WO2019170066A1 - 一种车载相机外参确定方法、装置、设备及系统 - Google Patents
一种车载相机外参确定方法、装置、设备及系统 Download PDFInfo
- Publication number
- WO2019170066A1 WO2019170066A1 PCT/CN2019/076915 CN2019076915W WO2019170066A1 WO 2019170066 A1 WO2019170066 A1 WO 2019170066A1 CN 2019076915 W CN2019076915 W CN 2019076915W WO 2019170066 A1 WO2019170066 A1 WO 2019170066A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- marker
- camera
- coordinate system
- positional relationship
- vehicle
- Prior art date
Links
Images
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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30204—Marker
- G06T2207/30208—Marker matrix
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30248—Vehicle exterior or interior
- G06T2207/30252—Vehicle exterior; Vicinity of vehicle
Definitions
- the present application relates to the field of assisted driving technology, and in particular, to a method, device, device and system for determining an external reference of an in-vehicle camera.
- the assisted driving system usually includes a plurality of in-vehicle cameras disposed at different positions of the vehicle body to collect images in different directions; and the images of the different directions are stitched into a large viewing angle by using parameters of the plurality of in-vehicle cameras The image is displayed to the driver and can be used to assist the driver.
- the parameters of the on-board camera include the external reference, and the external reference is the positional relationship between the on-board camera and the vehicle body.
- the solution for acquiring the external camera of the vehicle camera generally includes: parking the vehicle at the designated position, and setting a marker at another designated position near the vehicle, and predetermining a positional relationship between the parking position of the vehicle and the position of the marker; setting in the vehicle
- Each of the in-vehicle cameras is aligned with the marker for image acquisition, and the external reference of the on-board camera is calculated based on the position of the marker in the acquired image and the predetermined positional relationship.
- the parking position of the vehicle and the position of the marker are fixed. If the parking position of the vehicle is slightly deviated from the designated position, the accuracy of the acquired external reference is low.
- the purpose of the embodiments of the present application is to provide a method, device, device and system for determining an external reference of an on-vehicle camera to improve the accuracy of obtaining an external reference.
- an embodiment of the present application provides a method for determining an external parameter of an in-vehicle camera, including:
- a positional relationship between the plurality of on-vehicle cameras and the vehicle body is determined.
- an embodiment of the present application further provides an in-vehicle camera external parameter determining apparatus, including:
- An acquisition module configured to acquire a calibration image corresponding to each vehicle camera; wherein, the calibration image corresponding to each vehicle camera is obtained according to the image captured by the vehicle camera, and the calibration image corresponding to the adjacent vehicle camera includes the same marker;
- An identification module configured to identify a position of the marker in the calibration image corresponding to the onboard camera for each onboard camera
- a first determining module configured to determine, according to the location, a positional relationship between the onboard camera and the identified marker
- a conversion module configured to convert the plurality of on-vehicle cameras into the same coordinate system based on a determined position relationship between each on-vehicle camera and the marker;
- a second determining module configured to determine a positional relationship between the plurality of onboard cameras and the vehicle body in the same coordinate system.
- an embodiment of the present application further provides an electronic device, including a processor and a memory;
- a memory for storing a computer program
- the processor when used to execute a program stored on the memory, implements any of the above-described methods for determining the external parameters of the on-vehicle camera.
- the embodiment of the present application further provides a computer readable storage medium, where the computer readable storage medium stores a computer program, and when the computer program is executed by the processor, implements any of the above-mentioned vehicle camera external parameters. Determine the method.
- an embodiment of the present application further provides an executable program code for being executed to execute any of the above-described on-board camera external parameter determining methods.
- an embodiment of the present application further provides an assisted driving system, including: a processing device and a plurality of onboard cameras;
- Each in-vehicle camera for transmitting the acquired image to the processing device;
- the processing device is configured to obtain, according to the images collected by the plurality of on-board cameras, a calibration image corresponding to the plurality of on-vehicle cameras; wherein the calibration image corresponding to the adjacent on-vehicle camera includes the same marker;
- An in-vehicle camera that recognizes a position of a marker in a calibration image corresponding to the on-vehicle camera, and determines a positional relationship between the on-vehicle camera and the recognized marker according to the position; each of the on-vehicle camera and the marker according to the determined
- a positional relationship between the plurality of onboard cameras is converted into the same coordinate system based on one marker; and in the same coordinate system, a positional relationship between the plurality of onboard cameras and the vehicle body is determined.
- the calibration image corresponding to the adjacent vehicle camera includes the same marker, and for each vehicle camera, the position of the marker is identified in the calibration image corresponding to the vehicle camera, and the location is determined according to the location
- the positional relationship between the in-vehicle camera and the identified tag, the determined positional relationship includes the positional relationship between the same tag and the different in-vehicle cameras, so that a plurality of in-vehicle cameras can be converted based on one marker
- the positional relationship between the plurality of on-vehicle cameras and the vehicle body is determined in the same coordinate system.
- FIG. 1 is a schematic flowchart of a method for determining an external parameter of an in-vehicle camera according to an embodiment of the present application
- FIG. 2 is a schematic diagram of imaging provided by an embodiment of the present application.
- FIG. 3 is a schematic diagram of an application scenario provided by an embodiment of the present application.
- FIG. 4 is a schematic diagram of a vehicle body position according to an embodiment of the present application.
- FIG. 5 is a schematic structural diagram of an external camera determining device for an in-vehicle camera according to an embodiment of the present disclosure
- FIG. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure.
- FIG. 7 is a schematic structural diagram of an assist driving system according to an embodiment of the present application.
- an embodiment of the present application provides a method, device, device, and system for determining an external reference of an in-vehicle camera.
- the method can be applied to a processing device communicatively coupled to a plurality of in-vehicle cameras, or can be applied to any one of the in-vehicle cameras that are communicatively coupled to other in-vehicle cameras.
- a plurality of markers may be disposed in the vicinity of the vehicle, and the image captured by the adjacent vehicle camera includes the same marker, so that the plurality of vehicle cameras can be converted into the same coordinate system based on one marker.
- the positional relationship between the plurality of on-vehicle cameras and the vehicle body is determined. It can be seen that in this solution, only the adjacent vehicle camera is required to collect the same marker, and it is not necessary to fix the parking position of the vehicle and the position of the marker, thereby improving the accuracy of obtaining the external reference.
- FIG. 1 is a schematic flowchart of a method for determining an external parameter of an in-vehicle camera according to an embodiment of the present application, including:
- S101 Acquire a calibration image corresponding to each onboard camera.
- the calibration image corresponding to each on-vehicle camera is obtained according to the image acquired by the on-vehicle camera, and the calibration image corresponding to the adjacent on-vehicle camera includes the same marker.
- the original image captured by the on-board camera can be directly obtained as a calibration image; or, as another implementation manner, the original image collected by the multiple on-board cameras can be acquired, and the acquired multiple original images are distorted. Correction to get multiple calibration images.
- the in-vehicle camera is a fisheye camera
- the following image can be used to correct the distortion of the original image, that is, the fisheye image, to obtain a calibration image:
- ⁇ d ⁇ (1+k 1 ⁇ 2 +k 2 ⁇ 4 +k 3 ⁇ 6 +k 4 ⁇ 8 );
- x′ ( ⁇ d /r)a;
- y′ ( ⁇ d /r)b;
- X c , Y c , Z c represent the coordinates in the camera coordinate system
- X, Y, Z represent the coordinates in the world coordinate system
- k1-k4 represent the internal parameter distortion coefficient of the on-board camera
- u, v represent the image coordinate system Imaging coordinates
- a, b represent coordinates in the corrected image
- x', y' represent the coordinates after distortion
- R1 represents the amount of rotation of the coordinate in the world coordinate system to the coordinates of the camera coordinate system
- T1 represents the coordinate in the world coordinate system. Converts the amount of translation to the coordinates in the camera coordinate system.
- the on-board camera is a non-fisheye camera and conforms to the pinhole model
- the original image can be corrected for distortion using the following equation to obtain a calibration image:
- x, y, and z represent coordinates in the camera coordinate system
- X, Y, and Z represent coordinates in the world coordinate system
- k1-k6 represent internal parameter distortion coefficients of the on-vehicle camera
- p 1 and p 2 represent internal tangential distortion coefficients.
- c x and c y are the coordinates of the principal points in the camera internal reference
- the principal point is the point where the camera's main optical axis intersects the image plane
- u, v represents the imaging coordinates in the image coordinate system
- x′′, y′′ represents the distortion coordinate of.
- R1 represents the amount of rotation of coordinates in the world coordinate system to coordinates in the camera coordinate system
- T1 represents the amount of translation of coordinates in the world coordinate system to coordinate in the camera coordinate system.
- S102 For each on-vehicle camera, identify a position of the marker in the calibration image corresponding to the on-vehicle camera, and determine a positional relationship between the on-vehicle camera and the recognized marker according to the position.
- the marker can be an object such as a calibration cloth, a checkerboard, etc. If the marker is a checkerboard, the marker in the image can be identified by a black and white square, and the marker is obtained in the image coordinate system. s position.
- S102 may include: constructing a covariance matrix of the marker in a world coordinate system in which the marker is located; obtaining an initial rotation matrix R and an initial translation vector of the marker according to the covariance matrix T. Iteratively optimizing the initial rotation matrix R and the initial translation vector T by using a reprojection error between the coordinates of the marker in the calibration image and the coordinates in the world coordinate system in which the marker is located, to obtain the The positional relationship between the on-board camera and the marker.
- the feature vector corresponding to the minimum eigenvalue of the covariance matrix and the coordinate mean of the covariance matrix may be calculated;
- the feature vector and the coordinate mean value are transformed to obtain an initial rotation matrix R of the marker; and then the initial translation vector T of the marker is calculated according to the initial rotation matrix R and the coordinate mean value; and the marker is used in the calibration image.
- the re-projection error between the coordinates and the coordinates in the world coordinate system in which the marker is located, the initial rotation matrix R and the initial translation vector T are iteratively optimized to obtain the coordinates and coordinates of the marker in the calibration image.
- the Jacobian matrix can be constructed according to the partial and partial data of R and T, and the coordinates of the marker in the calibration image and the coordinates in the world coordinate system where the marker is located are used.
- the principle of minimum re-projection error is to optimize R and T multiple iterations to obtain optimized R and T.
- the specific number of iterations can be set according to the actual situation.
- the optimized R and T are obtained, and the mapping relationship between the coordinates of the marker in the calibration image and the coordinates in the world coordinate system in which the marker is located is obtained.
- the conversion relationship between the image coordinate system and the camera coordinate system can be obtained, combined with the above-mentioned "the mapping relationship between the coordinates of the marker in the calibration image and the coordinates in the world coordinate system in which the marker is located"
- the conversion relationship between the camera coordinate system and the world coordinate system in which the marker is located is obtained, that is, the positional relationship between the in-vehicle camera and the marker.
- the camera coordinate system is the XcYcZc coordinate system
- the world coordinate system of the marker is the XYZ coordinate system
- the camera is aligned with the marker for image acquisition, and the acquired image is distorted.
- a calibration image is obtained, and the image coordinate system of the calibration image is a uv coordinate system, wherein point c is the center point of the calibration image.
- the conversion relationship between the image coordinate system (uv coordinate system) and the camera coordinate system (XcYcZc coordinate system) can be obtained, and the image coordinate system (uv coordinate system) obtained above and the world coordinates of the marker object can be obtained.
- the mapping relationship between the system (XYZ coordinate system) and the conversion relationship between the camera coordinate system (XcYcZc coordinate system) and the world coordinate system (XYZ coordinate system) where the marker is located that is, between the on-board camera and the marker Positional relationship.
- S103 Convert the plurality of onboard cameras into the same coordinate system based on the determined positional relationship between each on-vehicle camera and the marker based on one marker.
- the plurality of in-vehicle cameras can be converted into the same coordinate system based on one marker.
- S101 may include: acquiring a first calibration image corresponding to the first camera, a second calibration image corresponding to the second camera, and a third calibration image corresponding to the third camera; wherein, in the first calibration image Included in the first calibration object, the second calibration image includes the first marker and the second marker, and the third calibration image includes the second marker;
- the positional relationship between the first camera and the first marker, the positional relationship between the second camera and the first marker, the positional relationship between the second camera and the second marker, and the third camera and the first are determined in S102.
- S103 includes: converting a world coordinate system in which the second marker is located according to a positional relationship between the second camera and the first marker and a positional relationship between the second camera and the second marker
- the world coordinate system in which the first marker is located obtains the coordinates of each onboard camera in the world coordinate system in which the first marker is located.
- a specific process of converting three cameras to the same coordinate system is described by taking three cameras as an example. If the number of in-vehicle cameras is greater than three, the process of converting a plurality of in-vehicle cameras into the same coordinate system is similar, and reference may be made to the present embodiment. For example, if there are four on-board cameras, the present embodiment may be used to convert three of the on-vehicle cameras into the same coordinate system, and the remaining one on-board camera is adjacent to at least one of the three on-vehicle cameras.
- the corresponding image is included in the corresponding image between the adjacent cameras, so that the remaining one in-vehicle camera can also be converted into the same coordinate system based on the positional relationship between the adjacent cameras and the same marker.
- the situation of five in-vehicle cameras or more in-vehicle cameras is similar and will not be enumerated one by one.
- the coordinates of the second camera in the world coordinate system in which the first marker is located are P1 (x1, y1, z1); according to the second camera and the second marker
- the positional relationship is obtained by obtaining the coordinates of the second camera in the world coordinate system in which the second marker is located, P2 (x2, y2, z2), and converting the world coordinate system in which the second marker is located to the first marker by using the following formula
- the world coordinate system where the object is located is P1 (x1, y1, z1)
- an in-vehicle camera is respectively disposed on the front side, the rear side, the left side, and the right side of the vehicle body.
- the camera set on the front side is referred to as a front camera
- the camera disposed on the rear side is referred to as a rear camera.
- the camera set on the left side is called the left camera
- the camera set on the right side is called the right camera
- the left front side marker is recorded as the marker 1
- the right front marker is recorded as the marker 2
- the left rear marker is recorded.
- the right rear marker is referred to as marker 4.
- the front camera can image the marker 1 and the marker 2, the camera captures the markers 3 and 4, and the left camera can mark the marker 1 and the marker.
- the object 3 performs image acquisition, and the right camera can perform image acquisition on the marker 2 and the marker 4.
- the image contains the marker 3 and the marker 4
- the left calibration image contains the marker 1 and the marker 3
- the right calibration image contains the marker 2 and the marker 4.
- the coordinates of the front camera in the world coordinate system of the marker 1 are determined, denoted as A; according to the positional relationship between the front camera and the marker 2, the front camera is determined in the world of the marker 2
- the coordinates in the coordinate system are denoted as B; according to the relationship between A and B, the world coordinate system in which the marker 2 is located is converted to the world coordinate system in which the marker 1 is located.
- the coordinates of the left camera in the world coordinate system of the marker 1 are determined, denoted as C; according to the positional relationship between the left camera and the marker 3, the left camera is determined to be at the marker 3
- the coordinates in the world coordinate system are recorded as D; according to the relationship between C and D, the world coordinate system in which the marker 3 is located is converted to the world coordinate system in which the marker 1 is located.
- the positional relationship between the right camera and the marker 2 and the conversion relationship between the world coordinate system in which the marker 1 is obtained and the world coordinate system in which the marker 2 is located, it is possible to be in the world coordinate system in which the marker 1 is located. , determine the coordinates of the right camera.
- the positional relationship between the rear camera and the marker 3 and the conversion relationship between the world coordinate system in which the marker 1 is obtained and the world coordinate system in which the marker 3 is located, it is possible to be in the world coordinate system in which the marker 1 is located. , determine the coordinates of the rear camera.
- a car camera on the front, rear, left and right sides of the car body, that is, a total of four car cameras, and for larger cars, such as buses, trucks, buses, etc. Since the vehicle body is long, a plurality of pairs of symmetrical car cameras can be disposed on the left and right sides of the vehicle body, or the vehicle cameras disposed on the left and right sides can also be asymmetrically distributed, and the distribution of the car camera is not limited.
- the world coordinate system in which the marker 4 is located needs to be converted to the world coordinate system in which the marker 1 is located, and any of the following methods may be employed:
- the coordinates of the rear camera in the world coordinate system of the marker 3 are determined, denoted as E; and the rear camera is determined according to the positional relationship between the rear camera and the marker 4.
- the coordinates in the world coordinate system in which the marker 4 is located are denoted as F; according to the relationship between E and F, and the world coordinate system in which the marker 3 obtained in the above content is located and the world coordinate system in which the marker 1 is located.
- the conversion relationship converts the world coordinate system in which the marker 4 is located to the world coordinate system in which the marker 1 is located.
- the coordinates of the right camera in the world coordinate system of the marker 2 are determined, denoted as G; according to the positional relationship between the right camera and the marker 4, the right camera is determined.
- the coordinate in the world coordinate system in which the marker 4 is located is denoted by H; according to the relationship between G and H, and the world coordinate system in which the marker 2 obtained in the above content is located and the world coordinate system in which the marker 1 is located.
- the conversion relationship converts the world coordinate system in which the marker 4 is located to the world coordinate system in which the marker 1 is located.
- S104 Determine the positional relationship between the plurality of onboard cameras and the vehicle body in the same coordinate system.
- a pair of on-board cameras can be symmetrically arranged on the left and right sides of the vehicle body, and the pair of on-board cameras are placed at half the length of the vehicle body, and on the front side of the vehicle body, and at half the width of the vehicle body.
- Set an in-vehicle camera front camera
- the coordinates of the middle point of the left and right camera coordinates are the center coordinates of the vehicle body, according to the coordinates of the center of the vehicle and the coordinates of the front camera.
- the angle between the connection and the north-south direction may be used as the rotation angle of the vehicle body, or the angle between the connection and the east-west direction may be used as the rotation angle of the vehicle body, and the like, and the like.
- an in-vehicle camera is respectively disposed on the front side, the rear side, the left side, and the right side of the vehicle body, and the coordinates of each in-vehicle camera are respectively determined in the world coordinate system where the left front side marker is located;
- S104 includes:
- a positional relationship between the right camera and the vehicle body is obtained based on the fourth coordinate and the fifth coordinate and the vehicle body rotation angle.
- the left and right cameras are symmetrically arranged, and the front and rear cameras are asymmetrically arranged.
- Make the connection between the left camera and the right camera denote L1
- make the vertical line L2 of L1 pass the camera to make the vertical line D1 of L1
- the camera makes the vertical line D2 of L1
- calculate the midpoint of D1 and D2 also It is the calculation (D1+D2)/2
- the midpoint of D1 and D2 is the vertical line L3 of L2
- the intersection of L2 and L3 is the coordinate of the center position of the vehicle body, that is, the fifth coordinate.
- the angle between L2 and the horizontal or vertical line, or the angle between L3 and the horizontal or vertical line can be used as the body rotation angle.
- the coordinate origin is the position of the marker 1, and the coordinates of the Ov point are determined in the coordinate system, that is, the coordinates of the center position of the vehicle body, that is, the fifth coordinate, in the coordinate system.
- the angle ⁇ between the vertical line L2 of the front and rear camera lines and the coordinate H w is determined, that is, the body rotation angle.
- the coordinates of the center position of the vehicle body and the rotation angle of the vehicle body are obtained, that is, the positional relationship between the vehicle body and the marker 1 is obtained, and the coordinates of each vehicle camera in the world coordinate system where the marker 1 is located have been obtained in the above content. That is, the positional relationship between each in-vehicle camera and the marker 1 is obtained, and thus, the positional relationship between each in-vehicle camera and the vehicle body is obtained.
- the calibration image corresponding to the adjacent vehicle camera includes the same marker, and for each vehicle camera, the position of the marker is identified in the calibration image corresponding to the vehicle camera, according to the location, Determining a positional relationship between the in-vehicle camera and the identified marker, the determined positional relationship includes a positional relationship between the same marker and a different in-vehicle camera, and therefore, the plurality of in-vehicles can be based on one marker
- the camera is switched to the same coordinate system, and in this same coordinate system, the positional relationship between the plurality of on-vehicle cameras and the vehicle body is determined.
- the embodiment of the present application further provides an in-vehicle camera external parameter determining device, as shown in FIG. 5, including:
- the obtaining module 501 is configured to acquire a calibration image corresponding to each vehicle camera; wherein the calibration image corresponding to each vehicle camera is obtained according to the image captured by the vehicle camera, and the calibration image corresponding to the adjacent vehicle camera includes the same marker;
- the identification module 502 is configured to identify, for each on-vehicle camera, a position of the marker in the calibration image corresponding to the on-vehicle camera;
- a first determining module 503, configured to determine, according to the location, a positional relationship between the in-vehicle camera and the identified tag;
- the conversion module 504 is configured to convert the plurality of on-vehicle cameras into the same coordinate system based on the determined positional relationship between each on-vehicle camera and the marker based on one marker;
- the second determining module 505 is configured to determine a positional relationship between the plurality of onboard cameras and the vehicle body in the same coordinate system.
- the acquiring module 501 may be specifically configured to:
- Distortion correction is performed on the acquired plurality of original images to obtain a plurality of calibration images.
- the first determining module 503 may include: a constructing submodule, an obtaining submodule, and an iterative submodule (not shown), where
- Obtaining a submodule configured to obtain an initial rotation matrix and an initial translation vector of the marker according to the covariance matrix
- the obtaining submodule may be specifically used to:
- the iterative sub-module can be specifically used to:
- the initial rotation matrix and the initial translation vector are iteratively optimized to obtain a marker in the calibration image.
- the acquiring module 501 may be specifically configured to:
- first calibration image corresponding to the first camera, a second calibration image corresponding to the second camera, and a third calibration image corresponding to the third camera; wherein the first calibration image includes a first marker, the second The first image and the second marker are included in the calibration image, and the second marker is included in the third calibration image;
- the conversion module 504 can be specifically configured to:
- the acquiring module 501 may be specifically configured to:
- the left calibration side image includes the left front side marker and the left rear side marker
- the right calibration image includes the right front side marker and the right rear marker
- the posterior calibration image includes the Describe the left posterior marker and the right posterior marker
- the conversion module 504 can be specifically configured to:
- the second determining module 505 is specifically configured to:
- the calibration image corresponding to the adjacent vehicle camera includes the same marker, and for each vehicle camera, the position of the marker is identified in the calibration image corresponding to the vehicle camera, according to the location, Determining a positional relationship between the in-vehicle camera and the identified marker, the determined positional relationship includes a positional relationship between the same marker and a different in-vehicle camera, and therefore, the plurality of in-vehicles can be based on one marker
- the camera is switched to the same coordinate system, and in this same coordinate system, the positional relationship between the plurality of on-vehicle cameras and the vehicle body is determined. It can be seen that in this solution, only the adjacent vehicle camera is required to collect the same marker, and it is not necessary to fix the parking position of the vehicle and the position of the marker, thereby improving the accuracy of obtaining the external reference.
- the embodiment of the present application further provides an electronic device, which may include: a processor and a memory; a memory for storing a computer program; and a processor for executing the program stored on the memory to implement any of the foregoing Car camera external parameter determination method.
- the electronic device can include a processor 601, a communication interface 602, a memory 603, and a communication bus 604, as shown in FIG. 6, wherein the processor 601, the communication interface 602, and the memory 603 are completed by the communication bus 604.
- Communication a processor 601, a communication interface 602, a memory 603, and a communication bus 604, as shown in FIG. 6, wherein the processor 601, the communication interface 602, and the memory 603 are completed by the communication bus 604.
- the processor 601 is configured to implement any of the above-described on-board camera external parameter determining methods when executing the program stored on the memory 603.
- the communication bus mentioned in the above electronic device may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus.
- PCI Peripheral Component Interconnect
- EISA Extended Industry Standard Architecture
- the communication bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only one thick line is shown in the figure, but it does not mean that there is only one bus or one type of bus.
- the communication interface is used for communication between the above electronic device and other devices.
- the memory may include a random access memory (RAM), and may also include a non-volatile memory (NVM), such as at least one disk storage.
- RAM random access memory
- NVM non-volatile memory
- the memory may also be at least one storage device located away from the aforementioned processor.
- the above processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; or may be a digital signal processing (DSP), dedicated integration.
- CPU central processing unit
- NP network processor
- DSP digital signal processing
- ASIC Application Specific Integrated Circuit
- FPGA Field-Programmable Gate Array
- the embodiment of the present application further provides a computer readable storage medium, where the computer readable storage medium stores a computer program, and when the computer program is executed by the processor, implements any of the above-described vehicle camera external parameter determining methods.
- the embodiment of the present application further provides an executable program code for being executed to execute any of the above-described on-board camera external parameter determining methods.
- the embodiment of the present application further provides an assist driving system, as shown in FIG. 7 , including: a processing device and a plurality of in-vehicle cameras: an in-vehicle camera 1 , an in-vehicle camera 2, an in-vehicle camera N, wherein
- Each in-vehicle camera for transmitting the acquired image to the processing device;
- the processing device is configured to obtain, according to the images collected by the plurality of on-board cameras, a calibration image corresponding to the plurality of on-vehicle cameras; wherein the calibration image corresponding to the adjacent on-vehicle camera includes the same marker;
- An in-vehicle camera that recognizes a position of a marker in a calibration image corresponding to the on-vehicle camera, and determines a positional relationship between the on-vehicle camera and the recognized marker according to the position; each of the on-vehicle camera and the marker according to the determined
- a positional relationship between the plurality of onboard cameras is converted into the same coordinate system based on one marker; and in the same coordinate system, a positional relationship between the plurality of onboard cameras and the vehicle body is determined.
- the processing device can also be used to perform any of the above-described on-board camera external parameter determination methods.
Abstract
Description
Claims (18)
- 一种车载相机外参确定方法,其特征在于,包括:获取每个车载相机对应的标定图像;其中,每个车载相机对应的标定图像根据该车载相机采集的图像得到,相邻车载相机所对应的标定图像中包含同一标记物;针对每个车载相机,在该车载相机对应的标定图像中识别标记物的位置,根据所述位置,确定该车载相机与所识别的标记物之间的位置关系;根据所确定的每个车载相机与标记物之间的位置关系,以一个标记物为基准,将所述多个车载相机转换至同一坐标系中;在所述同一坐标系中,确定所述多个车载相机与车身的位置关系。
- 根据权利要求1所述的方法,其特征在于,所述获取每个车载相机对应的标定图像,包括:获取多个车载相机采集的原始图像;对所获取的多张原始图像进行畸变校正,得到多张标定图像。
- 根据权利要求1所述的方法,其特征在于,所述根据所述位置,确定该车载相机与所识别的标记物之间的位置关系,包括:在标记物所在的世界坐标系中,构建标记物的协方差矩阵;根据所述协方差矩阵,得到标记物的初始旋转矩阵和初始平移向量;利用标记物在标定图像中的坐标与在标记物所在的世界坐标系中的坐标之间的重投影误差,将所述初始旋转矩阵及所述初始平移向量进行迭代优化,得到该车载相机与标记物之间的位置关系。
- 根据权利要求3所述的方法,其特征在于,所述根据所述协方差矩阵,得到标记物的初始旋转矩阵和初始平移向量,包括:计算所述协方差矩阵的最小特征值对应的特征向量、以及所述协方差矩阵的坐标均值;利用所述特征向量及所述坐标均值,变换得到标记物的初始旋转矩阵;根据所述初始旋转矩阵及所述坐标均值,计算标记物的初始平移向量;所述利用标记物在标定图像中的坐标与在标记物所在的世界坐标系中的坐标之间的重投影误差,将所述初始旋转矩阵及所述初始平移向量进行迭代优化,得到该车载相机与标记物之间的位置关系,包括:利用标记物在标定图像中的坐标与在标记物所在的世界坐标系中的坐标 之间的重投影误差,将所述初始旋转矩阵及所述初始平移向量进行迭代优化,得到标记物在标定图像中的坐标与在标记物所在的世界坐标系中的坐标之间的映射关系;根据所述映射关系以及该车载相机的内参,得到该车载相机与标记物之间的位置关系。
- 根据权利要求1所述的方法,其特征在于,所述获取每个车载相机对应的标定图像,包括:获取第一相机对应的第一标定图像、第二相机对应的第二标定图像以及第三相机对应的第三标定图像;其中,所述第一标定图像中包含第一标记物,所述第二标定图像中包含所述第一标记物和第二标记物,所述第三标定图像中包含所述第二标记物;所述根据所确定的每个车载相机与标记物之间的位置关系,以一个标记物为基准,将所述多个车载相机转换至同一坐标系中,包括:根据所述第二相机与所述第一标记物的位置关系以及所述第二相机与所述第二标记物的位置关系,将所述第二标记物所在的世界坐标系转换至所述第一标记物所在的世界坐标系,得到每个车载相机在所述第一标记物所在的世界坐标系中的坐标。
- 根据权利要求1所述的方法,其特征在于,所述获取每个车载相机对应的标定图像,包括:获取前相机对应的前标定图像、左相机对应的左标定图像、右相机对应的右标定图像以及后相机对应的后标定图像;其中,所述前标定图像中包含左前侧标记物和右前侧标记物,所述左标定图像中包含所述左前侧标记物和左后侧标记物,所述右标定图像中包含所述右前侧标记物和右后侧标记物,所述后标定图像中包含所述左后侧标记物和所述右后侧标记物;所述根据所确定的每个车载相机与标记物之间的位置关系,以一个标记物为基准,将所述多个车载相机转换至同一坐标系中,包括:根据所述前相机与所述左前侧标记物的位置关系,在所述左前侧标记物所在的世界坐标系中,确定所述前相机的坐标;根据所述前相机与所述左前侧标记物的位置关系以及所述前相机与所述右前侧标记物的位置关系,将所述右前侧标记物所在的世界坐标系转换至所述左前侧标记物所在的世界坐标系,再结合所述右相机与所述右前侧标记物 的位置关系,在所述左前侧标记物所在的世界坐标系中,确定所述右相机的坐标;根据所述左相机与所述左前侧标记物的位置关系,在所述左前侧标记物所在的世界坐标系中,确定所述左相机的坐标;根据所述左相机与所述左前侧标记物的位置关系以及所述左相机与所述左后侧标记物的位置关系,将所述左后侧标记物所在的世界坐标系转换至所述左前侧标记物所在的世界坐标系,再结合所述后相机与所述左后侧标记物的位置关系,在所述左前侧标记物所在的世界坐标系中,确定所述后相机的坐标。
- 根据权利要求6所述的方法,其特征在于,在所述同一坐标系中,确定所述多个车载相机与车身的位置关系,包括:在所述左前侧标记物所在的世界坐标系中,根据所述前相机的第一坐标、所述后相机的第二坐标、所述左相机的第三坐标以及所述右相机的第四坐标,计算车身中心位置的第五坐标、以及车身旋转角;根据所述第一坐标与所述第五坐标及所述车身旋转角,得到所述前相机与车身的位置关系;根据所述第二坐标与所述第五坐标及所述车身旋转角,得到所述后相机与车身的位置关系;根据所述第三坐标与所述第五坐标及所述车身旋转角,得到所述左相机与车身的位置关系;根据所述第四坐标与所述第五坐标及所述车身旋转角,得到所述右相机与车身的位置关系。
- 一种车载相机外参确定装置,其特征在于,包括:获取模块,用于获取每个车载相机对应的标定图像;其中,每个车载相机对应的标定图像根据该车载相机采集的图像得到,相邻车载相机所对应的标定图像中包含同一标记物;识别模块,用于针对每个车载相机,在该车载相机对应的标定图像中识别标记物的位置;第一确定模块,用于根据所述位置,确定该车载相机与所识别的标记物之间的位置关系;转换模块,用于根据所确定的每个车载相机与标记物之间的位置关系, 以一个标记物为基准,将所述多个车载相机转换至同一坐标系中;第二确定模块,用于在所述同一坐标系中,确定所述多个车载相机与车身的位置关系。
- 根据权利要求8所述的装置,其特征在于,所述获取模块,具体用于:获取多个车载相机采集的原始图像;对所获取的多张原始图像进行畸变校正,得到多张标定图像。
- 根据权利要求8所述的装置,其特征在于,所述第一确定模块,包括:构建子模块,用于在标记物所在的世界坐标系中,构建标记物的协方差矩阵;获得子模块,用于根据所述协方差矩阵,得到标记物的初始旋转矩阵和初始平移向量;迭代子模块,用于利用标记物在标定图像中的坐标与在标记物所在的世界坐标系中的坐标之间的重投影误差,将所述初始旋转矩阵及所述初始平移向量进行迭代优化,得到该车载相机与标记物之间的位置关系。
- 根据权利要求10所述的装置,其特征在于,所述获得子模块,具体用于:计算所述协方差矩阵的最小特征值对应的特征向量、以及所述协方差矩阵的坐标均值;利用所述特征向量及所述坐标均值,变换得到标记物的初始旋转矩阵;根据所述初始旋转矩阵及所述坐标均值,计算标记物的初始平移向量;所述迭代子模块,具体用于:利用标记物在标定图像中的坐标与在标记物所在的世界坐标系中的坐标之间的重投影误差,将所述初始旋转矩阵及所述初始平移向量进行迭代优化,得到标记物在标定图像中的坐标与在标记物所在的世界坐标系中的坐标之间的映射关系;根据所述映射关系以及该车载相机的内参,得到该车载相机与标记物之间的位置关系。
- 根据权利要求8所述的装置,其特征在于,所述获取模块,具体用于:获取第一相机对应的第一标定图像、第二相机对应的第二标定图像以及第三相机对应的第三标定图像;其中,所述第一标定图像中包含第一标记物,所述第二标定图像中包含所述第一标记物和第二标记物,所述第三标定图像中包含所述第二标记物;所述转换模块,具体用于:根据所述第二相机与所述第一标记物的位置关系以及所述第二相机与所述第二标记物的位置关系,将所述第二标记物所在的世界坐标系转换至所述第一标记物所在的世界坐标系,得到每个车载相机在所述第一标记物所在的世界坐标系中的坐标。
- 根据权利要求8所述的装置,其特征在于,所述获取模块,具体用于:获取前相机对应的前标定图像、左相机对应的左标定图像、右相机对应的右标定图像以及后相机对应的后标定图像;其中,所述前标定图像中包含左前侧标记物和右前侧标记物,所述左标定图像中包含所述左前侧标记物和左后侧标记物,所述右标定图像中包含所述右前侧标记物和右后侧标记物,所述后标定图像中包含所述左后侧标记物和所述右后侧标记物;所述转换模块,具体用于:根据所述前相机与所述左前侧标记物的位置关系,在所述左前侧标记物所在的世界坐标系中,确定所述前相机的坐标;根据所述前相机与所述左前侧标记物的位置关系以及所述前相机与所述右前侧标记物的位置关系,将所述右前侧标记物所在的世界坐标系转换至所述左前侧标记物所在的世界坐标系,再结合所述右相机与所述右前侧标记物的位置关系,在所述左前侧标记物所在的世界坐标系中,确定所述右相机的坐标;根据所述左相机与所述左前侧标记物的位置关系,在所述左前侧标记物所在的世界坐标系中,确定所述左相机的坐标;根据所述左相机与所述左前侧标记物的位置关系以及所述左相机与所述左后侧标记物的位置关系,将所述左后侧标记物所在的世界坐标系转换至所述左前侧标记物所在的世界坐标系,再结合所述后相机与所述左后侧标记物的位置关系,在所述左前侧标记物所在的世界坐标系中,确定所述后相机的坐标。
- 根据权利要求13所述的装置,其特征在于,所述第二确定模块,具体用于:在所述左前侧标记物所在的世界坐标系中,根据所述前相机的第一坐标、所述后相机的第二坐标、所述左相机的第三坐标以及所述右相机的第四坐标,计算车身中心位置的第五坐标、以及车身旋转角;根据所述第一坐标与所述第五坐标及所述车身旋转角,得到所述前相机与车身的位置关系;根据所述第二坐标与所述第五坐标及所述车身旋转角,得到所述后相机与车身的位置关系;根据所述第三坐标与所述第五坐标及所述车身旋转角,得到所述左相机与车身的位置关系;根据所述第四坐标与所述第五坐标及所述车身旋转角,得到所述右相机与车身的位置关系。
- 一种电子设备,其特征在于,包括处理器和存储器;存储器,用于存放计算机程序;处理器,用于执行存储器上所存放的程序时,实现权利要求1-7任一所述的方法步骤。
- 一种计算机可读存储介质,其特征在于,所述计算机可读存储介质内存储有计算机程序,所述计算机程序被处理器执行时实现权利要求1-7任一所述的方法步骤。
- 一种辅助驾驶系统,其特征在于,包括:处理设备及多个车载相机;每个车载相机,用于将采集的图像发送至所述处理设备;所述处理设备,用于根据所述多个车载相机采集的图像,得到所述多个车载相机对应的标定图像;其中,相邻车载相机所对应的标定图像中包含同一标记物;针对每个车载相机,在该车载相机对应的标定图像中识别标记物的位置,根据所述位置,确定该车载相机与所识别的标记物之间的位置关系;根据所确定的每个车载相机与标记物之间的位置关系,以一个标记物为基准,将所述多个车载相机转换至同一坐标系中;在所述同一坐标系中,确定所述多个车载相机与车身的位置关系。
- 一种可执行程序代码,其特征在于,所述可执行程序代码用于被运行以执行权利要求1-7任一所述的方法步骤。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810185257.7 | 2018-03-07 | ||
CN201810185257.7A CN110246184B (zh) | 2018-03-07 | 2018-03-07 | 一种车载相机外参确定方法、装置、设备及系统 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2019170066A1 true WO2019170066A1 (zh) | 2019-09-12 |
Family
ID=67846857
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2019/076915 WO2019170066A1 (zh) | 2018-03-07 | 2019-03-05 | 一种车载相机外参确定方法、装置、设备及系统 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN110246184B (zh) |
WO (1) | WO2019170066A1 (zh) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111815719A (zh) * | 2020-07-20 | 2020-10-23 | 北京百度网讯科技有限公司 | 图像采集设备的外参数标定方法、装置、设备及存储介质 |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113327292A (zh) * | 2021-06-11 | 2021-08-31 | 杭州鸿泉物联网技术股份有限公司 | 车载环视设备标定方法及装置 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8212878B2 (en) * | 2006-06-29 | 2012-07-03 | Hitachi, Ltd. | Calibration apparatus of on-vehicle camera, program, and car navigation system |
CN103593836A (zh) * | 2012-08-14 | 2014-02-19 | 无锡维森智能传感技术有限公司 | 一种摄像头参数计算方法及相机确定车体姿态的方法 |
CN103985118A (zh) * | 2014-04-28 | 2014-08-13 | 无锡观智视觉科技有限公司 | 一种车载环视系统摄像头参数标定方法 |
CN104833372A (zh) * | 2015-04-13 | 2015-08-12 | 武汉海达数云技术有限公司 | 一种车载移动测量系统高清全景相机外参数标定方法 |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8860818B1 (en) * | 2013-07-31 | 2014-10-14 | Apple Inc. | Method for dynamically calibrating rotation offset in a camera system |
EP2858035B1 (en) * | 2013-10-01 | 2020-04-22 | Application Solutions (Electronics and Vision) Limited | System, vehicle and method for online calibration of a camera on a vehicle |
CN104050650B (zh) * | 2014-06-19 | 2017-02-15 | 湖北汽车工业学院 | 基于坐标变换的完整成像的图像拼接方法 |
CN105631853B (zh) * | 2015-11-06 | 2018-01-30 | 湖北工业大学 | 车载双目相机标定及参数验证方法 |
CN105654422A (zh) * | 2015-12-23 | 2016-06-08 | 北京观著信息技术有限公司 | 点云配准方法和系统 |
CN106570907B (zh) * | 2016-11-22 | 2020-07-10 | 海信集团有限公司 | 一种相机标定方法及装置 |
CN106846410B (zh) * | 2016-12-20 | 2020-06-19 | 北京鑫洋泉电子科技有限公司 | 基于三维的行车环境成像方法及装置 |
CN107133988B (zh) * | 2017-06-06 | 2020-06-02 | 科大讯飞股份有限公司 | 车载全景环视系统中摄像头的标定方法及标定系统 |
-
2018
- 2018-03-07 CN CN201810185257.7A patent/CN110246184B/zh active Active
-
2019
- 2019-03-05 WO PCT/CN2019/076915 patent/WO2019170066A1/zh active Application Filing
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8212878B2 (en) * | 2006-06-29 | 2012-07-03 | Hitachi, Ltd. | Calibration apparatus of on-vehicle camera, program, and car navigation system |
CN103593836A (zh) * | 2012-08-14 | 2014-02-19 | 无锡维森智能传感技术有限公司 | 一种摄像头参数计算方法及相机确定车体姿态的方法 |
CN103985118A (zh) * | 2014-04-28 | 2014-08-13 | 无锡观智视觉科技有限公司 | 一种车载环视系统摄像头参数标定方法 |
CN104833372A (zh) * | 2015-04-13 | 2015-08-12 | 武汉海达数云技术有限公司 | 一种车载移动测量系统高清全景相机外参数标定方法 |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111815719A (zh) * | 2020-07-20 | 2020-10-23 | 北京百度网讯科技有限公司 | 图像采集设备的外参数标定方法、装置、设备及存储介质 |
CN111815719B (zh) * | 2020-07-20 | 2023-12-22 | 阿波罗智能技术(北京)有限公司 | 图像采集设备的外参数标定方法、装置、设备及存储介质 |
Also Published As
Publication number | Publication date |
---|---|
CN110246184A (zh) | 2019-09-17 |
CN110246184B (zh) | 2021-06-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2019184885A1 (zh) | 一种相机外参标定方法、装置及电子设备 | |
EP3633539A2 (en) | Method for position detection, device, and storage medium | |
US9536306B2 (en) | Vehicle vision system | |
JP4803449B2 (ja) | 車載カメラの校正装置、校正方法、並びにこの校正方法を用いた車両の生産方法 | |
US20140085409A1 (en) | Wide fov camera image calibration and de-warping | |
CN109741241B (zh) | 鱼眼图像的处理方法、装置、设备和存储介质 | |
US20230215187A1 (en) | Target detection method based on monocular image | |
CN109544629A (zh) | 摄像头位姿确定方法和装置以及电子设备 | |
US10803621B2 (en) | Method and device for building camera imaging model, and automated driving system for vehicle | |
JP2020113268A (ja) | 牽引ヒッチの位置を計算する方法 | |
WO2019170066A1 (zh) | 一种车载相机外参确定方法、装置、设备及系统 | |
CN111489288B (zh) | 一种图像的拼接方法和装置 | |
CN113988112B (zh) | 车道线的检测方法、装置、设备及存储介质 | |
CN112348902A (zh) | 路端相机的安装偏差角标定方法、装置及系统 | |
CN113409396B (zh) | 一种adas单目相机的标定方法 | |
TWI778368B (zh) | 車載鏡頭的自動校正方法以及車載鏡頭裝置 | |
CN110956585B (zh) | 全景图像拼接方法、装置以及计算机可读存储介质 | |
CN112330576A (zh) | 车载鱼眼摄像头畸变矫正方法、装置、设备及存储介质 | |
CN111368927A (zh) | 一种标注结果处理方法、装置、设备及存储介质 | |
CN116630401A (zh) | 鱼眼相机测距方法及终端 | |
KR101431378B1 (ko) | 전방위 영상의 생성 방법 및 장치 | |
CN116152347A (zh) | 一种车载摄像头安装姿态角标定方法及系统 | |
WO2023168747A1 (zh) | 基于域控制器平台的自动泊车的停车位标注方法及装置 | |
CN113610927B (zh) | 一种avm摄像头参数标定方法、装置及电子设备 | |
CN115601449A (zh) | 标定方法、环视图像生成方法、装置、设备及存储介质 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 19764296 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 19764296 Country of ref document: EP Kind code of ref document: A1 |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 19764296 Country of ref document: EP Kind code of ref document: A1 |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 19.03.2021) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 19764296 Country of ref document: EP Kind code of ref document: A1 |