Method and system for dynamically calibrating intelligent vehicle lamp

Abstract

The invention discloses an intelligent vehicle lamp dynamic calibration method and system, and the method comprises the steps: obtaining a to-be-projected image, and carrying out the preprocessing of the to-be-projected image: carrying out the correction of the to-be-projected image through employing a projection image correction parameter if there is a projection image correction parameter in a memory, and then generating a pseudo-random stripe basic image; if the projection image correction parameter does not exist in the memory, directly generating a pseudo-random stripe basic image; and linearly superposing an image to be projected onto the pseudo-random stripe basic image, and projecting the superposed composite image. According to the invention, the to-be-projected image is linearly superposed on the pseudo-random stripe basic image to project the superposed image, the projection image of the pseudo-random stripe basic image is collected, and the projection image correction parameter is obtained according to the projection image of the pseudo-random stripe basic image to correct the to-be-projected image, so that a better information image projection effect is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent vehicle lamp applications, and in particular to a method and system for dynamic calibration of intelligent vehicle lamps. Background Art

[0002] With the development of headlight technology, car companies and related enterprises have made many useful attempts to project images on the road through headlights. The projection of headlights on the road surface not only plays the role of illuminating the road surface, but also can display information such as the distance to the vehicle in front, the vehicle's passable area, the vehicle's planned trajectory, and the vehicle's avoidance of pedestrians. It has become a tool for two-way communication between drivers or self-driving vehicles and surrounding pedestrians, cyclists, drivers and other traffic participants, improving the safety and interactivity of vehicle operation.

[0003] A method and a motor vehicle for illuminating a lane area by projecting a trajectory, wherein the trajectory route is projected onto the lane as part of the light distribution, and the brightness distribution corresponding to the light distribution is irradiated onto the lane, and the light distribution is used to indicate not only the trajectory but also the acceleration of the motor vehicle parallel to its longitudinal direction. This method only considers the vehicle's own trajectory planning and the projection of the light, but ignores factors such as the ups and downs and slopes that may exist on the road, resulting in position deformation or displacement of the image projected onto the ground, leading to guidance errors or misunderstandings, and reducing the use effect of the smart headlights. Summary of the invention

[0004] The object of the present invention is to provide a method and system for dynamic calibration of intelligent vehicle lights, so as to at least partially solve the above-mentioned problems in the prior art.

[0005] To achieve the above object, the present invention provides a method for dynamic calibration of intelligent vehicle lights, comprising:

[0006] Step S1, obtaining an image to be projected, and performing preprocessing on the image to be projected: if there are projection image correction parameters in the memory, the image to be projected is corrected using the projection image correction parameters, and then step S2 is executed; if there are no projection image correction parameters in the memory, step S2 is directly executed;

[0007] Step S2, generating a pseudo-random stripe basic image; wherein the size of the pseudo-random stripe basic image is not less than the size of the image to be projected;

[0008] Step S3, linearly superimposing the image to be projected onto the pseudo-random stripe basic image, and projecting the superimposed composite image;

[0009] Step S4, collecting the projected live image, demodulating the projection of the pseudo-random fringe image, analyzing the projection of the pseudo-random fringe image and the deformation of the pseudo-random fringe image, and obtaining the projection image correction parameter;

[0010] Step S5, storing the projection image correction parameters in the memory. If the projection image correction parameters already exist in the memory, the projection image correction parameters obtained in step S4 are used to update the existing projection image correction parameters.

[0011] In one embodiment, the method may further include: generating an image to be projected according to a vehicle operating state.

[0012] Among them, in step S1, using the projection image correction parameter to correct the projection image includes:

[0013] According to the time when the projection image correction parameters are generated and the running speed of the vehicle, the displacement of the vehicle is calculated, and the projection image correction parameters are updated;

[0014] The image to be projected is corrected using the updated projection image correction parameters.

[0015] In one embodiment, the step S1 of correcting the projection image using the projection image correction parameter includes:

[0016] According to the time when the projection image correction parameter is generated, the time when the image to be projected is received, and the running speed of the vehicle, the displacement of the vehicle is calculated, and the projection image correction parameter is updated;

[0017] The image to be projected is corrected using the updated projection image correction parameters.

[0018] In one embodiment, correcting the projection image using the updated projection image correction parameters includes:

[0019] The spline interpolation distortion algorithm is used to construct the surface parameters of the perspective transformation of the image to be projected. The obtained homography matrix is ​​multiplied pixel by pixel with the image to be projected to achieve rotation and translation and generate a corrected projection image.

[0020] In one embodiment, the size of the pseudo-random stripe basic image is the same as the size of the image to be projected.

[0021] In one embodiment, the pseudo-random stripe base image is not perceptible to the human eye and includes horizontal and / or vertical stripes.

[0022] In one implementation, the stripe width value of the pseudo-random stripe base image is a pseudo-random number between 0.2 m and 0.5 m.

[0023] The present invention also provides a system for dynamic calibration of intelligent vehicle lights, which is used to execute the method for dynamic calibration of intelligent vehicle lights described in any of the above embodiments, and is characterized by comprising: a vehicle light, a calibration controller, and a camera; wherein:

[0024] The calibration controller is used to: obtain the image to be projected, and perform preprocessing on the image to be projected: if there are projection image correction parameters in the memory, use the projection image correction parameters to correct the image to be projected, and then generate a pseudo-random stripe basic image; if there are no projection image correction parameters in the memory, directly generate the pseudo-random stripe basic image; and linearly superimpose the image to be projected on the pseudo-random stripe basic image; wherein the size of the pseudo-random stripe basic image is not less than the size of the image to be projected;

[0025] The vehicle lights are used to: project the superimposed composite image;

[0026] The camera is used to: collect projected live images;

[0027] The calibration controller is also used to: demodulate the projection of the pseudo-random stripe image according to the projected live image captured by the camera, analyze the projection of the pseudo-random stripe image and the deformation of the pseudo-random stripe image, and obtain projection image correction parameters; store the projection image correction parameters in a memory; if the projection image correction parameters already exist in the memory, update the existing projection image correction parameters.

[0028] In one embodiment, the system further includes an image generating device for generating an image to be projected according to the operating status of the vehicle.

[0029] The present invention adopts the above technical solution, which has the following advantages:

[0030] The image to be projected is linearly superimposed on the pseudo-random stripe basic image to project the superimposed image, and the projection image of the pseudo-random stripe basic image is collected. The projection image correction parameters are obtained according to the projection image of the pseudo-random stripe basic image to correct the image to be projected, so as to achieve a better information image projection effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A schematic diagram of the structure of a system for dynamic calibration of intelligent vehicle lights provided by the present invention.

[0032] Figure 2 A schematic flow chart of a method for dynamic calibration of intelligent vehicle lights provided by the present invention. DETAILED DESCRIPTION

[0033] In the drawings, the same or similar reference numerals are used to represent the same or similar elements or elements with the same or similar functions. The embodiments of the present invention are described in detail below in conjunction with the drawings.

[0034] The automotive projection lighting display system based on visual processing and sensor technology uses image visual acquisition and detection and multi-sensor signal acquisition modules as input, projection lighting display modules as output, and the vehicle-mounted microprocessor control unit as the core. According to the vehicle's traffic conditions and combined with the vehicle's lighting working mode, it sends out corresponding intelligent lighting drive signals, and adjusts the brightness, graphics, and images of the projection lighting display module in real time and adaptively to achieve lamp projection display. The light projection lighting status monitoring module is used to monitor the operation status.

[0035] Figure 1 This is a schematic diagram of the structure of a system for dynamic calibration of intelligent vehicle lights provided by the present invention. Figure 1 As shown, the intelligent vehicle lamp dynamic calibration system provided in this embodiment includes a vehicle lamp 10, a calibration controller 20, and a camera 30; wherein,

[0036] The calibration controller 20 is used to: obtain the image to be projected, and perform preprocessing on the image to be projected: if there are projection image correction parameters in the memory, use the projection image correction parameters to correct the image to be projected, and then generate a pseudo-random stripe basic image; if there are no projection image correction parameters in the memory, directly generate the pseudo-random stripe basic image; and linearly superimpose the image to be projected on the pseudo-random stripe basic image; wherein the size of the pseudo-random stripe basic image is not less than the size of the image to be projected;

[0037] The vehicle lamp 10 is used to: project the superimposed composite image;

[0038] The camera 30 is used to: collect projected live images;

[0039] The calibration controller 20 is also used to: demodulate the projection of the pseudo-random stripe image according to the projected live image captured by the camera, analyze the projection of the pseudo-random stripe image and the deformation of the pseudo-random stripe image, and obtain projection image correction parameters; store the projection image correction parameters in a memory; if the projection image correction parameters already exist in the memory, update the existing projection image correction parameters.

[0040] In one embodiment, the system further includes an image generating device 40 for generating an image to be projected according to the vehicle operation state. The image generating device is, for example, an onboard computer or a control unit. The onboard computer or control unit mainly realizes the generation of vehicle trajectory and risk analysis results according to the driver's operation control or automatic driving perception, prediction, planning and control, and generates original projection image information according to the safety strategy, such as the planned trajectory of the car, the stop line, etc. At this time, the calibration controller 20 obtains the original projection image information from the onboard computer or the control unit.

[0041] In one embodiment, the calibration controller 20 includes a computing unit, a communication interface unit, etc., which are used to obtain the original projection image information and correct the projection image in combination with the projection image correction parameters; generate a pseudo-random stripe basic image; superimpose the projection image on the pseudo-random stripe basic image to generate a composite image; receive the scene image, demodulate the pseudo-random stripe image, analyze the deformation of the pseudo-random stripe image, and its relationship with the ground geometric features to obtain the projection image correction parameters.

[0042] In one embodiment, the vehicle lamp 10 is used to project the composite image onto the lane and the surrounding ground to display various information to pedestrians, cyclists and drivers. The preset camera is used to collect the live image of the graphic projection area in front of the vehicle and the surrounding area. The image information includes the lane and surrounding environment information and various projected information, which is digitized and sent to the calibration controller.

[0043] In other embodiments, the image to be projected may also be generated by the vehicle lamp 10. In this case, the vehicle lamp 10 may generate the original projection image information according to the vehicle operation status information transmitted by the onboard computer and in accordance with the safety strategy.

[0044] Figure 2 A flowchart of a method for dynamic calibration of intelligent vehicle lights provided by an embodiment of the present invention is shown in FIG. Figure 2 As shown, a method for dynamic calibration of intelligent vehicle lights provided by an embodiment of the present invention may include:

[0045] Step S1, obtaining an image to be projected, and performing preprocessing on the image to be projected.

[0046] The preprocessing includes: if there are projection image correction parameters in the memory, correcting the image to be projected by using the projection image correction parameters, and then executing step S2; if there are no projection image correction parameters in the memory, directly executing step S2.

[0047] In this step, the calibration controller obtains the image to be projected, searches the memory for projection image correction parameters, and performs correction if they exist, otherwise directly executes step S2. The memory may be an independent device or built into other devices, for example, the memory may be built into the image calibration controller.

[0048] In addition, before step S1, the method may further include: generating an image to be projected according to the running state of the vehicle.

[0049] For example, the vehicle light controller obtains the vehicle running status information and generates the corresponding image to be projected. The image calibration controller can be a separate controller or the same controller as the vehicle light controller.

[0050] The headlight controller is, for example, an onboard computer or control unit, which generates vehicle trajectory and risk analysis results based on driver operation control or autonomous driving perception, prediction, planning and control, and generates the original image to be projected according to the safety strategy.

[0051] When there are no projection image correction parameters due to reasons such as the first startup of the system or the shutdown of the dynamic calibration system, there is no need to correct the original projection image information, and the process is directly forwarded to the next step S2.

[0052] In one embodiment, using the projection image correction parameter to correct the projection image includes:

[0053] The displacement of the vehicle is calculated by combining the time when the projection image correction parameters are generated, the time when the original projection image information is received after being generated, and the instantaneous speed of the vehicle. The position of the projection image correction parameters is offset and extrapolated according to the actual position of the vehicle to generate the projection image correction parameters after position correction.

[0054] The calibration controller corrects the image to be projected in combination with the correction parameters of the projected image after position correction, and uses the spline interpolation distortion algorithm to construct the surface parameters of the perspective transformation of the projected image, and multiplies the obtained homography matrix with the image to be projected pixel by pixel to achieve rotation and translation, and generate a corrected projection image. If the road in front of the vehicle is in a raised state, the image will be longitudinally compressed or the image will become closer when the graphics are projected onto the raised part of the ground. Therefore, the image needs to be stretched to ensure that the image is not deformed or shifted when projected onto the ground. In particular, when the projected image is a curve with actual geometric significance such as a planned trajectory of a car or a stop line, the adjustment of the response can avoid causing misjudgment of pedestrians, cyclists and drivers. In the embodiment of the present invention, the entire image is corrected instead of being processed in blocks to avoid jumps in the boundaries of different blocks.

[0055] Step S2, generating a pseudo-random stripe basic image.

[0056] The calibration controller generates stripes of varying brightness and colors according to the settings. The changes in the stripes are not easily noticeable to the human eye. For example, they are composed of light other than visible light or their intensity is weak. This does not affect the acquisition of image information by pedestrians, cyclists and drivers, but can be extracted through subsequent processing.

[0057] The stripe type may include horizontal and vertical stripes, or other types such as circular stripes, to adapt to the direction of deformation of the object to be sensed. For road scenes, linear stripes may be preferably used. In addition, the stripe type may also include a combination of stripes in different directions, such as a combination of horizontal and vertical stripes to form a grid stripe.

[0058] The width of the stripe is a pseudo-random number, and its value is between 0.2m and 0.5m (settable) when projected onto the ground. The horizontal and vertical stripes are generated with this width value, which are used to extract the horizontal and vertical correction parameters respectively. And because the stripe width changes dynamically, the correction parameters at different positions can be obtained by accumulating data at different time points, so as to make corrections more complete and comprehensive.

[0059] The size of the pseudo-random stripe basic image is not less than the size of the image to be projected. Preferably, the size of the pseudo-random stripe basic image is consistent with the size of the projected image.

[0060] Step S3, linearly superimposing the image to be projected onto the pseudo-random stripe basic image, and projecting the superimposed composite image.

[0061] The calibration controller linearly superimposes the projected image and the pseudo-random stripe base image of the same size to generate a composite image, and sends the composite image to the smart headlight, which projects the image onto the lane and surrounding ground to display various information to pedestrians, cyclists, and drivers.

[0062] Step S4, collecting the projected on-site image, demodulating the projection of the pseudo-random fringe image, analyzing the projection of the pseudo-random fringe image and the deformation of the pseudo-random fringe image, and obtaining the projection image correction parameters.

[0063] The camera on the car collects live images of the graphic projection area in front of the vehicle and its surroundings. The image information includes lane and surrounding environment information and various projected information, which is digitized and sent to the calibration controller.

[0064] After receiving the captured image, the calibration controller obtains stripes with larger gradients of brightness or color changes by comparing and calculating the horizontal and vertical brightness and color changes. It then makes a difference between the original generated pseudo-random stripe basic image and the obtained stripes, and calculates the positional deviations of the horizontal and vertical stripes at different points, thereby extracting the deformation of the pseudo-random stripe image, including changes in stripe position, parallelism, distortion, etc.

[0065] In one embodiment, to simplify the processing, the longitudinal deviation is identified by the longitudinal stripes, and the lateral deviation is identified by the lateral stripes. At the same time, the stripes are compared with ground geometric features with accurate coordinate values, such as lane lines, road edges, detected vehicle edges, etc. Through comprehensive analysis, the deformation or displacement of the image at different positions can be obtained, that is, the projection image correction parameters can be obtained.

[0066] Step S5, storing the projection image correction parameters in the memory. If the projection image correction parameters already exist in the memory, the projection image correction parameters obtained in step S4 are used to update the existing projection image correction parameters.

[0067] After step S5, if the vehicle continues to project, it returns to step S1 and repeats steps S1-S5 until the vehicle is turned off or the smart headlights are manually turned off.

[0068] By adopting the system and method for dynamic calibration of intelligent vehicle lights provided by the present invention, the image to be projected is linearly superimposed on the pseudo-random stripe basic image to project the superimposed image, and the projection image of the pseudo-random stripe basic image is collected. The projection image correction parameters are obtained according to the projection image of the pseudo-random stripe basic image to correct the image to be projected, thereby achieving a better information image projection effect.

[0069] Finally, it should be pointed out that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Those skilled in the art should understand that the technical solutions described in the above embodiments may be modified, or some of the technical features thereof may be replaced by equivalents; these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for dynamic calibration of intelligent vehicle lights, characterized in that: include: Step S1, obtaining an image to be projected, and performing preprocessing on the image to be projected: if there are projection image correction parameters in the memory, the image to be projected is corrected using the projection image correction parameters, and then step S2 is performed; If the projection image correction parameter does not exist in the memory, directly execute step S2; Step S2, generating a pseudo-random stripe basic image; wherein the size of the pseudo-random stripe basic image is not less than the size of the image to be projected; Step S3, linearly superimposing the image to be projected onto the pseudo-random stripe basic image, and projecting the superimposed composite image; Step S4, collecting the projected live image, demodulating the projection of the pseudo-random fringe image, analyzing the projection of the pseudo-random fringe image and the deformation of the pseudo-random fringe image, and obtaining the projection image correction parameter; Step S5, storing the projection image correction parameters in the memory. If the projection image correction parameters already exist in the memory, the projection image correction parameters obtained in step S4 are used to update the existing projection image correction parameters.

2. The method for dynamic calibration of intelligent vehicle lights according to claim 1, characterized in that: Also includes: The image to be projected is generated according to the vehicle operating status.

3. The method for dynamic calibration of intelligent vehicle lights according to claim 1, characterized in that: In step S1, using the projection image correction parameter to correct the projection image includes: According to the time when the projection image correction parameters are generated and the running speed of the vehicle, the displacement of the vehicle is calculated, and the projection image correction parameters are updated; The image to be projected is corrected using the updated projection image correction parameters.

4. The method for dynamic calibration of intelligent vehicle lights according to claim 1, characterized in that: In step S1, using the projection image correction parameter to correct the projection image includes: According to the time when the projection image correction parameter is generated, the time when the image to be projected is received, and the running speed of the vehicle, the displacement of the vehicle is calculated, and the projection image correction parameter is updated; The image to be projected is corrected using the updated projection image correction parameters.

5. The method for dynamic calibration of intelligent vehicle lights according to claim 3 or 4, characterized in that: Correcting the projected image using the updated projection image correction parameters includes: The spline interpolation distortion algorithm is used to construct the surface parameters of the perspective transformation of the image to be projected. The obtained homography matrix is ​​multiplied pixel by pixel with the image to be projected to achieve rotation and translation and generate a corrected projection image.

6. The method for dynamic calibration of intelligent vehicle lights according to claim 1, characterized in that: The size of the pseudo-random stripe base image is the same as the size of the image to be projected.

7. The method for dynamic calibration of intelligent vehicle lights according to any one of claims 1 to 4, characterized in that: The pseudo-random stripe base image is not perceptible to the human eye and includes horizontal and / or vertical stripes.

8. The method for dynamic calibration of intelligent vehicle lights as claimed in claim 7, characterized in that: The stripe width value of the pseudo-random stripe base image is a pseudo-random number between 0.2m and 0.5m.

9. A system for dynamic calibration of intelligent vehicle lights, used to execute the method for dynamic calibration of intelligent vehicle lights according to any one of claims 1 to 8, characterized in that: include: A headlight, a calibration controller, and a camera; wherein: The calibration controller is used to: obtain the image to be projected, and perform preprocessing on the image to be projected: if there are projection image correction parameters in the memory, use the projection image correction parameters to correct the image to be projected, and then generate a pseudo-random stripe basic image; if there are no projection image correction parameters in the memory, directly generate the pseudo-random stripe basic image; and linearly superimpose the image to be projected on the pseudo-random stripe basic image; wherein the size of the pseudo-random stripe basic image is not less than the size of the image to be projected; The vehicle lights are used to: project the superimposed composite image; The camera is used to: collect projected live images; The calibration controller is also used to: demodulate the projection of the pseudo-random stripe image according to the projected live image captured by the camera, analyze the projection of the pseudo-random stripe image and the deformation of the pseudo-random stripe image, and obtain projection image correction parameters; store the projection image correction parameters in a memory; if the projection image correction parameters already exist in the memory, update the existing projection image correction parameters.

10. The system for dynamic calibration of intelligent vehicle lights according to claim 9, characterized in that: It also includes an image generating device for generating an image to be projected according to the running state of the vehicle.

Images