CN117036507A - Vehicle-mounted camera calibration system and method based on video injection - Google Patents
Vehicle-mounted camera calibration system and method based on video injection Download PDFInfo
- Publication number
- CN117036507A CN117036507A CN202311100716.4A CN202311100716A CN117036507A CN 117036507 A CN117036507 A CN 117036507A CN 202311100716 A CN202311100716 A CN 202311100716A CN 117036507 A CN117036507 A CN 117036507A
- Authority
- CN
- China
- Prior art keywords
- camera
- vehicle
- calibration
- calibrated
- controller
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000002347 injection Methods 0.000 title claims abstract description 74
- 239000007924 injection Substances 0.000 title claims abstract description 74
- 238000000034 method Methods 0.000 title claims abstract description 34
- 230000000007 visual effect Effects 0.000 claims abstract description 30
- 238000012544 monitoring process Methods 0.000 claims abstract description 22
- 238000004088 simulation Methods 0.000 claims description 32
- 230000004927 fusion Effects 0.000 claims description 28
- 230000002093 peripheral effect Effects 0.000 claims description 13
- 238000004891 communication Methods 0.000 claims description 9
- 239000011159 matrix material Substances 0.000 claims description 9
- 238000009434 installation Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 description 26
- 102100034112 Alkyldihydroxyacetonephosphate synthase, peroxisomal Human genes 0.000 description 8
- 101000799143 Homo sapiens Alkyldihydroxyacetonephosphate synthase, peroxisomal Proteins 0.000 description 8
- 238000000848 angular dependent Auger electron spectroscopy Methods 0.000 description 8
- 230000000694 effects Effects 0.000 description 7
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000033772 system development Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Studio Devices (AREA)
Abstract
The invention provides a vehicle-mounted camera calibration system and method based on video injection, wherein the system comprises a scene generating unit, a first video injection unit, a front-view camera controller, a second video injection unit, a visual box, an upper computer monitoring unit, an Ethernet converter and a to-be-calibrated controller; the scene generating unit is provided with two output branches, one is a first video injection unit and a front-view camera controller which are sequentially connected, and the other is a second video injection unit and a visual box which are sequentially connected; the front-view camera controller is in wireless connection with the controller to be calibrated through a vehicle-mounted Ethernet, and the vision box is also electrically connected with the controller to be calibrated; the controller to be calibrated and the vision box are both in wireless connection with an Ethernet converter through a vehicle-mounted Ethernet, and the output end of the Ethernet converter is electrically connected with the upper computer monitoring unit; the invention can improve the calibration accuracy of the vehicle-mounted camera, effectively reduce the calibration cost and has the universality and the flexibility.
Description
Technical Field
The invention relates to the technical field of vehicle-mounted camera calibration, in particular to a vehicle-mounted camera calibration system and method based on video injection.
Background
As the development of intelligent automobiles becomes more and more advanced, more or less advanced driving assistance systems (Advanced Driver Assistance System, ADAS) have been installed in various factories. While various sensors are basically mounted on ADAS products, cameras are widely used in ADAS-mounted vehicles because of their low price and small weather influence. How to ensure the reliability and safety of the ADAS is a key issue that needs to be concerned, and thus, the test on the ADAS is an essential part, and the test on the ADAS currently mainly includes two means, i.e., a real-vehicle test and a Hardware-in-Loop (HIL) test. Generally, in the system development stage, a controller of a vehicle with an ADAS function is repeatedly tested and verified on an HIL test bench until each function test of the controller is stable and reliable, and then is formally assembled on a real vehicle for testing.
The camera is used as an important sensor carried on a real vehicle, and the camera is required to be calibrated before the real vehicle is put off line for production. The calibration of the real vehicle camera is generally carried out in a standard calibration room, each calibration plate is shot by adjusting the camera, and the internal parameters and the external parameters of the camera are continuously adjusted. The calibration of the real vehicle camera requires that an operator repeatedly adjusts the real vehicle, the camera or the calibration plate to achieve a satisfactory calibration result, and has long time consumption, complex operation and higher cost for building a calibration room.
The prior art discloses a camera emulation test system, includes: the system comprises an upper computer, a video signal acquisition sub-board, a programmable logic gate array (Field-Programmable Gate Array, FPGA) video processing module, a camera signal analog output sub-board and a measured controller; generating a simulation camera video signal of the virtual camera through the vision software of the upper computer, processing and outputting the simulation camera video signal generated by the upper computer to a tested controller to test by a video signal acquisition sub-board, an FPGA video processing module and a camera signal simulation output sub-board, and directly transmitting the simulation camera video signal generated by the upper computer to the tested controller to test by the camera simulation test system; however, in the system in the prior art, only the camera data is acquired in a simulation way, the parameters of the simulation camera are not adjusted and set, and the real vehicle is matched, so that the acquired camera data format is not matched with or not matched with the camera data on the real vehicle, the simulation test result is affected, and the simulation precision and the robustness are low.
The prior art also discloses a camera calibration method and device based on the video injection scheme, wherein the method comprises the following steps: constructing a calibration scene and a target; constructing a controller area network (Controller Area Network, CAN) model; injecting calibration data into a controller by a video injection method; running a simulation environment and starting a calibration process; although the camera calibration method proposed by the prior art calibrates the cameras based on video injection, the cameras are calibrated only through 4 checkerboard targets, the information of all directions of a vehicle cannot be covered on the whole, the generated panoramic image has a blind area, meanwhile, fusion images of simulation scenes shot by all cameras are not generated, and the calibration precision is low.
Disclosure of Invention
The invention aims to provide a vehicle-mounted camera calibration system based on video injection, which is used for solving the defects of time and labor waste, low efficiency and low calibration precision of the vehicle-mounted camera calibration in the prior art; the second aim is to provide a vehicle-mounted camera calibration method based on video injection.
In order to solve the technical problems, the technical scheme of the invention is as follows:
a video injection-based vehicle camera calibration system, comprising: the device comprises a scene generating unit, a first video injection unit, a front-view camera controller, a to-be-calibrated controller, an Ethernet converter and an upper computer monitoring unit;
the front-view camera controller is used for carrying out simulation calibration on the front-view camera in the vehicle-mounted camera;
the scene generating unit, the first video injection unit and the front-view camera controller are sequentially connected, and the front-view camera controller is in wireless connection with the controller to be calibrated through the vehicle-mounted Ethernet;
the controller to be calibrated is in wireless connection with the Ethernet converter through the vehicle-mounted Ethernet, and the output end of the Ethernet converter is electrically connected with the upper computer monitoring unit.
Preferably, the system further comprises a second video injection unit and a visual box which are connected in sequence, wherein the visual box is used for carrying out simulation calibration on other cameras in the vehicle-mounted camera;
the input end of the second video injection unit is electrically connected with the scene generation unit, and the output end of the visual box is electrically connected with the controller to be calibrated;
the first video injection unit, the front-view camera controller, the second video injection unit and the visual box form two parallel output branches of the scene generating unit;
the vision box is also connected with the Ethernet converter in a wireless way through the vehicle-mounted Ethernet.
Preferably, the other cameras comprise a rear view camera and a peripheral view camera; the peripheral camera comprises a left front camera, a left rear camera, a right front camera and a right rear camera.
Preferably, the communication protocol between the vision box and the controller to be calibrated is a CAN communication protocol.
Preferably, the scene generating unit is specifically a first computer with scene simulation software built in; the upper computer monitoring unit is specifically a second computer with monitoring software;
the scene simulation software is IPG active and CarMaker software;
the monitoring software is Matrix Client software.
The invention also provides a vehicle-mounted camera calibration method based on video injection, which is based on the vehicle-mounted camera calibration system based on video injection and comprises the following steps:
s1: the method comprises the steps of obtaining a real vehicle model to be calibrated, generating a real image of a calibration room in the scene generating unit, and building the calibration room, wherein the ground and all walls of the calibration room are provided with a plurality of calibration plates, and the size of the calibration room is matched with the proportion of the real vehicle model to be calibrated;
s2: placing the real vehicle model to be calibrated in the central position of a calibration room, and initializing parameters of a vehicle-mounted camera on the real vehicle model to be calibrated in the scene generating unit;
s3: shooting the calibration room by using the front-view camera and other cameras after initializing parameters to obtain a front-view image and a peripheral-view image;
s4: inputting a front view image into a controller to be calibrated through the first video injection unit and a front view camera controller, and inputting a peripheral view image into the controller to be calibrated through the second video injection unit and a visual box;
the controller to be calibrated fuses the front view image and the peripheral view image to obtain a fusion image between calibrations;
s5: inputting the fusion image between the calibrations into an upper computer monitoring unit through the Ethernet converter, comparing the fusion image with the real image between the calibrations, judging whether the fusion image between the calibrations meets the preset condition, if not, repeating the steps S2-S4, and calibrating again after adjusting the parameters of the vehicle-mounted camera; if the parameters of the vehicle-mounted camera are met, the parameters of the vehicle-mounted camera are stored to serve as the optimal parameters of the vehicle-mounted camera, and simulation calibration of the vehicle-mounted camera is completed.
Preferably, in the step S1, each calibration plate includes a plurality of two-dimensional code pictures arranged in a matrix;
each two-dimensional code picture is a unique picture which is randomly generated.
Preferably, the parameters of the vehicle-mounted camera in the step S2 include an internal parameter and an external parameter of the vehicle-mounted camera;
the internal parameters comprise a focal length and a visual field size, and the external parameters comprise an installation position and a pitching angle;
the size of the vehicle-mounted camera is matched with the proportion of the real vehicle model to be calibrated.
Preferably, the second video injection unit and the vision box are closed when only the front-view camera is calibrated using the method.
The invention also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the above method.
Compared with the prior art, the technical scheme of the invention has the beneficial effects that:
the invention provides a vehicle-mounted camera calibration system and method based on video injection, which comprises the steps of firstly obtaining a real vehicle model to be calibrated, generating a real image of a calibration room in a scene generating unit and building the calibration room; placing the real vehicle model to be calibrated in the central position of a calibration room, and initializing parameters of a vehicle-mounted camera on the real vehicle model to be calibrated in the scene generating unit; shooting the calibration room by using the front-view camera and other cameras after initializing parameters to obtain a front-view image and a peripheral-view image; inputting a front view image into a controller to be calibrated through the first video injection unit and a front view camera controller, and inputting a peripheral view image into the controller to be calibrated through the second video injection unit and a visual box; the controller to be calibrated fuses the front view image and the peripheral view image to obtain a fusion image between calibrations; inputting the fusion image between the calibration areas into an upper computer monitoring unit through the Ethernet converter, comparing the fusion image with the real image between the calibration areas, judging whether a preset condition is met, if not, repeating the steps, and calibrating again after adjusting the parameters of the vehicle-mounted camera; if the parameters of the vehicle-mounted camera are met, the parameters of the vehicle-mounted camera are saved as the optimal parameters of the vehicle-mounted camera, and simulation calibration of the vehicle-mounted camera is completed;
according to the invention, 6 paths of images are shot through the calibrated vehicle-mounted camera and panoramic image fusion is carried out, so that visual and accurate driving-assisting image information is obtained, and the calibration accuracy is improved; meanwhile, the invention can simulate specific scenes and scenes with safety risks in actual life, thereby improving the testing efficiency and guaranteeing the personal safety of the testers; in addition, the same scene can be repeatedly tested for multiple times, so that the robustness of the system is effectively improved, the resources of the real vehicle test are greatly saved, and the workload of the real vehicle tester is reduced.
Drawings
Fig. 1 is a schematic diagram of a vehicle-mounted camera calibration system based on video injection according to embodiment 1.
Fig. 2 is a flowchart of a video injection-based vehicle-mounted camera calibration method provided in embodiment 2.
Fig. 3 is a graph of the actual vehicle inter-calibration effect provided in example 2.
Fig. 4 is a layout of the floor calibration plate between calibration provided in example 2.
Fig. 5 is a layout of the calibration plates of the left and right walls of the calibration room provided in example 2.
Fig. 6 is a layout of the calibration plates of the front wall surface and the rear wall surface of the calibration room provided in example 2.
Fig. 7 is a two-dimensional code image of the ground calibration plate and the wall calibration plate provided in embodiment 2.
Fig. 8 is a drawing showing the production of the floor calibration plate according to example 2.
Fig. 9 is a drawing of the wall surface calibration plate provided in example 2.
FIG. 10 is a plot of the resulting inter-calibration effects provided in example 2.
FIG. 11 is a fusion image between calibrations provided in example 2.
Fig. 12 is a simulated panoramic fusion image provided in example 2.
Detailed Description
The drawings are for illustrative purposes only and are not to be construed as limiting the present patent;
for the purpose of better illustrating the embodiments, certain elements of the drawings may be omitted, enlarged or reduced and do not represent the actual product dimensions;
it will be appreciated by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The technical scheme of the invention is further described below with reference to the accompanying drawings and examples.
Example 1
The embodiment provides a vehicle-mounted camera calibration system based on video injection, which comprises: the device comprises a scene generating unit, a first video injection unit, a front-view camera controller, a to-be-calibrated controller, an Ethernet converter and an upper computer monitoring unit;
the front-view camera controller is used for carrying out simulation calibration on the front-view camera in the vehicle-mounted camera;
the scene generating unit, the first video injection unit and the front-view camera controller are sequentially connected, and the front-view camera controller is in wireless connection with the controller to be calibrated through the vehicle-mounted Ethernet;
the controller to be calibrated is in wireless connection with the Ethernet converter through the vehicle-mounted Ethernet, and the output end of the Ethernet converter is electrically connected with the upper computer monitoring unit;
the system also comprises a second video injection unit and a visual box which are connected in sequence, wherein the visual box is used for carrying out simulation calibration on other cameras in the vehicle-mounted camera;
the input end of the second video injection unit is electrically connected with the scene generation unit, and the output end of the visual box is electrically connected with the controller to be calibrated;
the first video injection unit, the front-view camera controller, the second video injection unit and the visual box form two parallel output branches of the scene generating unit;
the visual box is also in wireless connection with the Ethernet converter through the vehicle-mounted Ethernet;
the other cameras comprise a rear-view camera and a peripheral-view camera; the peripheral camera comprises a left front camera, a left rear camera, a right front camera and a right rear camera;
the communication protocol between the visual box and the controller to be calibrated is a CAN communication protocol;
the scene generating unit is specifically a first computer with scene simulation software built in; the upper computer monitoring unit is specifically a second computer with monitoring software;
the scene simulation software is IPG active and CarMaker software;
the monitoring software is Matrix Client software.
In a specific implementation process, as shown in fig. 1, the vehicle-mounted camera includes a front-view camera and other cameras; the front-view camera controller is used for carrying out simulation calibration on the front-view camera, and the visual box is used for carrying out simulation calibration on other cameras, so that the system can carry out simulation calibration on cameras with 6 directions in total in the front, back and peripheral 4 sides at the same time;
the scene generating unit is used for generating various scenes and a calibration room, the first video injection unit and the second video injection unit are video injection boxes, and an external 12V power supply is used; the data of the 2 video injection boxes are input through software in a scene generating unit, in the embodiment, the scene generating unit is a computer with built-in scene simulation software, two independent display cards are arranged, and the scene simulation software is IPG active and CarMaker software;
the scene generating unit is provided with two output branches, one branch is a first video injection unit and a front-view camera controller which are sequentially connected, and the other branch is a second video injection unit and a visual box which are sequentially connected; the 2 video injection boxes are connected with the scene generation unit through HDMI high definition video lines, 6 HDMI lines are needed for the 6 cameras, and a main screen for test control is additionally added, so that 7 HDMI lines are needed in total;
the front-view camera controller is in wireless connection with the controller to be calibrated through a vehicle-mounted Ethernet;
the visual box is also electrically connected with the controller to be calibrated, and the communication protocol between the visual box and the controller to be calibrated is CAN communication protocol; after video data is input into the video injection box, the video data is transmitted to a controller to be calibrated through an LVDS cable after internal conversion;
the controller to be calibrated and the vision box are both in wireless connection with an Ethernet converter through a vehicle-mounted Ethernet, and the output end of the Ethernet converter is electrically connected with the upper computer monitoring unit; the upper computer monitoring unit is a computer with built-in monitoring software, the monitoring software is Matrix Client software and is used for monitoring video data in the controller and checking the data state of the 6-path cameras and the specific effect of the fused image;
in the calibration process of the vehicle-mounted camera, after being analyzed and processed by 2 video injection boxes, video signals are respectively input into a front-view camera controller and a visual box, video information is respectively written into the to-be-calibrated controller through vehicle-mounted Ethernet and CAN communication, finally, monitoring of video data is carried out through an upper computer monitoring unit, parameters of the vehicle-mounted camera are adjusted according to monitoring results, and finally calibration is completed;
in addition, the system can not only simulate and calibrate the high-power ADAS (comprising a front-view camera, a rear-view camera and a peripheral-view camera), but also independently simulate and calibrate the front-view camera, and only the vision box and the second video injection unit are required to be shielded at the moment, so that the system has compatibility and flexibility;
according to the system, 6 paths of images are shot through the calibrated vehicle-mounted camera and panoramic image fusion is carried out, visual and accurate driving-assisting image information is obtained, and calibration accuracy is improved; meanwhile, the system can simulate specific scenes and scenes with safety risks in actual life, so that the testing efficiency is improved, and the personal safety of testing personnel is ensured; in addition, the same scene can be repeatedly tested for multiple times, so that the robustness of the system is effectively improved, the resources of the real vehicle test are greatly saved, and the workload of the real vehicle tester is reduced.
Example 2
As shown in fig. 2, the present embodiment provides a video injection-based vehicle-mounted camera calibration method, which is based on the video injection-based vehicle-mounted camera calibration system described in embodiment 1, and includes the following steps:
s1: the method comprises the steps of obtaining a real vehicle model to be calibrated, generating a real image of a calibration room in the scene generating unit, and building the calibration room, wherein the ground and all walls of the calibration room are provided with a plurality of calibration plates, and the size of the calibration room is matched with the proportion of the real vehicle model to be calibrated;
s2: placing the real vehicle model to be calibrated in the central position of a calibration room, and initializing parameters of a vehicle-mounted camera on the real vehicle model to be calibrated in the scene generating unit;
s3: shooting the calibration room by using the front-view camera and other cameras after initializing parameters to obtain a front-view image and a peripheral-view image;
s4: inputting a front view image into a controller to be calibrated through the first video injection unit and a front view camera controller, and inputting a peripheral view image into the controller to be calibrated through the second video injection unit and a visual box;
the controller to be calibrated fuses the front view image and the peripheral view image to obtain a fusion image between calibrations;
s5: inputting the fusion image between the calibrations into an upper computer monitoring unit through the Ethernet converter, comparing the fusion image with the real image between the calibrations, judging whether the fusion image between the calibrations meets the preset condition, if not, repeating the steps S2-S4, and calibrating again after adjusting the parameters of the vehicle-mounted camera; if the parameters of the vehicle-mounted camera are met, the parameters of the vehicle-mounted camera are saved as the optimal parameters of the vehicle-mounted camera, and simulation calibration of the vehicle-mounted camera is completed;
in the step S1, each calibration board includes a plurality of two-dimensional code pictures arranged in a matrix;
each two-dimensional code picture is a unique picture which is randomly generated;
the parameters of the vehicle-mounted camera in the step S2 comprise internal parameters and external parameters of the vehicle-mounted camera;
the internal parameters comprise a focal length and a visual field size, and the external parameters comprise an installation position and a pitching angle;
the size of the vehicle-mounted camera is matched with the proportion of the real vehicle model to be calibrated;
when the method is used for calibrating the front-view camera only, the second video injection unit and the vision box are closed.
In the specific implementation process, firstly, a real vehicle model to be calibrated is obtained, a real image of a calibration room is generated in the scene generating unit, and the calibration room is built;
the scene simulation software mainly applied in the embodiment is IPG active and CarMaker, the actual real vehicle calibration room is used as a template to generate in the IPG CarMaker-Scenario Editor, and the effect diagram of the real vehicle calibration room is shown in FIG. 3;
the method comprises the steps that a plurality of calibration plates are arranged on the ground and all walls of a calibration room, each calibration plate is formed by arranging different two-dimensional code pictures one by one according to a certain sequence, a whole image is generated by arranging the two-dimensional code pictures according to specific size requirements, the whole image is imported into an IPG (internet protocol) CarMaker-Scenario Editor, and the positions and the inclination angles of simulation calibration plates are set according to the proportion between the simulation calibration plates and a real vehicle so as to achieve the consistency with the real vehicle proportion; each two-dimensional code on the calibration wall and the calibration floor has global uniqueness, the two-dimensional codes on the calibration wall and the calibration floor are arranged according to a certain sequence, and the position of each two-dimensional code is uniquely fixed and consistent with the calibration plate of the real vehicle;
as shown in fig. 4 to 10, fig. 4 is a layout of a ground calibration plate between calibration rooms, fig. 5 is a layout of calibration plates of a left wall surface and a right wall surface between calibration rooms, fig. 6 is a layout of calibration plates of a front wall surface and a rear wall surface between calibration rooms, fig. 7 is a two-dimensional code picture of the ground calibration plate and the wall surface calibration plate, fig. 8 is a manufacturing drawing of the ground calibration plate, fig. 9 is a manufacturing drawing of the wall surface calibration plate, and fig. 10 is a simulation inter-calibration effect drawing with the same proportion as a real vehicle which is built in an IPG Movie;
after the simulation calibration room is built, placing the real vehicle model to be calibrated at the central position of the calibration room, initializing parameters of the vehicle-mounted cameras on the real vehicle model to be calibrated in the scene generating unit, and adjusting the installation positions and pitching angles of the front-view camera, the rear-view camera, the left front camera, the left rear camera, the right front camera and the right rear camera to achieve that the shot pictures are consistent with pictures shot by the real vehicle; the pictures shot during real vehicle calibration can be imported into an IPG active, the transparency is set to be 30, and then the angles of 6 paths of cameras are respectively adjusted, so that each path of shot pictures are overlapped with the actual pictures shot by the cameras on the real vehicle;
after the installation position and the pitching angle of the camera are adjusted, 6 paths of pictures of the camera are switched out from the IPG active, and the 6 paths of pictures are cut into a controller to be calibrated; the method comprises the steps that the picture information of 6 paths of cameras is monitored and checked through Matrix Client software in an upper computer monitoring unit, the 6 paths of pictures are guaranteed to correspond to front view, rear view, left front view, left rear view, right front view and right rear view one by one respectively, and fusion failure can be caused by any path of missing or wrong corresponding relation;
after the position of the camera is adjusted, starting a CarMaker, running a Testrun function, starting to transmit data to a video injection box, and checking a fused image at a Matrix Client after data fusion is obtained; adjusting the view angle of the corresponding camera by observing the fusion effect of the fusion image;
after image data of 6 cameras in the simulation calibration room are transmitted to a video injection box, the 6 images are spliced and fused to generate a 360-degree calibration room fusion image, and as shown in fig. 11, the image comprises 6 azimuth image information, namely 6 azimuth information of front, rear, left front, left rear, right front and right rear in sequence, so that all information around a vehicle to be calibrated can be completely input into a controller to be calibrated, and the effect of simulating surrounding environment information shot by a camera sensor on a real vehicle is achieved;
after all cameras are calibrated, the calibrated vehicle-mounted cameras are utilized to acquire scene information around the vehicle, the scene information is input into a controller to be calibrated, real-time scene fusion images of 6 paths of cameras are shown in fig. 12, simulation tests are carried out in the mode, and the real-time scene fusion images are closer to actual driving scenes; the simulation test system improves the video black box scheme applied before, is closer to an actual driving scene, has more comprehensive scene data simulation, smaller limitation and more credible test result, is easier to test the problems in the driving scene, and is more suitable for ADAS automatic test;
in addition, when the method in the embodiment is used for only calibrating the front-view camera, only the front-view camera is required to be reduced, and other cameras, the second video injection unit and the vision box are closed;
according to the method, 6 paths of images are shot through the calibrated vehicle-mounted camera and panoramic image fusion is carried out, visual and accurate driving-assisting image information is obtained, and calibration accuracy is improved; meanwhile, the method can simulate specific scenes and scenes with safety risks in actual life, so that the testing efficiency is improved, and the personal safety of testing personnel is ensured; in addition, the same scene can be repeatedly tested for multiple times, so that the robustness of the system is effectively improved, the resources of the real vehicle test are greatly saved, and the workload of the real vehicle tester is reduced.
The same or similar reference numerals correspond to the same or similar components;
the terms describing the positional relationship in the drawings are merely illustrative, and are not to be construed as limiting the present patent;
it is to be understood that the above examples of the present invention are provided by way of illustration only and not by way of limitation of the embodiments of the present invention. Other variations or modifications of the above teachings will be apparent to those of ordinary skill in the art. It is not necessary here nor is it exhaustive of all embodiments. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the invention are desired to be protected by the following claims.
Claims (10)
1. A video injection-based vehicle camera calibration system, comprising: the device comprises a scene generating unit, a first video injection unit, a front-view camera controller, a to-be-calibrated controller, an Ethernet converter and an upper computer monitoring unit;
the front-view camera controller is used for carrying out simulation calibration on the front-view camera in the vehicle-mounted camera;
the scene generating unit, the first video injection unit and the front-view camera controller are sequentially connected, and the front-view camera controller is in wireless connection with the controller to be calibrated through the vehicle-mounted Ethernet;
the controller to be calibrated is in wireless connection with the Ethernet converter through the vehicle-mounted Ethernet, and the output end of the Ethernet converter is electrically connected with the upper computer monitoring unit.
2. The vehicle-mounted camera calibration system based on video injection according to claim 1, further comprising a second video injection unit and a vision box which are sequentially connected, wherein the vision box is used for performing simulation calibration on other cameras in the vehicle-mounted camera;
the input end of the second video injection unit is electrically connected with the scene generation unit, and the output end of the visual box is electrically connected with the controller to be calibrated;
the first video injection unit, the front-view camera controller, the second video injection unit and the visual box form two parallel output branches of the scene generating unit;
the vision box is also connected with the Ethernet converter in a wireless way through the vehicle-mounted Ethernet.
3. The video injection-based on-board camera calibration system of claim 2, wherein the other cameras include a rear-view camera and a peripheral-view camera; the peripheral camera comprises a left front camera, a left rear camera, a right front camera and a right rear camera.
4. A video injection based vehicle camera calibration system according to claim 3, wherein the communication protocol between the vision box and the controller to be calibrated is the CAN communication protocol.
5. The video injection-based vehicle camera calibration system according to claim 4, wherein the scene generating unit is a first computer with scene simulation software built in; the upper computer monitoring unit is specifically a second computer with monitoring software;
the scene simulation software is IPG active and CarMaker software;
the monitoring software is Matrix Client software.
6. A video injection-based vehicle-mounted camera calibration method based on the video injection-based vehicle-mounted camera calibration system as claimed in any one of claims 2 to 5, characterized by comprising the following steps:
s1: the method comprises the steps of obtaining a real vehicle model to be calibrated, generating a real image of a calibration room in the scene generating unit, and building the calibration room, wherein the ground and all walls of the calibration room are provided with a plurality of calibration plates, and the size of the calibration room is matched with the proportion of the real vehicle model to be calibrated;
s2: placing the real vehicle model to be calibrated in the central position of a calibration room, and initializing parameters of a vehicle-mounted camera on the real vehicle model to be calibrated in the scene generating unit;
s3: shooting the calibration room by using the front-view camera and other cameras after initializing parameters to obtain a front-view image and a peripheral-view image;
s4: inputting a front view image into a controller to be calibrated through the first video injection unit and a front view camera controller, and inputting a peripheral view image into the controller to be calibrated through the second video injection unit and a visual box;
the controller to be calibrated fuses the front view image and the peripheral view image to obtain a fusion image between calibrations;
s5: inputting the fusion image between the calibrations into an upper computer monitoring unit through the Ethernet converter, comparing the fusion image with the real image between the calibrations, judging whether the fusion image between the calibrations meets the preset condition, if not, repeating the steps S2-S4, and calibrating again after adjusting the parameters of the vehicle-mounted camera; if the parameters of the vehicle-mounted camera are met, the parameters of the vehicle-mounted camera are stored to serve as the optimal parameters of the vehicle-mounted camera, and simulation calibration of the vehicle-mounted camera is completed.
7. The method for calibrating a vehicle-mounted camera based on video injection according to claim 6, wherein in the step S1, each calibration plate comprises a plurality of two-dimensional code pictures arranged in a matrix;
each two-dimensional code picture is a unique picture which is randomly generated.
8. The method for calibrating an in-vehicle camera based on video injection according to claim 7, wherein the parameters of the in-vehicle camera in step S2 include in-vehicle camera parameters and out-vehicle camera parameters;
the internal parameters comprise a focal length and a visual field size, and the external parameters comprise an installation position and a pitching angle;
the size of the vehicle-mounted camera is matched with the proportion of the real vehicle model to be calibrated.
9. The video injection-based on-vehicle camera calibration method of claim 8, wherein the second video injection unit and the visual box are closed when only the front-view camera is calibrated using the method.
10. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, carries out the steps of the method according to any one of claims 6 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311100716.4A CN117036507A (en) | 2023-08-30 | 2023-08-30 | Vehicle-mounted camera calibration system and method based on video injection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311100716.4A CN117036507A (en) | 2023-08-30 | 2023-08-30 | Vehicle-mounted camera calibration system and method based on video injection |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117036507A true CN117036507A (en) | 2023-11-10 |
Family
ID=88635369
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311100716.4A Pending CN117036507A (en) | 2023-08-30 | 2023-08-30 | Vehicle-mounted camera calibration system and method based on video injection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117036507A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117893618A (en) * | 2024-01-11 | 2024-04-16 | 长春众鼎科技有限公司 | Advanced driving assistance system camera software calibration and training |
-
2023
- 2023-08-30 CN CN202311100716.4A patent/CN117036507A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117893618A (en) * | 2024-01-11 | 2024-04-16 | 长春众鼎科技有限公司 | Advanced driving assistance system camera software calibration and training |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109413415B (en) | Camera controller testing system and testing method | |
| US20220136930A1 (en) | System for testing intelligent vehicles | |
| CN108319259B (en) | Test system and test method | |
| CN216145095U (en) | ADAS simulation test system | |
| CN112987703A (en) | System and method for developing and testing in-loop automatic driving of whole vehicle in laboratory | |
| CN117036507A (en) | Vehicle-mounted camera calibration system and method based on video injection | |
| CN111026099A (en) | Intelligent networking automobile function simulation test system and test method | |
| CN112422936B (en) | Vehicle panoramic image calibration system, method, server and storage medium | |
| CN113484851A (en) | Simulation test system and method for vehicle-mounted laser radar and complete vehicle in-loop test system | |
| CN110389041B (en) | System and method for testing vehicle look-around and ultrasonic integrated system | |
| CN115016323A (en) | Automatic driving simulation test system and method | |
| CN113704119B (en) | Test method, device and system for intelligent driving and storage medium | |
| CN110244687A (en) | A kind of test macro of camera controller | |
| CN113706964A (en) | Intelligent driving teaching training system and automatic driving vehicle control method | |
| CN116107288A (en) | Intelligent driving simulation test system and method thereof | |
| CN112698582A (en) | ADAS ECU simulation test method and system | |
| CN113835410A (en) | Finished automobile testing method and system based on signal simulation device | |
| CN115880142A (en) | Image generation method and device of trailer, storage medium and terminal | |
| CN113689695B (en) | Method and system for data acquisition, visualization and calibration of roadside sensing system | |
| ES2886957T3 (en) | An automotive test system, method and software product | |
| Grapinet et al. | Characterization and simulation of optical sensors | |
| CN114488047A (en) | Vehicle sensor calibration system | |
| CN112711245A (en) | Calibration method and device for automobile diagnosis equipment and computing equipment | |
| CN110580722A (en) | Calibration test method and calibration test device for panoramic system and vehicle | |
| Bedjiah et al. | Methodology to Validate the Radiated Immunity of Very Complex Systems by a Succession of Simple Component Radiated Immunity Tests at System Level |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |