Method and system for determining pitch angle and yaw angle of vehicle-mounted camera

Abstract

The invention relates to a vehicle-mounted camera correction technology, and discloses a method and system for determining a pitch angle and a yaw angle of a vehicle-mounted camera, and the method comprises the steps: obtaining an ADAS camera and a wheel image, and obtaining the ADAS camera and a rear wheel image through arranging an external camera; parameters of an ADAS camera and a rear wheel image are extracted, and parameters of the obtained ADAS camera and the rear wheel image are extracted; the pitch angle theta pitch and the deviation angle theta yaw are determined, the pitch angle and the yaw angle are calculated according to the extracted image parameters of the ADAS camera and the rear wheel, and therefore the pitch angle theta pitch and the deviation angle theta yaw of the ADAS camera are determined. According to the method for determining the azimuth angle of the vehicle camera through imaging of the external camera, the azimuth angle of the ADAS camera is accurately and efficiently determined, and an invasive calibration process is not needed; and the calibration process of the ADAS technology is further optimized.

Description

Technical Field

[0001] The present invention relates to a vehicle-mounted camera correction technology, and in particular to a method and system for determining the pitch angle and yaw angle of a vehicle-mounted camera. Background Art

[0002] With the development of smart cars, automatic assisted driving technology has become increasingly important. Among them, on-board cameras are a very important part of the field of automatic assisted driving, and the calibration of on-board cameras is crucial to restore the three-dimensional information of objects in the acquired two-dimensional images. There is a one-to-one correspondence between the spatial points in the camera's imaging model and the image points on the image plane, and this correspondence is usually determined by the camera's parameters, including intrinsic parameters and extrinsic parameters. The intrinsic parameters are the camera's own parameters and generally do not need to be calibrated, while the extrinsic parameters mainly include the camera's pitch angle, rotation angle and yaw angle.

[0003] Accurately determining the orientation angle of a vehicle's cameras is crucial for optimizing ADAS features, including lane departure warning, collision avoidance, and autonomous driving systems. Traditional methods for calibrating and determining the orientation angle of these cameras typically require specialized equipment or invasive calibration procedures, which are time-consuming and costly.

[0004] For example, prior art 1: CN201811289485.5 discloses parking a vehicle equipped with the vehicle-mounted camera at a predetermined location; receiving a calibration start instruction; taking a photo of a target using the vehicle-mounted camera, performing calibration calculations based on the photos; and if the calibration is successful, automatically writing the calibration calculation results into a configuration file.

[0005] There is a need for a non-intrusive, efficient, and accurate solution to determine the pitch and yaw angles of a vehicle camera using external imaging technology. Observing the geometry of the camera lens and vehicle components using an external camera can provide a solution to this problem. Summary of the Invention

[0006] The present invention addresses the problem that methods for determining the azimuth angle of a camera in the prior art usually require dedicated equipment or invasive calibration procedures, which are time-consuming and costly. A method and system for determining the pitch and yaw angles of a vehicle-mounted camera are provided.

[0007] In order to solve the above technical problems, the present invention is solved by the following technical solutions:

[0008] A method for determining a pitch angle and a yaw angle of a vehicle-mounted camera, wherein the vehicle-mounted camera is a forward-facing ADAS camera installed on a vehicle; the method comprises:

[0009] Acquisition of ADAS camera and wheel images, by setting up an external camera to acquire ADAS camera and rear wheel images;

[0010] Extracting parameters of the ADAS camera and rear wheel images, and extracting parameters from the acquired ADAS camera and rear wheel images;

[0011] The pitch angle θ_pitch and the yaw angle θ_yaw are determined by calculating the pitch angle and the yaw angle based on the extracted ADAS camera and rear wheel image parameters, thereby determining the pitch angle θ_pitch and the yaw angle θ_yaw of the ADAS camera.

[0012] Preferably, the parameters for extracting the ADAS camera and rear wheel images include:

[0013] Acquire the edges of the ADAS camera and rear wheel images using edge detection algorithms;

[0014] Extracting the closed contour of the ADAS camera and rear wheels: extracting the closed contour of the edges of the acquired ADAS camera and rear wheel images to obtain the closed contour of the ADAS camera and rear wheels;

[0015] Ellipse fitting: fitting the closed contour of the ADAS camera and rear wheels using the least squares method;

[0016] Obtaining the ellipse parameters, obtaining the ellipse parameters for the fitted reunion.

[0017] Preferably, the ellipse parameters include the central coordinate axis (x, y), the major axis length a, the minor axis length b and the direction angle φ;

[0018] Where x is the horizontal axis of the central coordinate axis, y is the vertical axis of the central coordinate axis, a is the length of the major axis, b is the length of the minor axis, and φ is the angle between the major axis and the horizontal axis of the image.

[0019] Preferably, the pitch angle θ_pitch and the yaw angle θ_yaw are determined as follows:

[0020] The pitch angle θ_pitch and the yaw angle θ_yaw are determined by the parameters of the ADAS camera and the rear wheel image;

[0021] The pitch angle θ_pitch is determined by the long axis length a and the short axis length b of the ADAS camera and the rear wheel image;

[0022]

[0023] For the deflection angle θ_yaw, θ_yaw is determined by the angle φ between the long axis and the horizontal axis of the image;

[0024] θ_yaw=φ。

[0025] Preferably, the wheel image is an image of a rear wheel of the vehicle.

[0026] Preferably, the method further includes calibration of a camera, and before determining the pitch angle θ_pitch and the yaw angle θ_yaw, calibration of an external camera is performed.

[0027] Preferably, the camera is calibrated by performing perspective calibration on the external camera through perspective transformation.

[0028] In order to solve the above technical problems, the present invention also provides a system for determining the pitch angle and yaw angle of a vehicle-mounted camera, wherein the vehicle-mounted camera is a forward-facing ADAS camera installed on a vehicle; the system comprises:

[0029] An ADAS camera and wheel image acquisition module is configured to acquire images of the ADAS camera and rear wheels by setting an external camera;

[0030] Extract ADAS camera and rear wheel image parameter module, extract parameters from the acquired ADAS camera and rear wheel images;

[0031] The pitch angle θ_pitch and yaw angle determination module calculates the pitch angle and yaw angle based on the extracted ADAS camera and rear wheel image parameters, thereby determining the pitch angle θ_pitch and yaw angle θ_yaw of the ADAS camera.

[0032] Preferably, a camera calibration module is further included to calibrate the external camera before determining the pitch angle θ_pitch and the yaw angle θ_yaw.

[0033] The present invention has significant technical effects due to the adoption of the above technical solutions:

[0034] The present invention uses a method for determining the azimuth angle of a vehicle camera through imaging with an external camera to accurately and efficiently determine the azimuth angle of an ADAS camera without the need for an invasive calibration process, further optimizing the calibration process of ADAS technology.

[0035] The present invention can determine the azimuth angle of the vehicle-mounted camera without requiring physical access to the ADAS camera or the interior of the vehicle.

[0036] The present invention can quickly determine the azimuth angle without complicated settings.

[0037] The present invention utilizes geometric properties and precise calculations to achieve reliable results.

[0038] The present invention is applicable to a variety of vehicle types and camera configurations. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 This is a diagram showing the position of the external camera relative to the vehicle according to the present invention.

[0040] Figure 2 This is a lens diagram showing an ADAS camera in an external camera according to the present invention.

[0041] Figure 3 This is a lens diagram showing the vehicle wheels through an external camera according to the present invention.

[0042] Figure 4 It is a detailed view of the lens or wheel ellipse of the present invention. DETAILED DESCRIPTION

[0043] The present invention is further described in detail below with reference to the accompanying drawings and embodiments.

[0044] Example 1

[0045] A method for determining a pitch angle and a yaw angle of a vehicle-mounted camera, wherein the vehicle-mounted camera is a forward-facing ADAS camera installed on a vehicle; the method comprises:

[0046] Acquisition of ADAS camera and wheel images, by setting up an external camera to acquire ADAS camera and rear wheel images;

[0047] Extracting parameters of the ADAS camera and rear wheel images, and extracting parameters from the acquired ADAS camera and rear wheel images;

[0048] The pitch angle θ_pitch and the yaw angle θ_yaw are determined by calculating the pitch angle and the yaw angle based on the extracted ADAS camera and rear wheel image parameters, thereby determining the pitch angle θ_pitch and the yaw angle θ_yaw of the ADAS camera.

[0049] Figure 1 In [1], the method involves positioning an external camera in a position to capture an image of the ADAS camera lens and the wheel on one side of the vehicle. Due to the effect of perspective distortion, the lens and wheel appear as ellipses in the image.

[0050] By analyzing the geometric properties of these ellipses, specifically their axis lengths and orientations, the pitch and yaw angles of the ADAS camera can be calculated. The vehicle's true heading is determined by analyzing the ellipse of the non-steering rear wheels as a reference.

[0051] Parameters for extracting ADAS camera and rear wheel images include:

[0052] Before edge detection, the image needs to be preprocessed by using a noise reduction filter to enhance the image contrast and brightness, thus obtaining a preprocessed image.

[0053] Acquire the edges of the ADAS camera and rear wheel images. Use the edge detection algorithm to obtain the edges of the ADAS camera and rear wheel images after preprocessing.

[0054] Extracting the closed contour of the ADAS camera and rear wheels: extracting the closed contour of the edges of the acquired ADAS camera and rear wheel images to obtain the closed contour of the ADAS camera and rear wheels;

[0055] Ellipse fitting: fitting the closed contour of the ADAS camera and rear wheels using the least squares method;

[0056] Obtaining the ellipse parameters, obtaining the ellipse parameters for the fitted reunion.

[0057] The ellipse parameters include the central coordinate axis (x, y), the major axis length a, the minor axis length b and the direction angle φ;

[0058] Where x is the horizontal axis of the central coordinate axis, y is the vertical axis of the central coordinate axis, a is the length of the major axis, b is the length of the minor axis, and φ is the angle between the major axis and the horizontal axis of the image.

[0059] Determination of pitch angle θ_pitch and yaw angle θ_yaw:

[0060] The pitch angle θ_pitch and the yaw angle θ_yaw are determined by the parameters of the ADAS camera and the rear wheel image;

[0061] The pitch angle θ_pitch is determined by the long axis length a and the short axis length b of the ADAS camera and the rear wheel image;

[0062]

[0063] For the deflection angle θ_yaw, θ_yaw is determined by the angle φ between the long axis and the horizontal axis of the image;

[0064] θ_yaw=φ。

[0065] The wheel image is the image of the vehicle's rear wheels. Because the vehicle's rear wheels do not turn during operation, they define the vehicle's true direction of travel. Figure 2 In the image captured by the external camera, the multiple lenses of the ADAS camera appear as ellipses due to perspective distortion. Figure 3 In the figure, the wheels on one side of the vehicle are also shown as ovals.

[0066] Example 2

[0067] Based on Example 1, Figure 4In this embodiment, the calibration of the camera is also included. Before the pitch angle θ_pitch and the yaw angle θ_yaw are determined, the calibration of the external camera is performed.

[0068] Camera calibration: Perspective calibration of the external camera is performed using a perspective transformation. Before calculations, the external camera is calibrated using known calibration techniques and calibration patterns to compensate for lens distortion. Perspective correction: If necessary, a perspective transformation is applied to compensate for any tilt due to the external camera's position.

[0069] Example 3

[0070] This embodiment is a system for determining the pitch angle and yaw angle of a vehicle-mounted camera, where the vehicle-mounted camera is a forward-facing ADAS camera installed on a vehicle; the system includes:

[0071] An ADAS camera and wheel image acquisition module is configured to acquire images of the ADAS camera and rear wheels by setting an external camera;

[0072] Extract ADAS camera and rear wheel image parameter module, extract parameters from the acquired ADAS camera and rear wheel images;

[0073] The pitch angle θ_pitch and yaw angle determination module calculates the pitch angle and yaw angle based on the extracted ADAS camera and rear wheel image parameters, thereby determining the pitch angle θ_pitch and yaw angle θ_yaw of the ADAS camera.

[0074] Example 4

[0075] Based on the third embodiment, this embodiment further includes a camera calibration module, which performs calibration of an external camera before determining the pitch angle θ_pitch and the yaw angle θ_yaw.

Claims

1. A method for determining the pitch and yaw angles of a vehicle-mounted camera, wherein the vehicle-mounted camera is a forward-facing ADAS camera mounted on a vehicle; The methods include: Acquisition of ADAS camera and wheel images, by setting up an external camera to acquire ADAS camera and wheel images; Extracting parameters of the ADAS camera and rear wheel images, and extracting parameters from the acquired ADAS camera and rear wheel images; The pitch angle θ_pitch and the yaw angle θ_yaw are determined by calculating the pitch angle and the yaw angle based on the extracted ADAS camera and rear wheel image parameters, thereby determining the pitch angle θ_pitch and the yaw angle θ_yaw of the ADAS camera.

2. The method for determining the pitch angle and yaw angle of a vehicle-mounted camera according to claim 1, wherein: Parameters for extracting ADAS camera and rear wheel images include: Acquire the edges of the ADAS camera and rear wheel images using edge detection algorithms; Extracting the closed contour of the ADAS camera and rear wheels: extracting the closed contour of the edges of the acquired ADAS camera and rear wheel images to obtain the closed contour of the ADAS camera and rear wheels; Ellipse fitting: fitting the closed contour of the ADAS camera and rear wheels using the least squares method; Obtaining the ellipse parameters, obtaining the ellipse parameters for the fitted reunion.

3. The method for determining the pitch angle and yaw angle of a vehicle-mounted camera according to claim 2, wherein: The ellipse parameters include the central coordinate axis (x, y), the major axis length a, the minor axis length b and the direction angle φ; Where x is the horizontal axis of the central coordinate axis, y is the vertical axis of the central coordinate axis, a is the length of the major axis, b is the length of the minor axis, and φ is the angle between the major axis and the horizontal axis of the image.

4. The method for determining the pitch angle and yaw angle of a vehicle-mounted camera according to claim 3, wherein: Determination of pitch angle θ_pitch and yaw angle θ_yaw: The pitch angle θ_pitch and the yaw angle θ_yaw are determined by the parameters of the ADAS camera and the rear wheel image; The pitch angle θ_pitch is determined by the long axis length a and the short axis length b of the ADAS camera and the rear wheel image; For the deflection angle θ_yaw, θ_yaw is determined by the angle φ between the long axis and the horizontal axis of the image; θ_yaw=φ。 5. The method for determining the pitch angle and yaw angle of a vehicle-mounted camera according to claim 1, wherein: The wheel image is an image of the rear wheel of the vehicle.

6. The method for determining the pitch angle and yaw angle of a vehicle-mounted camera according to claim 1, wherein: It also includes camera calibration. Before determining the pitch angle θ_pitch and the yaw angle θ_yaw, the external camera is calibrated.

7. The method for determining the pitch angle and yaw angle of a vehicle-mounted camera according to claim 6, characterized in that: Camera calibration, perspective calibration of the external camera is performed through perspective transformation.

8. A system for determining the pitch and yaw angles of a vehicle-mounted camera, wherein the vehicle-mounted camera is a forward-facing ADAS camera installed on a vehicle; characterized in that: include: An ADAS camera and wheel image acquisition module is configured to acquire images of the ADAS camera and rear wheels by setting an external camera; Extract ADAS camera and rear wheel image parameter module, extract parameters from the acquired ADAS camera and rear wheel images; The pitch angle θ_pitch and yaw angle determination module calculates the pitch angle and yaw angle based on the extracted ADAS camera and rear wheel image parameters, thereby determining the pitch angle θ_pitch and yaw angle θ_yaw of the ADAS camera.

9. The method for determining the pitch angle and yaw angle of a vehicle-mounted camera according to claim 8, wherein: It also includes a camera calibration module, which calibrates the external camera before determining the pitch angle θ_pitch and the yaw angle θ_yaw.

Images