularly to an intelligent car rearview mirror. Background Art

[0002] In the field of traditional car rearview mirrors, most existing rearview mirrors are optical reflection type with single functions. Even with the development of vehicle intelligence, some rearview mirrors with simple electronic auxiliary functions have emerged, but there are still many defects. For example, in image processing, in the face of bad weather or low light environment, the images collected by the camera are blurred, and it is difficult to effectively correct over-dark or over-bright areas. For example, at night, the strong light of the vehicle behind makes the local image over-bright and loses details, and the road conditions in the dark are difficult to see clearly. At the same time, there is a lack of reasonable division of the image detection area, and the whole image needs to be processed when detecting target objects, resulting in low efficiency; in road condition recognition, the judgment of the driving state of the vehicle behind is inaccurate, the speed, acceleration and relative distance of the vehicle behind cannot be accurately obtained, and it is easy to misjudge due to the lack of comprehensive analysis of the driving trajectory. When reversing, the detection range of road obstacles behind is limited; and there is a lack of warning function. Summary of the Invention

[0003] In order to solve the above-mentioned disadvantages in the prior art, the present invention proposes an intelligent car rearview mirror.

[0004] In order to solve the above technical problems, the technical solutions adopted by the present invention are as follows:

[0005] An intelligent car rearview mirror, comprising a rearview mirror body and a camera, wherein the rearview mirror body includes a display screen, a processor, a speaker and a microphone; the processor is provided with an image processing module, a road condition recognition module and a warning module;

[0006] The camera is used for collecting image information behind the vehicle and transmitting it to the image processing module;

[0007] The image processing module is used for processing the collected image information, including image enhancement and target recognition;

[0008] The road condition recognition module is connected to the image processing module and recognizes road conditions based on the processed image information, such as the driving state of the vehicle behind, the lane change trend, road obstacles, etc.;

[0009] The warning module is connected to the road condition recognition module and generates a warning message when a potential dangerous road condition is recognized;

[0010] The display screen is used for displaying the processed image information, the road condition recognition result and the warning message;

[0011] The microphone is used for receiving the voice commands of the driver;

[0012] The loudspeaker is used to play warning information and feedback information in response to voice commands.

[0013] Further, the image processing module includes an image enhancement unit, an object recognition unit, and a lane division unit; the image enhancement unit uses adaptive filtering and histogram equalization algorithms to enhance images blurred due to bad weather or low light, and corrects the brightness of over-bright and over-dark areas of the picture; the object recognition unit uses deep learning algorithms to identify target objects in the image, including vehicles, pedestrians, traffic signs, and road obstacles; the lane division unit uses the Canny edge detection algorithm combined with the Hough transform to identify traffic markings in the image, and determines the position ranges of the current lane and the adjacent lanes on both sides through the recognized traffic markings, so as to divide virtual lanes; when traffic markings are not recognized in the image, the area occupied by the virtual lanes in the image is divided according to preset virtual markings;

[0014] Based on the lanes divided by the lane division unit, the image processing module determines the detection areas of the current lane and the adjacent lanes on both sides. When detecting vehicles, it only detects and recognizes the images within the detection areas occupied by the lanes, improving efficiency.

[0015] Further, the processor is also provided with a vehicle state detection module, which is connected to the image processing module. By recognizing the images collected by the camera and processed by the image processing module, it judges the moving state of the vehicle, and the moving state includes reversing, normal forward movement, and turning.

[0016] Further, the road condition recognition module includes a following distance analysis unit, a lane change trend judgment unit, and a reversing obstacle detection unit; the following distance analysis unit judges the following distance of the vehicle behind according to the position of the target object in the image; the lane change trend judgment unit judges whether the vehicle behind has the intention of changing lanes by analyzing the driving trajectory and turn signal state of the vehicle behind for a period of time; the reversing obstacle detection unit detects the vehicle moving state judged by the vehicle state detection module, and when the vehicle is reversing, it conducts target recognition on the area occupied by the virtual lane in the middle at the rear to identify obstacles on the virtual lane.

[0017] Further, when the warning module detects that the following distance of the vehicle behind is less than the warning distance, the vehicle behind changes lanes, or the vehicle behind turns on the turn signal, it controls the loudspeaker to play the corresponding warning voice;

[0018] The processor is also provided with a vehicle speed detection module, and the warning module is provided with the ability to automatically adjust the warning distance according to the driving speed of the vehicle.

[0019] Further, the camera is disposed on the roof fin, the lens of the camera faces the rear of the vehicle, and the camera is provided with a cleaning brush driven by a motor. The processor is provided with an occlusion detection module for detecting whether there are stains blocking the view in the picture. The cleaning brush contacts the lens of the camera and can rotate under the drive of the motor to wipe off stains such as dust and rain on the lens surface, so as to ensure the clarity of image acquisition.

[0020] Further, the display screen is a touch display screen, and the driver can adjust the display content, brightness, and contrast through touch operations. The display content includes real-time images, road condition parameters, warning icons, and voice command operation interfaces.

[0021] Further, when the image processing module performs detection and recognition on the image, the following steps are executed:

[0022] S1: Preprocess the collected image, read the image collected by the camera, grayscale it, use Gaussian filtering or median filtering to remove noise, and use histogram equalization to enhance the image contrast;

[0023] S2: Lane traffic marking recognition: Use the Canny edge detection algorithm to detect the image edge, detect the straight line feature through the Hough transform, and then use the least squares method to fit the lane line to divide the lane;

[0024] S3: Locate and recognize the contour of the vehicle in the rear lane. Use the object detection algorithm based on deep learning to locate the license plate, determine the approximate position of the vehicle according to the license plate position, and then use the image segmentation algorithm to recognize the vehicle contour;

[0025] S4: Rear vehicle signal light recognition: Locate the signal light area within the recognized vehicle contour, extract the color features, and judge the signal light state according to the color features and flashing rules;

[0026] If it is the first frame image, perform complete lane traffic marking recognition, vehicle location and contour recognition, and rear vehicle signal light recognition. If it is not the first frame image, use the contour information recognized in the previous frame to assist in recognition;

[0027] Lane traffic markings: Based on the position of the lane line in the previous frame, perform edge detection and straight line fitting in a small range near it to update the lane line information;

[0028] Vehicle location and contour: Within the range where the vehicle may move, according to the vehicle contour and motion estimation in the previous frame, narrow the target detection and segmentation range, quickly locate the vehicle, and update the contour;

[0029] Rear vehicle signal light: Search near the signal light area in the previous frame, and quickly judge the signal light state by combining color and flashing features;

[0030] Vehicle distance judgment: Establish a geometric model based on camera parameters, combine the position of the vehicle in the image and the license plate size, and calculate the vehicle distance according to the geometric model;

[0031] Result output and feedback: Integrate information such as lane division, vehicle position, vehicle distance, and signal light status, display it on the display screen, and send corresponding warning signals through the warning module.

[0032] Furthermore, when the road condition recognition module detects the following situations, the speaker emits corresponding voice prompts:

[0033] When it is detected that the vehicle is changing lanes, a voice prompt of "You are changing lanes, please pay attention to the vehicles behind" is emitted;

[0034] When it is detected that the following vehicle is too close, reaching the corresponding threshold set by the warning module, a voice prompt of "The following vehicle is too close, please maintain a safe vehicle distance" is emitted;

[0035] When it is detected that the following vehicle turns on the turn signal, according to, a voice prompt of "The turn signal of the vehicle on the [left / right] side behind has been turned on" is emitted.

[0036] Furthermore, it also includes a storage module and a communication module; the storage module is connected to the processor and is used to store the processed image data, road condition recognition records, vehicle status data, and system setting information, supporting local storage and cloud backup of data; the communication module supports Bluetooth, Wi-Fi, and 4G / 5G communications, and can perform data interaction with smartphones, in-vehicle systems, or cloud servers to achieve functions such as remote control, software upgrade, and data sharing.

[0037] The beneficial effects of an intelligent automotive rearview mirror of the present invention are as follows: It is provided with an image processing module. Through the image enhancement unit, adaptive filtering and histogram equalization algorithms are adopted to improve the problem of blurred images in bad weather and low light conditions. It can also calculate the average brightness value and sub-region brightness values of the image to perform targeted correction on overly dark or bright regions, enhancing the image clarity and the ability to present details. The lane division unit uses advanced algorithms to accurately identify when there are traffic markings, and when there are no markings, it divides virtual lanes according to a preset geometric model and determines the detection area, only detecting and identifying images within the relevant lane areas, improving the image processing efficiency and reducing waste of computing resources. In terms of road condition recognition, the rear vehicle status analysis unit in the road condition recognition module can accurately judge the driving speed, acceleration, and relative distance of the rear vehicle by analyzing the position and movement information of the target object, providing comprehensive rear vehicle status information for the driver. The lane change trend judgment unit comprehensively analyzes the driving trajectory and turn signal status of the rear vehicle, significantly improving the accuracy of judging the rear vehicle's lane change intention. The reverse obstacle detection unit accurately identifies road obstacles behind the vehicle during reverse according to the vehicle movement status judged by the vehicle status detection module, expanding the detection range and enhancing reverse safety. In terms of the warning function, the warning module is connected to the vehicle speed detection module and can automatically adjust the warning distance according to the vehicle speed, giving an early warning at high speeds and avoiding false alarms at low speeds, improving the timeliness and accuracy of warnings. The rich warning voice prompts for behaviors such as lane changes and turning on turn signals of vehicles in different lanes enable the driver to clearly understand the complex road conditions behind and make preparations in advance, greatly enhancing driving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 is a schematic diagram of the hardware connection of an intelligent automotive rearview mirror of the present invention;

[0039] Figure 2 is a system architecture diagram of the processor of an intelligent automotive rearview mirror of the present invention;

[0040] Figure 3 is a system architecture diagram of the image processing module of an intelligent automotive rearview mirror of the present invention;

[0041] Figure 4 is a system price diagram of the road condition recognition module of an intelligent automotive rearview mirror of the present invention.

[0042] Among them, 1 is the rearview mirror body; 2 is the camera; 11 is the display screen; 12 is the processor; 13 is the speaker; 14 is the microphone; 121 is the image processing module; 1211 is the image enhancement unit; 1212 is the target recognition unit; 1213 is the lane division unit; 122 is the road condition recognition module; 1221 is the vehicle distance analysis unit; 1222 is the lane change trend judgment unit; 1223 is the reverse obstacle detection unit; 123 is the warning module; 124 is the occlusion detection module; 125 is the vehicle state detection module; 126 is the vehicle speed detection module; 15 is the indicator light; 21 is the cleaning brush. Detailed implementation manner

[0043] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. 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.

[0044] As Figures 1-4 shown, an intelligent vehicle rearview mirror includes a rearview mirror body 1 and a camera 2. The rearview mirror body 1 includes a display screen 11, a processor 12, a speaker 13, and a microphone 14; the processor 12 is provided with an image processing module 121, a road condition recognition module 122, and a warning module 123;

[0045] The camera 2 is used to collect image information of the rear of the vehicle and transmit it to the image processing module 121;

[0046] The image processing module 121 is used to process the collected image information, including image enhancement and target recognition;

[0047] The road condition recognition module 122 is connected to the image processing module 121 and recognizes road conditions based on the processed image information, such as the driving state of the following vehicle, the lane change trend, road obstacles, etc.;

[0048] The warning module 123 is connected to the road condition recognition module 122 and generates a warning message when a potential dangerous road condition is recognized;

[0049] The display screen 11 is used to display the processed image information, the road condition recognition result, and the warning message;

[0050] The microphone 14 is used to receive the voice commands of the driver;

[0051] The speaker 13 is used to play the warning message and the feedback message in response to the voice command.

[0052] Further, the image processing module 121 includes an image enhancement unit 1211, an object recognition unit 1212, and a lane division unit 1213; the image enhancement unit 1211 uses adaptive filtering and histogram equalization algorithms to enhance images blurred due to bad weather or low light, and corrects the brightness of over-bright and over-dark areas of the picture; the object recognition unit 1212 uses deep learning algorithms to identify target objects in the image, including vehicles, pedestrians, traffic signs, and road obstacles; the lane division unit 1213 uses the Canny edge detection algorithm combined with the Hough transform to identify traffic markings in the image, determines the position ranges of the current lane and the adjacent lanes on both sides through the identified traffic markings, and divides virtual lanes accordingly; when traffic markings are not recognized in the image, the area occupied by the virtual lanes in the image is divided according to preset virtual markings.

[0053] Based on the lanes divided by the lane division unit 1213, the image processing module 121 determines the detection areas of the current lane and the adjacent lanes on both sides. When detecting vehicles, it only detects and identifies the images within the detection areas occupied by the lanes, improving efficiency.

[0054] Further, the processor is also provided with a vehicle state detection module 125. The vehicle state detection module 125 is connected to the image processing module 121, and determines the moving state of the vehicle by recognizing the images collected by the camera 2 and processed by the image processing module 121. The moving states include reversing, normal forward movement, and turning.

[0055] Further, the road condition recognition module 122 includes a following distance analysis unit 1221, a lane change trend judgment unit 1222, and a reversing obstacle detection unit 1223; the following distance analysis unit 1221 determines the following distance of the vehicle behind according to the position of the target object in the image; the lane change trend judgment unit 1222 determines whether the vehicle behind has an intention to change lanes by analyzing the driving trajectory and turn signal state of the vehicle behind for a period of time; the reversing obstacle detection unit 1223 detects the target objects in the area occupied by the virtual lane in the middle at the rear when the vehicle is reversing according to the moving state of the vehicle judged by the vehicle state detection module 125, and recognizes the obstacles on the virtual lane.

[0056] Further, the warning module 123 controls the speaker 13 to play corresponding warning voices when it detects that the following distance of the vehicle behind is less than the warning distance, the vehicle behind changes lanes, or the vehicle behind turns on the turn signal.

[0057] When the following distance of the vehicle behind is less than the warning distance (set as X meters) automatically adjusted by the warning module 123 according to the vehicle speed, it issues a voice prompt of "The following distance behind is less than X meters. Please speed up."

[0058] When it is detected that the vehicle in the directly rear lane turns on the left / right turn signal, a voice prompt of "The vehicle in the directly rear lane turns on the left / right turn signal" is issued;

[0059] When it is detected that the vehicle in the directly rear lane changes lanes to the left / right, a voice prompt of "The vehicle in the directly rear lane changes lanes to the left / right" is issued;

[0060] When it is detected that the vehicle in the left lane turns on the right turn signal, a voice prompt of "The vehicle in the left lane turns on the right turn signal" is issued;

[0061] When it is detected that the vehicle in the left lane changes lanes to the right, a voice prompt of "The vehicle in the left lane changes lanes to the right" is issued;

[0062] When it is detected that the vehicle in the right lane turns on the left turn signal, a voice prompt of "The vehicle in the right lane turns on the left turn signal" is issued;

[0063] When it is detected that the vehicle in the right lane changes lanes to the left, a voice prompt of "The vehicle in the right lane changes lanes to the left" is issued;

[0064] The driver can also use the indicator light for prompting. There are 4 indicator lights 15 provided at the top of the rearview mirror body 1, and the on / off of the indicator lights 15 is used to prompt the dynamic of the following vehicle;

[0065] The processor 12 is further provided with a vehicle speed detection module 126, and the warning module 123 is provided with the function of automatically adjusting the warning distance according to the driving speed of the vehicle.

[0066] Further, the camera 2 is arranged on the roof fin, the lens of the camera 2 faces the rear of the vehicle, and the camera 2 is provided with a cleaning brush 21 driven by a motor. The processor 12 is provided with an occlusion detection module 124, and the occlusion detection module 124 is used to detect whether there are stains blocking the field of view in the picture. The cleaning brush 21 contacts the lens of the camera 2, and the cleaning brush 21 can rotate under the drive of the motor to wipe off stains such as dust and rainwater on the lens surface to ensure the clarity of image acquisition.

[0067] Further, the display screen 11 is a touch display screen, and the driver can adjust the display content, brightness and contrast through touch operations. The display content includes real-time images, road condition parameters, warning icons and a voice command operation interface.

[0068] Further, when the image processing module 121 performs detection and recognition on the image, the following steps are executed:

[0069] S1: Perform preprocessing on the collected image, read the image collected by the camera 2, grayscale it, use Gaussian filtering or median filtering to remove noise, and use histogram equalization to enhance the image contrast;

[0070] S2: Lane traffic marking recognition: Use the Canny edge detection algorithm to detect the image edges, detect the straight-line features through the Hough transform, and then fit the lane lines using the least squares method to divide the lanes;

[0071] S3: Locate and recognize the contours of the vehicles in the rear lane. Use the object detection algorithm based on deep learning to locate the license plate, determine the approximate position of the vehicle according to the license plate position, and then use the image segmentation algorithm to recognize the vehicle contour;

[0072] S4: Rear vehicle signal light recognition: Locate the signal light area within the recognized vehicle contour, extract the color features, and judge the signal light status according to the color features and blinking rules;

[0073] If it is the first frame of the image, perform complete lane traffic marking recognition, vehicle location and contour recognition, and rear vehicle signal light recognition. If it is not the first frame of the image, use the contour information recognized in the previous frame to assist in recognition;

[0074] Lane traffic markings: Based on the position of the lane lines in the previous frame, perform edge detection and straight-line fitting within a small range near it to update the lane line information;

[0075] Vehicle location and contour: Within the range where the vehicle may move, according to the vehicle contour and motion estimation in the previous frame, narrow the target detection and segmentation range, quickly locate the vehicle and update the contour;

[0076] Rear vehicle signal lights: Search near the signal light area in the previous frame, and quickly judge the signal light status by combining the color and blinking features;

[0077] Vehicle distance judgment: Establish a geometric model according to the parameters of camera 2, combine the position of the vehicle in the image and the size of the license plate, and calculate the vehicle distance based on the geometric model;

[0078] Result output and feedback: Integrate information such as lane division, vehicle position, vehicle distance, and signal light status, display it on the display screen 11, and send out corresponding warning signals through the warning module 123.

[0079] Furthermore, when the road condition recognition module 122 detects the following situations, the speaker 13 issues corresponding voice prompts:

[0080] When it is detected that the vehicle is changing lanes, issue a voice prompt of "You are changing lanes, please pay attention to the vehicles behind";

[0081] When it is detected that the following vehicle is following too closely and reaches the corresponding threshold set by the warning module 123, issue a voice prompt of "The following vehicle is following too closely, please keep a safe distance";

[0082] When it is detected that the following vehicle turns on the turn signal, according to, issue a voice prompt of "The turn signal of the following [left / right] side vehicle has been turned on".

[0083] Further, it further includes a storage module and a communication module; the storage module is connected to the processor 12 and is used for storing the processed image data, road condition recognition records, vehicle state data, and system setting information, supporting local storage and cloud backup of the data; the communication module supports Bluetooth, Wi-Fi, and 4G / 5G communications, and can perform data interaction with a smart phone, a vehicle-mounted system, or a cloud server to implement functions such as remote control, software upgrade, and data sharing.

[0084] The beneficial effects of an intelligent vehicle rearview mirror of the present invention are as follows: It is provided with an image processing module. Through the image enhancement unit, the adaptive filtering and histogram equalization algorithms are adopted to improve the problem of blurred images in bad weather and low light. It can also calculate the average brightness value and sub-region brightness value of the image to perform targeted correction on the too dark or too bright regions, improving the image clarity and detail presentation ability; the lane division unit uses advanced algorithms to accurately identify when there are traffic markings, and divides virtual lanes according to a preset geometric model when there are no markings, and determines the detection area accordingly, and only detects and identifies the images within the relevant lane areas, improving the image processing efficiency and reducing the waste of computing resources; in terms of road condition recognition, the rear vehicle state analysis unit in the road condition recognition module can accurately judge the driving speed, acceleration, and relative distance of the rear vehicle by analyzing the position and movement information of the target object, providing comprehensive rear vehicle state information for the driver. The lane change trend judgment unit comprehensively analyzes the driving trajectory and turn signal state of the rear vehicle, significantly improving the accuracy of judging the rear vehicle's lane change intention; the reverse obstacle detection unit accurately identifies the road surface obstacles behind the vehicle during reverse according to the vehicle movement state judged by the vehicle state detection module, expanding the detection range and improving the reverse safety; in terms of the warning function, the warning module is connected to the vehicle speed detection module, and can automatically adjust the warning distance according to the vehicle speed, giving an early warning at high speed and avoiding false alarms at low speed, improving the timeliness and accuracy of the warning. The rich warning voice prompts for behaviors such as lane change and turning on the turn signal of vehicles in different lanes can enable the driver to clearly understand the complex road conditions behind and make preparations in advance, greatly improving driving safety.

[0085] In the description of the present invention, the meaning of several is one or more, the meaning of multiple is more than two, understand greater than, less than, exceeding, etc. as not including the present number, and understand above, below, within, etc. as including the present number. If there is a description of first and second, it is only for the purpose of distinguishing technical features and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features or implicitly indicating the sequence relationship of the indicated technical features.

[0086] In the description of the present invention, unless otherwise clearly defined, words such as setting, installing, and connecting should be understood in a broad sense, and those skilled in the art can reasonably determine the specific meanings of the above words in the present invention in combination with the specific content of the technical solution.

[0087] The above describes the present invention and its implementation manners. Such description is not restrictive. What is shown in the drawings is only one of the implementation manners of the present invention, and the actual content is not limited thereto. Generally speaking, if those of ordinary skill in the art are inspired by it and, without departing from the gist of the present invention, design similar structural manners and embodiments to this technical solution without creative efforts, they shall fall within the protection scope of the present invention.

Claims

1. An intelligent automobile rearview mirror, characterized in that: The rearview mirror comprises a rearview mirror body (1) and a camera (2); the rearview mirror body (1) comprises a display screen (11), a processor (12), a speaker (13) and a microphone (14); the processor (12) is provided with an image processing module (121), a road condition recognition module (122) and an early warning module (123); The camera (2) is used to collect image information of the rear of the vehicle and transmit it to the image processing module (121); The image processing module (121) is used to process the collected image information, including image enhancement and target recognition; The road condition recognition module (122) is connected to the image processing module (121) and recognizes road conditions, such as the driving status of the following vehicle, lane change trend, road obstacles, etc., based on the processed image information; The warning module (123) is connected to the road condition recognition module (122) and generates warning information when a potentially dangerous road condition is recognized; The display screen (11) is used to display processed image information, road condition recognition results and warning information; The microphone (14) is used to receive the driver's voice command; The loudspeaker (13) is used to play warning information and feedback information in response to voice commands.

2. The smart car rearview mirror according to claim 1, characterized in that: The image processing module (121) comprises an image enhancement unit (1211), a target recognition unit (1212) and a lane division unit (1213); the image enhancement unit (1211) uses adaptive filtering and histogram equalization algorithms to enhance images blurred due to bad weather or low light, and correct the brightness of overly bright and overly dark areas of the image; the target recognition unit (1212) uses a deep learning algorithm to recognize target objects in the image, including vehicles, pedestrians, traffic signs and road obstacles; the lane division unit (1213) uses a Canny edge detection algorithm combined with a Hough transform to recognize traffic markings in the image, and determines the position range of the current lane and adjacent lanes on both sides through the recognized traffic markings, thereby dividing the virtual lane; when no traffic markings are recognized in the image, the area occupied by the virtual lane in the image is divided according to the preset virtual markings; The image processing module (121) determines the detection areas of the current lane and adjacent lanes on both sides based on the lanes divided by the lane division unit (1213), and when detecting a vehicle, only detects and identifies the image within the detection area occupied by the lane, thereby improving efficiency.

3. The smart car rearview mirror according to claim 1, characterized in that: The processor is also provided with a vehicle state detection module (125), which is connected to the image processing module (121) and determines the movement state of the vehicle by identifying the image collected by the camera (2) and processed by the image processing module (121), wherein the movement state includes reversing, normal forward movement and turning.

4. The smart automobile rearview mirror according to claim 3, characterized in that: The road condition recognition module (122) comprises a vehicle distance analysis unit (1221), a lane change trend judgment unit (1222) and a back-up obstacle detection unit (1223); the vehicle distance analysis unit (1221) judges the following distance of the rear vehicle according to the position of the target object on the image; the lane change trend judgment unit (1222) judges whether the rear vehicle has the intention to change lanes by analyzing the driving track and the state of the turn signal of the rear vehicle over a period of time; the back-up obstacle detection unit (1223) detects the vehicle movement state judged by the vehicle state detection module (125), performs target recognition on the area occupied by the virtual lane in the middle of the rear when the vehicle is backing up, and identifies obstacles on the virtual lane.

5. The smart automobile rearview mirror according to claim 3, characterized in that: The warning module (123) controls the speaker (13) to play a corresponding warning voice when detecting that the following vehicle's following distance is less than the warning distance, the following vehicle changes lanes, or the following vehicle turns on its turn signal; The processor (12) is also provided with a vehicle speed detection module (126), and the warning module (123) is provided with a warning distance that can be automatically adjusted according to the vehicle's driving speed.

6. The smart automobile rearview mirror according to claim 1, characterized in that: The camera (2) is arranged on the roof tail fin, the lens of the camera (2) faces the rear of the car, and the camera (2) is provided with a cleaning brush (21) driven by a motor, the processor (12) is provided with an occlusion detection module (124), the occlusion detection module (124) is used to detect whether there is a stain that blocks the field of view on the picture, the cleaning brush (21) is in contact with the lens of the camera (2), and the cleaning brush (21) can rotate under the drive of the motor to remove stains such as dust and rain on the lens surface, so as to ensure the clarity of image acquisition.

7. The smart automobile rearview mirror according to claim 1, characterized in that: The display screen (11) is a touch display screen, and the driver can adjust the display content, brightness and contrast through touch operation. The display content includes real-time images, road condition parameters, warning icons and a voice command operation interface.

8. The smart automobile rearview mirror according to claim 1, characterized in that: The image processing module (121) performs the following steps when detecting and identifying an image: S1: preprocessing the captured image, reading the image captured by the camera (2), graying it, removing noise by Gaussian filtering or median filtering, and enhancing the image contrast by histogram equalization; S2: Lane traffic marking recognition: Use the Canny edge detection algorithm to detect image edges, use the Hough transform to detect straight line features, and then use the least squares method to fit lane lines to divide lanes; S3: Position and identify the contour of the vehicle in the rear lane, use the deep learning-based target detection algorithm to locate the license plate, determine the approximate position of the vehicle based on the license plate position, and then use the image segmentation algorithm to identify the vehicle contour; S4: Rear vehicle signal light recognition: locate the signal light area within the recognized vehicle outline, extract color features, and determine the signal light status based on the color features and flashing rules; If it is the first frame image, complete lane traffic marking recognition, vehicle positioning and contour recognition, and rear vehicle signal light recognition are performed. If it is not the first frame image, the contour information recognized in the previous frame is used to assist in recognition. Vehicle distance judgment: A geometric model is established based on the parameters of the camera (2), and the vehicle distance is calculated based on the geometric model in combination with the position of the vehicle on the image and the size of the license plate; The result is output and fed back, and information such as lane division, vehicle position, vehicle distance and signal light status are integrated, and the processed image is displayed on the display screen (11) in real time.

9. The smart automobile rearview mirror according to claim 1, characterized in that: When the road condition recognition module (122) detects the following situations, the speaker (13) issues a corresponding voice prompt: When the vehicle is detected changing lanes, a voice prompt "You are changing lanes, please pay attention to the vehicle behind" is issued; When it is detected that the distance between the rear vehicle and the vehicle is too close and reaches the corresponding threshold set by the warning module (123), a voice prompt "The distance between the rear vehicle and the vehicle is too close, please maintain a safe distance" is issued; When it is detected that the rear vehicle has turned on its turn signal, a voice prompt is issued: "The turn signal of the rear [left / right] vehicle is on." 10. The smart automobile rearview mirror according to claim 1, characterized in that: It also includes a storage module and a communication module; the storage module is connected to the processor (12) and is used to store processed image data, road condition recognition records, vehicle status data and system setting information, and supports local storage and cloud backup of data; the communication module supports Bluetooth, Wi-Fi and 4G / 5G communications, and can interact with smart phones, vehicle systems or cloud servers to achieve remote control, software upgrades and data sharing functions.

Images