Auxiliary driving method based on intelligent identification and auxiliary driving instrument adopting method

Abstract

The invention relates to an auxiliary driving method based on intelligent identification and an auxiliary driving instrument adopting the method. The auxiliary driving method based on intelligent identification comprises the following steps: S1, image data intelligent identification processing: carrying out CVAT data labeling on an image acquired by an auxiliary driving instrument, realizing semi-automatic labeling, covering labeling information of a required image detection segmentation algorithm, realizing intelligent identification, and adopting a yolk algorithm model in target detection to realize real-time accurate detection; s2, performing acceleration processing on the deep learning neural network by adopting a Hisilicon media SoC of a Hisilicon NNIE neural network inference engine, and providing high-precision low-delay perception output for a night vision system; and S3, completing calculation including semantic segmentation, target detection and target identification in real time, and realizing AI image data identification, so that signboards, pedestrians and vehicles encountered in the driving process can be accurately identified, and auxiliary driving is realized. According to the auxiliary driving method based on intelligent identification, intelligent identification and real-time accurate detection can be realized.

Description

Technical Field

[0001] The present invention relates to the technical field of assisted driving, and in particular to an assisted driving method based on intelligent recognition and an assisted driving instrument using the method. Background Art

[0002] At present, assisted driving devices do not have AI assisted driving capabilities and cannot actively identify pedestrians, vehicles, drones and other units and their locations and distance information in the video captured by the camera. At the same time, current night vision technology is mainly used in sights, helmet observation mirrors and handheld observation mirrors. There are few night vision technologies applied to the field of assisted driving devices, resulting in the lack of night vision capabilities in current assisted driving applications. In the military field, existing assisted driving devices cannot fully meet the actual use needs of night driving and cannot respond quickly to complex battlefield environments.

[0003] In order to conceal, drive safely and collect battlefield information during combat, military vehicles urgently need an all-weather night-time assisted driving device with a large field of view, moderate viewing distance, certain AI-assisted driving capabilities, and strong environmental adaptability. Summary of the invention

[0004] The technical problem to be solved by the present invention is how to provide an assisted driving device that is applicable to all weather conditions including night time, with a large field of view, moderate viewing distance, AI assisted driving capability, and strong environmental adaptability.

[0005] To solve the above technical problems, according to one aspect of the present invention, an assisted driving method based on intelligent recognition is provided, and the assisted driving method based on intelligent recognition is applied to an assisted driving instrument. The assisted driving method based on intelligent recognition includes the following steps: S1, intelligent recognition and processing of image data, CVAT data annotation of the images collected by the assisted driving instrument, and semi-automatic annotation is realized, covering the annotation information of the required image detection and segmentation algorithm, realizing intelligent recognition, and adopting the yolo algorithm model in target detection to realize real-time and accurate detection; S2, HiSilicon Media SoC using HiSilicon NNIE neural network inference engine accelerates the processing of deep learning neural network, and provides high-precision and low-latency perception output for night vision system; S3, real-time completion of calculations including semantic segmentation, target detection, and target recognition, realizing AI image data recognition, so as to accurately identify signs, pedestrians, and vehicles encountered during driving, and realize assisted driving.

[0006] According to an embodiment of the present invention, step S3 may also include: training a model, applying a gradually expanded data set during the calculation process to a model including CVAT data annotation, a YOLO algorithm model, and a deep learning neural network for training; as more and more data streams enter the training, the recognition success rate is improved and the recognition types are increased.

[0007] According to another aspect of the present invention, an assisted driving instrument is provided, which includes: a housing for supporting and including various components of the assisted driving instrument; a lens having two ends, one end facing the object side and the other end facing the eye side; a low-light camera CMOS, arranged close to the eye side, for use in an environment with an illumination of 1 to 10 -3 The low-light camera core of LUX realizes normal operation under low-light conditions, collects and processes the weak light reflected by the object and outputs it, and then the human eye can observe the low-light environment through the display screen; the millimeter-wave radar is installed and fixed on the shell, and is used to measure the distance, speed and direction; the display is used to display the identified signs, pedestrians, vehicles encountered during driving, or information to prompt the driver, so as to realize assisted driving; the control component receives and processes the image data output by the lens, the low-light camera CMOS and the millimeter-wave radar data, and uses the assisted driving method based on intelligent recognition as claimed in any one of claims 1-2 to perform intelligent recognition of assisted driving, realize AI image intelligent recognition, and integrate the strong light suppression technology, and output the information to be displayed to the display for display, so as to realize assisted driving under various weather conditions.

[0008] According to an embodiment of the present invention, the control component may be provided with an FPGA module, a HiSilicon VI ISP channel, an AI hardware module, a CAN controller, and an LB2022-00 image processing module, wherein the FPGA module converts and processes the video or image data received by the control component, and sends it to the HiSilicon VI ISP channel for HiSilicon standard sensor image processing, and then uses the AI ​​hardware module to complete a series of calculations including semantic segmentation, target detection, and target recognition in real time, accurately recognize information including signs, pedestrians, vehicles, and drones, and then uses the CAN controller to transmit the data, and finally uses the LB2022-00 image processing module to process the image signals sent to each image sensor, encode them, and then send them to the display for human eye observation.

[0009] According to an embodiment of the present invention, the FPGA system refers to a video processing system based on an FPGA minimum system, a system that accelerates image processing, reduces image transmission delay and other programmable functions. AI intelligent recognition is artificial intelligence recognition, which uses a deep learning algorithm to train an artificial intelligence model to achieve target recognition. The present invention refers to a multi-target, fast and high-accuracy intelligent recognition algorithm. CVAT data labeling refers to an algorithm for labeling data during target recognition, such as an algorithm for selecting the identified target and displaying the speed and distance of the target. The yolo algorithm model is a yolo algorithm model in target detection, a target detection system based on a single neural network. The HiSilicon NNIE neural network inference engine is a hardware unit for accelerating the processing of deep learning convolutional neural networks. The HiSilicon VI ISP channel is the name of the ISP channel of the HiSilicon VI chip module, which is used for video data transmission. The CAN controller is a dedicated chip for CAN bus communication. The LB2022-00 image processing module is an image processor, a module that processes and converts images and outputs them.

[0010] According to an embodiment of the present invention, the lens may include an objective lens barrel and a first lens, a cemented lens, a third lens, a fourth lens, and a fifth lens arranged in sequence from the object side to the eye side, wherein the first lens is a biconvex lens facing the object side, the third lens is a plano-convex lens with the convex surface facing the object side, the fourth lens is a biconcave lens, and the fifth lens is a plano-concave lens with the concave surface facing the object side. The objective lens barrel has a reserved through hole at its edge close to the object side, and the entire objective lens is fixed to the housing by threads. The objective lens barrel has a reserved threaded hole at its edge close to the image side, and the low-light camera core is fixed to the objective lens barrel through an adapter plate by screws.

[0011] According to an embodiment of the present invention, the low-light camera CMOS can be connected to the lens via a core adapter.

[0012] According to an embodiment of the present invention, an adjustment gasket can be set between the low-light camera CMOS and the lens to adjust the imaging clarity. The image captured by the low-light camera CMOS is transmitted to the control component, and the AI ​​image data carried by the control component is intelligently processed and output to superimpose the original video data, which is then displayed on a monitor.

[0013] According to an embodiment of the present invention, the control component may include a power management module, which supplies power to the low-light camera CMOS and millimeter-wave radar through the power management board module. The control component comprehensively processes the data of each module and outputs it to a display for human eye observation.

[0014] According to an embodiment of the present invention, the assisted autopilot may also include: a cable, through which communication and power supply are performed, the cable connection includes a cigarette lighter socket and an aviation socket for power supply, the cigarette lighter is connected through the cigarette lighter socket to power the host and the display, or the aviation socket is connected for communication between the display and the host to realize mutual data transmission.

[0015] According to an embodiment of the present invention, the assisted driving instrument may also include: a display bracket, a rubber suction cup, a radar cover, an adjustment bracket, a rubber-coated magnet and a shell cover, wherein the display bracket is connected to the display screen by screws, and when in use, the display is fixed to the front windshield of the vehicle or other plane by the rubber suction cup, and the bracket is adjusted to make the display screen at an angle convenient for human eye observation.

[0016] Compared with the prior art, the technical solution provided by the embodiments of the present invention can at least achieve the following beneficial effects:

[0017] According to the intelligent recognition-based assisted driving method of the present invention, a low-illumination COMS is used to -3 LUX's low-light camera core can work normally under low-light conditions. It collects and processes the weak light reflected by objects and outputs it, which allows the human eye to observe the low-light environment through the display screen.

[0018] According to the assisted driving method based on intelligent recognition of the present invention, it is implemented based on AI intelligent recognition. The AI ​​recognition technology uses CVAT (Computer Vision Annotation Tool) data annotation to realize semi-automatic annotation, covering the annotation information of the required image detection and segmentation algorithm; the present invention adopts the yolo algorithm model in target detection, and its rich weight model, real-time detection speed, precise detection performance and unique decoupling head processing method are very suitable for the practical application of the present invention. By adaptively improving the above-mentioned model, its performance and indicators are improved.

[0019] According to the intelligent recognition-based assisted driving method of the present invention, the HiSilicon NNIE neural network inference engine, i.e., a hardware unit in the HiSilicon media SoC that is specifically used for accelerating the processing of neural networks, especially deep learning neural networks, can provide high-precision and low-latency perception output for the night vision system. It can complete a series of calculations such as semantic segmentation, target detection, and target recognition in real time, and is the basis for accurately identifying signs, pedestrians, and vehicles, thereby realizing AI image data recognition based on the trained model.

[0020] According to the assisted driving device of the assisted driving method based on intelligent recognition of the present invention, in order to conceal, drive safely and collect battlefield information during combat, the vehicle can achieve assisted driving with a large field of view, moderate viewing distance, AI assisted driving capability, strong environmental adaptability, and all-weather applicability including nighttime. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings of the embodiments will be briefly introduced below. Obviously, the drawings in the following description only relate to some embodiments of the present invention, but are not intended to limit the present invention.

[0022] Figure 1 is a schematic diagram showing an assisted driving method based on intelligent recognition according to an embodiment of the present invention;

[0023] Figure 2 is a schematic diagram showing an auxiliary driving instrument according to an embodiment of the present invention;

[0024] Figure 3 is a cross-sectional view showing a main unit of a driver assistance device according to an embodiment of the present invention;

[0025] Figure 4 is a schematic diagram showing a display screen according to an embodiment of the present invention;

[0026] Figure 5 Schematic diagram of a display bracket according to an embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] Unless otherwise defined, the technical or scientific terms used herein shall have the common meanings understood by persons with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the patent application specification and claims of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "one" or "a" do not indicate a quantity limitation, but indicate the existence of at least one.

[0029] Figure 1 is a schematic diagram showing an assisted driving method based on intelligent recognition according to an embodiment of the present invention;

[0030] The assisted driving method based on intelligent recognition is applied to assisted driving instruments, such as Figure 1 As shown, the assisted driving method based on intelligent recognition includes the following steps:

[0031] S1, intelligent recognition and processing of image data, annotating the images collected by the assisted driving instrument with CVAT data, realizing semi-automatic annotation, covering the annotation information of the required image detection and segmentation algorithm, realizing intelligent recognition, and using the YOLO algorithm model in target detection to achieve real-time and accurate detection;

[0032] S2, HiSilicon Media SoC using HiSilicon NNIE neural network inference engine accelerates the processing of deep learning neural networks, providing high-precision and low-latency perception output for night vision systems;

[0033] S3. Complete calculations including semantic segmentation, target detection, and target recognition in real time to realize AI image data recognition, so as to accurately identify signs, pedestrians, and vehicles encountered during driving and realize assisted driving.

[0034] According to one or some embodiments of the present invention, step S3 also includes: training the model, applying the gradually expanded data set during the calculation process to the model including CVAT data annotation, YOLO algorithm model, and deep learning neural network for training; as more and more data streams enter the training, the recognition success rate is improved and the recognition types are increased.

[0035] According to the assisted driving method based on intelligent recognition of the present invention, it is implemented based on AI intelligent recognition. The AI ​​recognition technology uses CVAT (Computer Vision Annotation Tool) data annotation to realize semi-automatic annotation, covering the annotation information of the required image detection and segmentation algorithm; the present invention adopts the yolo algorithm model in target detection, and its rich weight model, real-time detection speed, precise detection performance and unique decoupling head processing method are very suitable for the practical application of the present invention. By adaptively improving the above-mentioned model, its performance and indicators are improved.

[0036] Figure 2 is a schematic diagram showing an auxiliary driving instrument according to an embodiment of the present invention; Figure 3 is a cross-sectional view showing a main unit of a driver assistance device according to an embodiment of the present invention; Figure 4 is a schematic diagram showing a display screen according to an embodiment of the present invention; Figure 5 Schematic diagram of a display bracket according to an embodiment of the present invention.

[0037] like Figure 2-5 As shown, the driver assistance device includes: a housing 1, a lens 2, a low-light camera CMOS 10, a millimeter-wave radar 3, a display 11 and a control component 7.

[0038] The housing 1 is used to support and contain various components of the auxiliary driving instrument.

[0039] The lens 2 has two ends, one end is opposite to the object side, and the other end is opposite to the eye side.

[0040] The low-light camera CMOS 10 is set close to the target and is used in environments with an illumination of 1 to 10 -3 LUX's low-light camera core can work normally under low-light conditions, collect and process the weak light reflected by the object and output it, and then the human eye can observe the low-light environment through the display screen.

[0041] The millimeter wave radar 3 is mounted and fixed on the housing 1 and is used to measure distance, speed and direction.

[0042] The display 11 is used to display the identified signs, pedestrians, vehicles encountered during driving, or information to prompt the driver, so as to realize assisted driving.

[0043] The control component 7 receives and processes the image data output by the lens 2, the low-light camera CMOS 10 and the data of the millimeter-wave radar 3, and adopts the assisted driving method based on intelligent recognition as claimed in any one of claims 1-2 to perform intelligent recognition of assisted driving, realize AI image intelligent recognition, and integrate the strong light suppression technology, and output the information to be displayed to the display 11 for display, so as to realize assisted driving under various weather conditions.

[0044] According to the intelligent recognition-based auxiliary driving instrument of the present invention, a low-illumination COMS is used to -3 LUX's low-light camera core can work normally under low-light conditions. It collects and processes the weak light reflected by objects and outputs it, which allows the human eye to observe the low-light environment through the display screen.

[0045] According to one or some embodiments of the present invention, the control component 7 is provided with an FPGA module, a HiSilicon VI ISP channel, an AI hardware module, a CAN controller, and an LB2022-00 image processing module, wherein the FPGA module converts and processes the video or image data received by the control component 7, and sends it to the HiSilicon VI ISP channel for HiSilicon standard sensor image processing, and then uses the AI ​​hardware module to complete a series of calculations including semantic segmentation, target detection, and target recognition in real time, accurately recognize information including signs, pedestrians, vehicles, and drones, and then uses the CAN controller to transmit the data, and finally uses the LB2022-00 image processing module to process the image signals sent to each image sensor and encode them before sending them to the display 11 for human eye observation.

[0046] Among them, the FPGA system refers to a video processing system based on the FPGA minimum system, a system that accelerates image processing, reduces image transmission delay and other programmable functions. AI intelligent recognition is artificial intelligence recognition, which uses deep learning algorithms to train artificial intelligence models to achieve target recognition. The present invention refers to a multi-target, fast and high-accuracy intelligent recognition algorithm. CVAT data labeling refers to an algorithm that labels and processes data during target recognition, such as an algorithm that selects the identified target and displays the speed and distance of the target. The yolo algorithm model is a yolo algorithm model in target detection, a target detection system based on a single neural network. The HiSilicon NNIE neural network inference engine is a hardware unit that accelerates the processing of deep learning convolutional neural networks. The HiSilicon VI ISP channel is the name of the ISP channel of the HiSilicon VI chip module, which is used for video data transmission. The CAN controller is a dedicated chip for CAN bus communication. The LB2022-00 image processing module is an image processor, a module that processes and converts images and outputs them.

[0047] According to the intelligent recognition-based assisted driving method of the present invention, the HiSilicon NNIE neural network inference engine, i.e., a hardware unit in the HiSilicon media SoC that is specifically used for accelerating the processing of neural networks, especially deep learning neural networks, can provide high-precision and low-latency perception output for the night vision system. It can complete a series of calculations such as semantic segmentation, target detection, and target recognition in real time, and is the basis for accurately identifying signs, pedestrians, and vehicles, thereby realizing AI image data recognition based on the trained model.

[0048] According to one or some embodiments of the present invention, the lens 2 includes an objective lens barrel and a first lens, a cemented lens, a third lens, a fourth lens, and a fifth lens arranged in sequence from the object direction to the eye direction.

[0049] Among them, the first lens is a biconvex lens facing the object side, the third lens is a plano-convex lens with the convex surface facing the object side, the fourth lens is a biconcave lens, and the fifth lens is a plano-concave lens with the concave surface facing the object side. A through hole is reserved on the edge of the objective lens barrel close to the object side, and the entire objective lens is fixed to the housing 1 by threads. A threaded hole is reserved on the edge of the objective lens barrel close to the image side, and the low-light camera core is fixed to the objective lens barrel through an adapter plate by screws.

[0050] According to one or some embodiments of the present invention, the low-light camera CMOS 10 is connected to the lens 2 via a movement adapter 9 .

[0051] According to one or some embodiments of the present invention, an adjustment gasket is arranged between the low-light camera CMOS 10 and the lens 2 for adjusting the imaging clarity. The image captured by the low-light camera CMOS 10 is transmitted to the control component 7, and the image data is intelligently processed by the AI ​​image data carried by the control component 7, and then output and superimposed with the original video data, which is displayed through the display 11.

[0052] According to one or some embodiments of the present invention, the control component 7 includes a power management module, which supplies power to the low-light camera CMOS 10 and the millimeter-wave radar 3 through the power management board module. The control component 7 comprehensively processes the data of each module and outputs it to the display 11 for human eye observation.

[0053] According to one or some embodiments of the present invention, the assisted autopilot also includes: a cable, through which communication and power supply are performed, the cable connection includes a cigarette lighter socket and an aviation socket for power supply, the cigarette lighter is connected through the cigarette lighter socket to power the host 13 and the display 11, or the aviation socket is connected for communication between the display 11 and the host 13 to realize mutual data transmission.

[0054] According to one or some embodiments of the present invention, the assisted autopilot further includes: a display bracket 12 , a rubber suction cup, a radar cover 4 , an adjustment bracket 5 , a rubber-coated magnet 6 and a housing cover 8 .

[0055] The display bracket 12 is connected to the display screen by screws. When in use, the display 11 is fixed to the front windshield of the vehicle or other planes by a rubber suction cup, and the bracket is adjusted to make the display screen at an angle convenient for human eyes to observe.

[0056] When in use, place the vehicle host 13 on the top of the vehicle, connect the host 13 to the roof through the rubber-coated magnet 6, connect the display screen and the host 13 with a cable, and then insert the cigarette lighter plug of the cable into the cigarette lighter. After the indicator light of the cigarette lighter plug is on, you can long press the display switch to turn on the device. After turning on the device, click the encoder to use the functions in the menu to adjust the brightness, turn on AI recognition, turn on the distance display, etc. When you are done using it, long press the switch to turn it off.

[0057] According to the assisted driving device of the assisted driving method based on intelligent recognition of the present invention, in order to conceal, drive safely and collect battlefield information during combat, the vehicle can achieve assisted driving with a large field of view, moderate viewing distance, AI assisted driving capability, strong environmental adaptability, and all-weather applicability including nighttime.

[0058] The above description is merely an exemplary embodiment of the present invention and is not intended to limit the protection scope of the present invention. The protection scope of the present invention is determined by the appended claims.

Claims

1. An assisted driving method based on intelligent recognition, wherein the assisted driving method based on intelligent recognition is applied to an assisted driving instrument, and the assisted driving method based on intelligent recognition comprises the following steps: S1, intelligent recognition and processing of image data, annotating the images collected by the assisted driving instrument with CVAT data, realizing semi-automatic annotation, covering the annotation information of the required image detection and segmentation algorithm, realizing intelligent recognition, and using the YOLO algorithm model in target detection to achieve real-time and accurate detection; S2, HiSilicon Media SoC using HiSilicon NNIE neural network inference engine accelerates the processing of deep learning neural networks, providing high-precision and low-latency perception output for night vision systems; S3. Complete calculations including semantic segmentation, target detection, and target recognition in real time to realize AI image data recognition, so as to accurately identify signs, pedestrians, and vehicles encountered during driving and realize assisted driving.

2. The assisted driving method based on intelligent recognition as claimed in claim 1, wherein: Step S3 also includes: training the model, applying the gradually expanded data set during the calculation process to the model including CVAT data annotation, YOLO algorithm model, and deep learning neural network for training; as more and more data streams enter the training, the recognition success rate is improved and the recognition types are increased.

3. A driver assistance device, wherein: include: A housing, used to support and contain various components of the auxiliary driving instrument; The lens has two ends, one end facing the object side and the other end facing the eye side; Low-light camera CMOS, set close to the target, is used in the environment with an illumination of 1 to 10 -3 LUX's low-light camera core can work normally under low-light conditions, collect and process the weak light reflected by the object and output it, and then the human eye can observe the low-light environment through the display screen; Millimeter wave radar, installed and fixed on the shell, used to measure distance, speed and direction; A display for displaying the identified signs, pedestrians, vehicles encountered during driving, or information to prompt the driver, so as to realize assisted driving; The control component receives and processes the image data output by the lens, the low-light camera CMOS and the data of the millimeter-wave radar, and uses the assisted driving method based on intelligent recognition as described in any one of claims 1-2 to perform intelligent recognition of assisted driving, realize AI image intelligent recognition, and integrate the strong light suppression technology to output the information to be displayed to the display for display, so as to realize assisted driving under various weather conditions.

4. The driver assistance device according to claim 3, wherein: The control component is equipped with an FPGA module, a HiSilicon VI ISP channel, an AI hardware module, a CAN controller, and an LB2022-00 image processing module. The FPGA module converts and processes the video or image data received by the control component, and sends it to the HiSilicon VI ISP channel for HiSilicon standard sensor image processing. Then, the AI ​​hardware module is used to complete a series of calculations including semantic segmentation, target detection, and target recognition in real time, and accurately recognize information including signs, pedestrians, vehicles, and drones. The CAN controller is then used to transmit the data. Finally, the LB2022-00 image processing module is used to process the image signals sent to each image sensor, encode them, and then send them to the display for human observation.

5. The driver assistance device according to claim 3, wherein: The lens comprises an objective lens tube and a first lens, a cemented lens, a third lens, a fourth lens, and a fifth lens arranged in sequence from the object direction to the eye direction. Among them, the first lens is a biconvex lens facing the object side, the third lens is a plano-convex lens with the convex side facing the object side, the fourth lens is a biconcave lens, and the fifth lens is a plano-concave lens with the concave side facing the object side. There is a reserved through hole on the edge of the objective lens barrel close to the object side, and the entire objective lens is fixed to the shell by threads. There is a reserved threaded hole on the edge of the objective lens barrel close to the image side, and the low-light camera core is fixed to the objective lens barrel through an adapter plate by screws.

6. The driver assistance device according to claim 3, wherein: The low-light camera CMOS is connected to the lens via a core adapter.

7. The driver assistance device according to claim 6, wherein: An adjustment gasket is arranged between the low-light camera CMOS and the lens for adjusting the imaging clarity. The image captured by the low-light camera CMOS is transmitted to the control component, and the image data is intelligently processed by the AI ​​image data carried by the control component, and then output and superimposed with the original video data, which is displayed on the display.

8. The driver assistance device according to claim 3, wherein: The control component includes a power management module, through which the low-light camera CMOS and millimeter-wave radar are powered. The control component comprehensively processes the data of each module and then outputs it to a display for human observation.

9. The driver assistance device according to claim 3, wherein: Also includes: Cable, communication and power supply are carried out through the cable, and the cable connection includes a cigarette lighter socket and an aviation socket for power supply. The cigarette lighter is connected through the cigarette lighter socket to power the host and the display, or the aviation socket connection is used for communication between the display and the host to realize mutual data transmission.

10. The driver assistance device according to claim 3, wherein: Also includes: Monitor bracket, rubber suction cup, radar cover, adjustment bracket, rubber-coated magnet and housing cover, The monitor bracket is connected to the display screen by screws. When in use, the monitor is fixed to the front windshield of the vehicle or other flat surface by a rubber suction cup, and the bracket is adjusted to make the display screen at an angle convenient for human eye observation.

Images