A fisheye camera-based vehicle orientation angle acquisition method and related device

Abstract

The application provides a fisheye camera-based vehicle orientation angle determination method and related equipment. A target image including a target vehicle is obtained based on a fisheye camera, and the target image is subjected to de-distortion processing according to internal and external parameters of the fisheye camera to obtain a de-distorted target image. Then, an incident angle of the target vehicle relative to the fisheye camera is determined according to the de-distorted target image and the internal and external parameters of the fisheye camera, and a local orientation of the target vehicle in the de-distorted target image is determined through a preset deep learning network. Finally, the orientation angle of the target vehicle is determined according to the local orientation and the incident angle of the target vehicle relative to the fisheye camera. Through the above method, when a vehicle obtains the movement orientation of the vehicle through the fisheye camera, the large image deformation in the fisheye camera imaging picture no longer has a negative impact on the prediction of the vehicle movement orientation.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a method and related equipment for obtaining the vehicle orientation angle based on a fisheye camera. Background Technology

[0002] With the rapid development of autonomous driving technology, it is widely used in various modes of transportation. In the process of autonomous vehicle control, the perception of the surrounding environment and the vehicle itself is crucial. Current autonomous driving control typically requires predicting the movement and orientation of other vehicles around the vehicle, and then controlling the vehicle's autonomous driving based on the movement and orientation of these vehicles. In existing technologies, this prediction is usually done using fisheye cameras to capture images of surrounding vehicles. However, due to the inherent limitations of fisheye cameras, the images containing surrounding vehicles are often significantly distorted, affecting the subsequent prediction of their movement and orientation, resulting in low accuracy.

[0003] Therefore, how to solve the problem of low accuracy in predicting the direction of vehicle movement in the process of autonomous driving control has become a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] In view of the above problems, in order to solve the problem of low accuracy in predicting the vehicle's orientation during autonomous driving control, this application provides a method and related equipment for obtaining the vehicle orientation angle based on a fisheye camera.

[0005] The embodiments of this application disclose the following technical solutions:

[0006] In a first aspect, this application discloses a method for obtaining the vehicle orientation angle based on a fisheye camera, including:

[0007] Acquire a target image; the target image is obtained based on a fisheye camera, and the target image includes the target vehicle;

[0008] Based on the intrinsic and extrinsic parameters of the fisheye camera, the target image is subjected to distortion correction processing to obtain a distortion-corrected target image.

[0009] Based on the distortion-corrected target image and the intrinsic and extrinsic parameters of the fisheye camera, the incident angle of the target vehicle relative to the fisheye camera is determined.

[0010] The local orientation of the target vehicle in the distortion-corrected target image is determined by a preset deep learning network.

[0011] The orientation angle of the target vehicle is determined based on the local orientation and the incident angle of the target vehicle relative to the fisheye camera.

[0012] Optionally, the distortion-corrected target image is displayed as a pinhole camera image; determining the incident angle of the target vehicle relative to the fisheye camera based on the distortion-corrected target image and the intrinsic and extrinsic parameters of the fisheye camera specifically includes:

[0013] The center pixel of the target vehicle in the distortion-corrected target image is obtained; the center pixel is based on the center pixel of the pinhole camera image.

[0014] Based on the intrinsic and extrinsic parameters of the fisheye camera, the center pixel is converted into a center pixel based on the fisheye camera image.

[0015] The incident angle of the target vehicle relative to the fisheye camera is determined based on the distortion table of the fisheye camera and the center pixel of the image captured by the fisheye camera.

[0016] Optionally, determining the incident angle of the target vehicle relative to the fisheye camera based on the distortion table of the fisheye camera and the center pixel of the image captured by the fisheye camera specifically includes:

[0017] The imaging ratio of the target vehicle is calculated based on the center pixel of the fisheye camera image.

[0018] The incident angle of the target vehicle relative to the fisheye camera is determined based on the distortion table of the fisheye camera and the imaging scale value of the target vehicle.

[0019] Optionally, after determining the orientation angle of the target vehicle based on the local orientation and the incident angle of the target vehicle relative to the fisheye camera, the method further includes:

[0020] The autonomous driving mode is adjusted according to the orientation angle of the target vehicle.

[0021] Optionally, adjusting the autonomous driving mode based on the orientation angle of the target vehicle specifically includes:

[0022] If the difference between the orientation angle of the target vehicle and the orientation angle of the vehicle exceeds a preset threshold, the autonomous driving mode will be adjusted to a preset safety warning mode.

[0023] Secondly, this application discloses a vehicle orientation angle acquisition device based on a fisheye camera, comprising:

[0024] An acquisition module is used to acquire a target image; the target image is obtained based on a fisheye camera, and the target image includes a target vehicle;

[0025] The distortion correction module is used to perform distortion correction processing on the target image based on the intrinsic and extrinsic parameters of the fisheye camera to obtain a distortion-corrected target image.

[0026] An incident angle determination module is used to determine the incident angle of the target vehicle relative to the fisheye camera based on the distortion-corrected target image and the intrinsic and extrinsic parameters of the fisheye camera.

[0027] The local orientation determination module is used to determine the local orientation of the target vehicle in the distortion-corrected target image through a preset deep learning network.

[0028] The orientation angle determination module is used to determine the orientation angle of the target vehicle based on the local orientation and the incident angle of the target vehicle relative to the fisheye camera.

[0029] Optionally, the incident angle determination module is specifically used for:

[0030] The center pixel of the target vehicle in the distortion-corrected target image is obtained; the center pixel is based on the center pixel of the pinhole camera image.

[0031] Based on the intrinsic and extrinsic parameters of the fisheye camera, the center pixel is converted into a center pixel based on the fisheye camera image.

[0032] The incident angle of the target vehicle relative to the fisheye camera is determined based on the distortion table of the fisheye camera and the center pixel of the image captured by the fisheye camera.

[0033] Optionally, the incident angle determination module is specifically used for:

[0034] The imaging ratio of the target vehicle is calculated based on the center pixel of the fisheye camera image.

[0035] The incident angle of the target vehicle relative to the fisheye camera is determined based on the distortion table of the fisheye camera and the imaging scale value of the target vehicle.

[0036] Thirdly, this application discloses an electronic device, which includes: a processor, a memory, and a system bus;

[0037] The processor and the memory are connected via the system bus;

[0038] The memory is used to store one or more programs, the one or more programs including instructions that, when executed by the processor, cause the processor to perform the vehicle heading angle acquisition method based on a fisheye camera.

[0039] Fourthly, this application discloses a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the aforementioned method for obtaining the vehicle orientation angle based on a fisheye camera.

[0040] Compared to existing technologies, this application offers the following advantages: This application provides a method and related equipment for determining the vehicle orientation angle based on a fisheye camera. It involves acquiring a target image captured by a fisheye camera, including the target vehicle, and then performing distortion correction processing on the target image based on the intrinsic and extrinsic parameters of the fisheye camera to obtain a distortion-corrected target image. Then, based on the distortion-corrected target image and the intrinsic and extrinsic parameters of the fisheye camera, the incident angle of the target vehicle relative to the fisheye camera is determined. Furthermore, a pre-set deep learning network is used to determine the local orientation of the target vehicle within the distortion-corrected target image. Finally, based on the local orientation and the incident angle of the target vehicle relative to the fisheye camera, the orientation angle of the target vehicle is determined. This method ensures that when the vehicle's motion orientation is obtained using a fisheye camera, the significant image distortion in the fisheye camera image no longer negatively impacts the prediction of the vehicle's motion orientation, thus improving the accuracy of the prediction. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 A flowchart illustrating a method for determining a vehicle orientation angle based on a fisheye camera, provided in an embodiment of this application;

[0043] Figure 2 A schematic diagram of a target image that has undergone distortion correction processing, provided as an embodiment of this application;

[0044] Figure 3 A schematic diagram of a vehicle orientation angle determination device based on a fisheye camera is provided for an embodiment of this application;

[0045] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0046] As described earlier, perception of the surrounding environment and the vehicle itself is crucial in the process of autonomous driving control. Current autonomous driving control typically requires predicting the movement and orientation of other vehicles around the vehicle, and then using this prediction to control the vehicle's autonomous driving based on the movement and orientation of each vehicle. In existing technologies, this prediction is usually done using fisheye cameras to capture images of surrounding vehicles. However, due to the inherent limitations of fisheye cameras, the images containing surrounding vehicles are often significantly distorted, affecting the subsequent prediction of their movement and orientation, resulting in low accuracy.

[0047] Therefore, how to solve the problem of low accuracy in predicting the direction of vehicle movement in the process of autonomous driving control has become a technical problem that urgently needs to be solved by those skilled in the art.

[0048] To address the aforementioned issues, this application provides a method and related equipment for determining vehicle orientation angle based on a fisheye camera. The method involves acquiring a target image including the target vehicle captured by a fisheye camera, and then performing distortion correction processing on the target image based on the intrinsic and extrinsic parameters of the fisheye camera to obtain a distortion-corrected target image. Next, based on the distortion-corrected target image and the extrinsic parameters of the fisheye camera, the incident angle of the target vehicle relative to the fisheye camera is determined. Then, using a pre-set deep learning network, the local orientation of the target vehicle within the distortion-corrected target image is determined. Finally, based on the local orientation and the incident angle of the target vehicle relative to the fisheye camera, the orientation angle of the target vehicle is determined. This method ensures that when the vehicle's motion orientation is obtained using a fisheye camera, the significant image distortion in the fisheye camera image no longer negatively impacts the prediction of the vehicle's motion orientation, thus improving the accuracy of vehicle motion orientation prediction.

[0049] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0050] Method Implementation Examples

[0051] S101: Acquire a target image; the target image is obtained based on a fisheye camera, and the target image includes the target vehicle.

[0052] First, before determining the vehicle's heading angle, a target image including the target vehicle is acquired, and this target image is captured by the vehicle's fisheye camera.

[0053] A fisheye camera is a panoramic camera that can independently achieve large-area, blind-spot-free monitoring. Different types of fisheye cameras can achieve different wide-angle shooting effects. Currently, the mainstream fisheye cameras use hanging and wall mounting methods to obtain 360-degree and 180-degree monitoring effects respectively. Therefore, they are widely used in vehicle cameras to help with the autonomous driving control of vehicles.

[0054] S102: Based on the intrinsic and extrinsic parameters of the fisheye camera, the target image is subjected to distortion correction processing to obtain a distortion-corrected target image.

[0055] After acquiring the target image containing the target vehicle, distortion correction processing is performed on the target image based on the intrinsic and extrinsic parameters of the fisheye camera. The target image displayed in the form of a fisheye camera image is transformed into an image displayed in the form of a pinhole camera image. This image displayed in the form of a pinhole camera image is the target image after distortion correction processing.

[0056] S103: Determine the incident angle of the target vehicle relative to the fisheye camera based on the distorted target image and the intrinsic and extrinsic parameters of the fisheye camera.

[0057] After completing the distortion correction process on the target image and obtaining the distorted target image, it is necessary to determine the incident angle of the target vehicle relative to the vehicle's own fisheye camera in the distorted image, based on the specific intrinsic and extrinsic parameters of the fisheye camera and in conjunction with the distorted target image. Specifically, this can be achieved through the following three steps:

[0058] Step 1: Obtain the center pixel of the target vehicle in the distorted target image; the center pixel is the center pixel of the image captured by the pinhole camera.

[0059] Step 2: Based on the intrinsic and extrinsic parameters of the fisheye camera, convert the center pixel into a center pixel based on the fisheye camera image.

[0060] Step 3: Determine the incident angle of the target vehicle relative to the fisheye camera based on the distortion table of the fisheye camera and the center pixel of the image based on the fisheye camera.

[0061] When determining the incident angle of the target vehicle relative to the fisheye camera, the center pixel of the target vehicle in the distorted target image is first obtained. Because the target image is displayed in the form of a pinhole camera image after distortion correction, when calculating the incident angle of the target vehicle relative to the fisheye camera, it is necessary to convert the center pixel of the target vehicle based on the pinhole camera image into the center pixel of the audio camera image.

[0062] After obtaining the center pixel of the target vehicle based on the fisheye camera image, the incident angle of the target vehicle relative to the fisheye camera can be determined by using the center pixel and the distortion table corresponding to the fisheye camera of the vehicle.

[0063] S104: Determine the local orientation of the target vehicle in the distortion-corrected target image using a preset deep learning network.

[0064] After obtaining the distortion-corrected image, a pre-set deep learning network is used to determine the local orientation of the target vehicle within the distortion-corrected image. The concept of local orientation will be explained below; please refer to [link to relevant documentation]. Figure 2 , Figure 2 This is a schematic diagram of a target image that has undergone distortion correction processing, provided as an embodiment of this application.

[0065] exist Figure 4 In this embodiment, S1 represents the target vehicle. The target vehicle can be extracted and processed separately, and its local orientation can be observed independently of the image. Currently, the target vehicle's local orientation is biased to the right. Based on vertical angle analysis, the target vehicle should be shifted approximately 15 degrees to the right. This 15-degree rightward shift is the target vehicle's local orientation.

[0066] For a pre-defined deep learning network, before determining the local orientation, a large number of images are collected to train the network. The images used for training the network are labeled with the vehicle's specific orientation angle, the angle of incidence relative to the fisheye camera, and the vehicle's local orientation. By inputting a large number of images for training and using the vehicle's local orientation as the output data, the training of the pre-defined deep learning network is complete. After training, it is applied to a real-world scene to predict the local orientation of vehicles, thus obtaining the target vehicle's local orientation.

[0067] S105: Determine the orientation angle of the target vehicle based on the local orientation and the incident angle of the target vehicle relative to the fisheye camera.

[0068] Finally, based on the local orientation of the target vehicle and the incident angle of the target vehicle relative to the fisheye camera, the orientation angle of the target vehicle can be determined. Specifically, the orientation angle of the target vehicle can be calculated using the following formula.

[0069] θ l =θ g -α

[0070] Where θ l θ represents the orientation angle of the target vehicle. g The local orientation of the target vehicle is indicated by α, and the incident angle of the target vehicle relative to the fisheye camera is indicated by α.

[0071] As an optional implementation, after step S105, the following step is further included:

[0072] Step 1: Adjust the autonomous driving mode according to the orientation angle of the target vehicle.

[0073] After obtaining the target vehicle's heading angle, the vehicle's autonomous driving mode can be adjusted based on this angle. Specifically, a preset threshold can be set. When the difference between the target vehicle's heading angle and the vehicle's own heading angle exceeds this threshold, a collision risk is identified. In this case, the vehicle's autonomous driving mode is switched to a preset safety warning mode. Under this mode, the driving system used for autonomous driving control adjusts the vehicle's speed and heading angle in real time based on the target vehicle's speed and heading angle, effectively ensuring driving safety during autonomous driving.

[0074] As an alternative implementation, determining the incident angle of the target vehicle relative to the fisheye camera can be accomplished through the following two steps:

[0075] Step 1: Calculate the imaging ratio of the target vehicle based on the center pixel of the fisheye camera image;

[0076] Step 2: Determine the incident angle of the target vehicle relative to the fisheye camera based on the distortion table of the fisheye camera and the imaging ratio of the target vehicle.

[0077] After obtaining the center pixel of the target vehicle image captured by the fisheye camera, the incident angle of the target vehicle relative to the fisheye camera can be further determined by calculating the imaging scale value of the target vehicle. Specifically, the imaging scale value of the target vehicle can be calculated using the following formula.

[0078]

[0079] In the formula, The image scale value of the target vehicle is represented by x. t x represents the x-coordinate of the center pixel of the target vehicle, and x represents the total pixel width of the target image after distortion correction.

[0080] After obtaining the imaging scale value of the target vehicle, the angle of incidence of the target vehicle relative to the fisheye camera can be obtained by looking up the distortion table of the fisheye camera using the imaging scale value of the target vehicle.

[0081] This embodiment provides a method and related equipment for determining the vehicle orientation angle based on a fisheye camera. The method involves acquiring a target image including the target vehicle captured by a fisheye camera, and then performing distortion correction processing on the target image based on the intrinsic and extrinsic parameters of the fisheye camera to obtain a distortion-corrected target image. Then, based on the distortion-corrected target image and the intrinsic and extrinsic parameters of the fisheye camera, the incident angle of the target vehicle relative to the fisheye camera is determined. A preset deep learning network is then used to determine the local orientation of the target vehicle within the distortion-corrected target image. Finally, based on the local orientation and the incident angle of the target vehicle relative to the fisheye camera, the orientation angle of the target vehicle is determined. This method ensures that when the vehicle's motion orientation is obtained using a fisheye camera, the large image distortion in the fisheye camera image no longer negatively impacts the prediction of the vehicle's motion orientation, thus improving the accuracy of vehicle motion orientation prediction.

[0082] The following describes a vehicle orientation angle determination device based on a fisheye camera, which is provided in an embodiment of this application. The vehicle orientation angle determination device based on a fisheye camera described below and the vehicle orientation angle determination method based on a fisheye camera described above can be referred to and correspond to each other.

[0083] Device Examples

[0084] The acquisition module 100 is used to acquire a target image; the target image is obtained based on a fisheye camera, and the target image includes a target vehicle;

[0085] The distortion correction module 200 is used to perform distortion correction processing on the target image according to the internal and external parameters of the fisheye camera to obtain the distortion-corrected target image.

[0086] The incident angle determination module 300 is used to determine the incident angle of the target vehicle relative to the fisheye camera based on the distortion-corrected target image and the intrinsic and extrinsic parameters of the fisheye camera.

[0087] The local orientation determination module 400 is used to determine the local orientation of the target vehicle in the distortion-corrected target image through a preset deep learning network.

[0088] Orientation angle determination module 500 is used to determine the orientation angle of the target vehicle based on the local orientation and the incident angle of the target vehicle relative to the fisheye camera.

[0089] Optionally, the incident angle determination module is specifically used for:

[0090] The center pixel of the target vehicle in the distortion-corrected target image is obtained; the center pixel is based on the center pixel of the pinhole camera image.

[0091] Based on the intrinsic and extrinsic parameters of the fisheye camera, the center pixel is converted into a center pixel based on the fisheye camera image.

[0092] The incident angle of the target vehicle relative to the fisheye camera is determined based on the distortion table of the fisheye camera and the center pixel of the image captured by the fisheye camera.

[0093] Optionally, the incident angle determination module is specifically used for:

[0094] The imaging ratio of the target vehicle is calculated based on the center pixel of the fisheye camera image.

[0095] The incident angle of the target vehicle relative to the fisheye camera is determined based on the distortion table of the fisheye camera and the imaging scale value of the target vehicle.

[0096] Optionally, the vehicle orientation angle acquisition device based on a fisheye camera further includes:

[0097] The adjustment module is used to adjust the autonomous driving mode according to the orientation angle of the target vehicle.

[0098] Optionally, the adjustment module is specifically used for:

[0099] When the difference between the orientation angle of the target vehicle and the orientation angle of the vehicle exceeds a preset threshold, the autonomous driving mode is adjusted to a preset safety warning mode.

[0100] Electronic device examples

[0101] See Figure 4 The figure is a schematic diagram of the structure of an electronic device provided in an embodiment of this application, including:

[0102] Memory 11 is used to store computer programs;

[0103] The processor 12 is used to execute the computer program to implement the steps of the vehicle orientation angle acquisition method based on a fisheye camera as described in any of the above method embodiments.

[0104] In this embodiment, the device can be an in-vehicle computer, a PC (Personal Computer), or a terminal device such as a smartphone, tablet computer, handheld computer, or portable computer.

[0105] The device may include a memory 11, a processor 12, and a bus 13.

[0106] The memory 11 includes at least one type of readable storage medium, such as flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the device, such as the hard disk of the device. In other embodiments, the memory 11 can also be an external storage device of the device, such as a plug-in hard disk, SmartMedia Card (SMC), Secure Digital (SD) card, Flash Card, etc. Furthermore, the memory 11 can include both internal and external storage units of the device. The memory 11 can be used not only to store application software and various types of data installed on the device, such as program code for executing fault prediction methods, but also to temporarily store data that has been output or will be output.

[0107] In some embodiments, processor 12 may be a central processing unit (CPU), controller, microcontroller, microprocessor or other data processing chip, used to run program code stored in memory 11 or process data, such as program code for executing a fault prediction method.

[0108] This bus 13 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 4 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0109] Furthermore, the device may also include a network interface 14, which may optionally include a wired interface and / or a wireless interface (such as a Wi-Fi interface, a Bluetooth interface, etc.), typically used to establish communication connections between the device and other electronic devices.

[0110] Optionally, the device may further include a user interface 15, which may include a display, an input unit such as a keyboard, and optionally, a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the device and to display a visual user interface.

[0111] Figure 4 Only devices with components 11-15 are shown; those skilled in the art will understand that... Figure 4 The structure shown does not constitute a limitation on the device and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0112] It should be noted that the various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for the method apparatus, electronic device, and vehicle, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the description of the method embodiments. The method apparatus, electronic device, and vehicle described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components indicated as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of the solution in this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0113] The above description is merely one specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A fisheye camera-based vehicle orientation angle acquisition method, characterized in that, The method comprises the following steps: acquiring a target image; the target image is obtained based on a fisheye camera, and the target image comprises a target vehicle; performing de-distortion processing on the target image according to internal and external parameters of the fisheye camera to obtain a de-distortion-processed target image; determining an incident angle of the target vehicle relative to the fisheye camera according to the de-distortion-processed target image and the internal and external parameters of the fisheye camera; determining a local orientation of the target vehicle in the de-distortion-processed target image by using a preset deep learning network; determining an orientation angle of the target vehicle according to the local orientation and the incident angle of the target vehicle relative to the fisheye camera; the de-distortion-processed target image is displayed in the form of a pinhole camera imaging picture; the step of determining the incident angle of the target vehicle relative to the fisheye camera according to the de-distortion-processed target image and the internal and external parameters of the fisheye camera specifically comprises the following steps: acquiring a center pixel point of the target vehicle in the de-distortion-processed target image; the center pixel point is based on a center pixel point of the pinhole camera imaging picture; converting the center pixel point into a center pixel point based on a fisheye camera imaging picture according to the internal and external parameters of the fisheye camera; determining the incident angle of the target vehicle relative to the fisheye camera according to a distortion table of the fisheye camera and the center pixel point based on the fisheye camera imaging picture.

2. The method of claim 1, wherein, the step of determining the incident angle of the target vehicle relative to the fisheye camera according to the distortion table of the fisheye camera and the center pixel point based on the fisheye camera imaging picture specifically comprises the following steps: calculating an imaging scale value of the target vehicle according to the center pixel point based on the fisheye camera imaging picture; determining the incident angle of the target vehicle relative to the fisheye camera according to the distortion table of the fisheye camera and the imaging scale value of the target vehicle.

3. The method of claim 1, wherein, after the step of determining the orientation angle of the target vehicle according to the local orientation and the incident angle of the target vehicle relative to the fisheye camera, the method further comprises the following step: adjusting an automatic driving mode according to the orientation angle of the target vehicle.

4. The method of claim 3, wherein, the step of adjusting the automatic driving mode according to the orientation angle of the target vehicle specifically comprises the following step: if a difference between the orientation angle of the target vehicle and an orientation angle of a host vehicle exceeds a preset threshold, adjusting the automatic driving mode to a preset safety warning mode.

5. A fisheye camera based vehicle orientation angle acquisition device, characterized in that, The method comprises the following steps: an acquiring module is configured to acquire a target image; the target image is obtained based on a fisheye camera, and the target image comprises a target vehicle; a de-distortion processing module is configured to perform de-distortion processing on the target image according to internal and external parameters of the fisheye camera to obtain a de-distortion-processed target image; an incident angle determining module is configured to determine an incident angle of the target vehicle relative to the fisheye camera according to the de-distortion-processed target image and the internal and external parameters of the fisheye camera; a local orientation determining module is configured to determine a local orientation of the target vehicle in the de-distortion-processed target image by using a preset deep learning network; The orientation angle determination module is configured to determine an orientation angle of the target vehicle according to the local orientation and an incident angle of the target vehicle relative to the fisheye camera. The incident angle determination module is specifically configured to: obtain a center pixel point of the target vehicle in the target image after the distortion processing; the center pixel point is a center pixel point based on a pinhole camera imaging picture; convert the center pixel point into a center pixel point based on a fisheye camera imaging picture according to internal and external parameters of the fisheye camera; determine the incident angle of the target vehicle relative to the fisheye camera according to a distortion table of the fisheye camera and the center pixel point based on the fisheye camera imaging picture.

6. The apparatus of claim 5, wherein, The incident angle determination module is specifically configured to: calculate an imaging scale value of the target vehicle according to the center pixel point based on the fisheye camera imaging picture; determine the incident angle of the target vehicle relative to the fisheye camera according to the distortion table of the fisheye camera and the imaging scale value of the target vehicle.

7. An electronic device, comprising: The device comprises a processor, a memory, and a system bus; The processor and the memory are connected through the system bus; The memory is configured to store one or more programs, the one or more programs comprising instructions that, when executed by the processor, cause the processor to perform the fisheye camera-based vehicle orientation angle acquisition method in any one of claims 1-4.

8. A computer readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the fisheye camera-based vehicle orientation angle acquisition method in any one of claims 1-4.

Images