WO2023284358A1

WO2023284358A1 - Camera calibration method and apparatus, electronic device, and storage medium

Info

Publication number: WO2023284358A1
Application number: PCT/CN2022/088601
Authority: WO
Inventors: 张保成
Original assignee: 北京迈格威科技有限公司
Priority date: 2021-07-16
Filing date: 2022-04-22
Publication date: 2023-01-19
Also published as: CN113658265A

Abstract

The present application provides a camera calibration method and apparatus, an electronic device, and a storage medium. The method comprises: obtaining information of bounding boxes of areas where at least two targets in a first image are located, wherein the first image is an image captured by a camera at a first moment, and the information of each bounding box comprises the position and the size of the corresponding bounding box in the first image; determining a first linear parameter group according to the information of the bounding boxes of the at least two targets, the first linear parameter group being used for representing a linear relationship between the position and the size of the bounding box in the first image; when the first linear parameter group is present in a predetermined correspondence between a linear parameter group and a camera parameter group, finding a first camera parameter group corresponding to the first linear parameter group from the correspondence, the first camera parameter group comprising the mounting height of the camera, the pitch angle of the camera, and the focal length of the camera; and determining a target camera parameter group according to the first camera parameter group. Camera calibration can be completed without placing a calibration object or controlling the camera to move.

Description

Camera calibration method, device, electronic equipment and storage medium

Cross References to Related Applications

This application claims priority to a Chinese patent application with application number 2021108077251 entitled "Camera Calibration Method, Device, Electronic Equipment, and Storage Medium" filed with the State Intellectual Property Office of China on July 16, 2021, the entire contents of which are incorporated by reference incorporated in this application.

technical field

The present application relates to the technical field of image processing, in particular, to a camera calibration method, device, electronic equipment and storage medium.

Background technique

In the image measurement process and machine vision applications, in order to determine the corresponding relationship between the three-dimensional geometric position of the point on the surface of the space object and the pixel point in the image, it is usually necessary to establish a geometric model of camera imaging, and the parameters of the geometric model are the camera parameters. The process of solving the camera parameters is called camera calibration.

There are two main types of related camera calibration methods. The first type of camera calibration method needs to rely on the placed calibration objects, and the manufacturing accuracy of the calibration objects will affect the calibration results. At the same time, some application scenarios (for example, security prevention and control scenarios) are not suitable for Placing a calibration object limits the application of this method; the second type of camera calibration method mainly uses the motion information of the camera to calibrate the camera. This method does not need to rely on the calibration object, but it needs to control the camera to do some special motion, and The method is not suitable for scenes where the motion information is unknown or the camera movement cannot be controlled (for example, a security prevention and control scene). From the above analysis, it can be seen that the application scene of the relevant camera calibration method is limited.

Contents of the invention

In view of this, embodiments of the present application provide a camera calibration method, device, electronic equipment, and storage medium, so as to at least solve the above problems.

Some embodiments of the present application provide a camera calibration method, the method may include: acquiring information of detection frames of at least two target areas in the first image; wherein, the first image is taken by the camera at the first moment The image received; the information of each detection frame includes: the position and size of the corresponding detection frame in the first image; according to the information of the detection frames of the at least two targets, the first linear parameter group is determined; wherein , the first linear parameter group is used to characterize the linear relationship between the position and size of the detection frame in the first image; the first linear parameter exists in the predetermined correspondence between the linear parameter group and the camera parameter group When grouping, find the first camera parameter group corresponding to the first linear parameter group from the corresponding relationship; wherein, the first camera parameter group includes: the erection height of the camera, the pitch of the camera angle and the focal length of the camera; determining a target camera parameter set according to the first camera parameter set.

In the above implementation process, the information of the detection frames in the area where at least two targets are located in the first image captured by the camera is obtained, and the information of each detection frame includes: the position and size of the corresponding detection frame in the first image ; According to the information of the detection frame of at least two targets, determine the first linear parameter group that characterizes the linear relationship between the position and size of the detection frame in the first image, and then from the predetermined linear parameter group and camera parameter group In the corresponding relationship, quickly find out the first camera parameter group corresponding to the first linear parameter group, and determine the target camera parameter group according to the first camera parameter group. The whole process does not need to place a calibration object, and does not need to control the camera movement. Can complete camera calibration.

Based on these embodiments, in a possible design, the method may further include: acquiring a plurality of different camera parameter groups; parameters corresponding to any two camera parameter groups in the plurality of different camera parameter groups The types are the same; for each camera parameter group, use the camera parameter group, the size and position information of at least two virtual targets to generate a detection frame simulation diagram; wherein, the detection frame simulation diagram includes: the at least two The detection frame of the virtual target; according to the position and size of at least two detection frames in the detection frame simulation diagram in the detection frame simulation diagram, a linear parameter group is determined; the relationship between the linear parameter group and the camera parameter group is established Correspondence.

In the above implementation process, after obtaining multiple different camera parameter sets, for each camera parameter set, use the camera parameter set, the size and position information of at least two virtual targets in the world coordinate system to generate a detection frame simulation Figure; then according to the position and size of at least two detection frames in the detection frame simulation diagram, determine the linear parameter group representing the linear relationship between the position and size of the detection frame in the detection frame simulation diagram, and set up the linear parameter group The corresponding relationship with the camera parameter set ensures that the camera parameter set of the camera can be quickly searched out subsequently according to the corresponding relationship.

Based on these embodiments, in a possible design, for each camera parameter group, using the camera parameter group, the size and position information of at least two virtual targets to generate a detection frame simulation diagram may include: The size distribution of the real target of the same type as the virtual target generates the size of the at least two virtual targets; generates the position information of the at least two virtual targets; for each camera parameter group, according to the camera parameter group, the at least The size and position information of the two virtual objects is used to generate the simulation diagram of the detection frame.

In the above implementation process, the sizes of at least two virtual targets are generated according to the size distribution of real targets of the same type as the virtual targets; for each camera parameter set, the positions of at least two virtual targets generated according to the camera parameter set Size and position information to generate a simulation map of the detection frame. Since the size of the virtual target is generated according to the size distribution of the same type of real target, it is ensured that the generated detection frame distribution map is more meaningful.

Based on these embodiments, in a possible design, obtaining a plurality of different camera parameter groups may include: based on the target sampling interval, performing discrete values for each camera parameter within a value range to obtain the plurality of camera parameters parameter group.

In the above implementation process, based on the target sampling interval, each camera parameter is discretely valued within a range of values, thereby ensuring that multiple different camera parameter groups can be obtained.

Based on these embodiments, in a possible design, when the first linear parameter group does not exist in the corresponding relationship, the method may further include: for each linear parameter group in the corresponding relationship, in When the difference between the values of the corresponding parameters in the linear parameter group and the first linear parameter group is less than the target threshold, determine that the linear parameter group is the second linear parameter group; from the correspondence, find out the relationship with A second camera parameter set corresponding to the second linear parameter set; determining the target camera parameter set according to the second camera parameter set.

For any two linear parameter groups, if the difference between the values of the corresponding parameters (parameters of the same category in the two linear parameter groups) in the two linear parameter groups is relatively small, then the two linear parameter groups are The difference between the values of the corresponding parameters in the corresponding camera parameter group will also be relatively small. Therefore, in the above implementation process, when the first linear parameter group does not exist in the corresponding relationship, it is determined from the corresponding relationship The difference between the values of the corresponding parameters in the first linear parameter group and the second linear parameter group whose values are smaller than the target threshold, and then find out the second camera parameter group corresponding to the second linear parameter group from the corresponding relationship , and then determine the target camera parameter set according to the second camera parameter set, so that when the first linear parameter set does not exist in the corresponding relationship, camera calibration can be completed relatively accurately.

Based on these embodiments, in a possible design, when the number of the second linear parameter groups is at least two, the second linear parameter group corresponding to the second linear parameter group is found from the corresponding relationship. The camera parameter group may include: for each second linear parameter group, finding the second camera parameter group corresponding to the second linear parameter group from the corresponding relationship; correspondingly, according to the second The camera parameter group, determining the target camera parameter group includes: obtaining the detection frame information of the target area in the second image; the second image is the image captured by the camera at the second moment; the first moment and the The difference between the second moments is smaller than the target time difference; for each second camera parameter group, based on the second camera parameter group, the information of the detection frame of the target in the first image, and the information in the second image The information of the detection frame of the target determines the final moving speed of the target captured by the camera from the first moment to the second moment; when it is determined that the final moving speed is within the normal moving speed range, determine the The second camera parameter set is qualified; according to the qualified second camera parameter set, the target camera parameter set is determined.

In the above implementation process, when the number of the second linear parameter groups is at least two, for each second linear parameter group, find out the second camera corresponding to the second linear parameter group from the corresponding relationship Parameter group, in order to eliminate unqualified camera parameter groups in all the second camera parameter groups found, so as to improve the accuracy of camera calibration results, therefore, this application considers that in most cases, the normal movement of the target The speed is within the normal moving speed range, and then obtain the detection frame information of the area where the target is located in the second image captured by the camera at the second moment, wherein the difference between the first moment and the second moment is smaller than the target time difference; then For each second camera parameter group, based on the second camera parameter group and the information of the detection frame of the target in the first image and the second image, determine the final moving speed, and when the final moving speed is determined to be within the normal moving speed range, determine that the second camera parameter set is qualified, and then determine the target camera parameter set according to the qualified second camera parameter set to ensure the accuracy of camera calibration.

Based on these embodiments, in a possible design, based on the second camera parameter set, the information of the detection frame of the target in the first image, and the information of the detection frame of the target in the second image, determine The final moving speed of the target captured by the camera from the first moment to the second moment may include: the information of the detection frame of the target in the first image, and the target's detection frame information in the second image. In the information of the detection frame, at least one detection frame information group of the target is determined; each detection frame information group includes: information of two detection frames corresponding to the target; information for each detection frame in each detection frame information group , based on the second camera parameter group and the information of the detection frame, determine the position information of the target corresponding to the detection frame in the world coordinate system; according to the information corresponding to each detection frame corresponding to the detection frame information group According to the position information of the target in the world coordinate system, the moving speed of the corresponding target between the first moment and the second moment is determined; according to the moving speed of the at least one target, the obtained Describe the final movement speed.

In the above implementation process, at least one target detection frame information group is determined from the target detection frame information in the first image and the second image, and for each detection frame information in each detection frame information group, Based on the second camera parameter group and the information of the detection frame, determine the position information of the target corresponding to the detection frame in the world coordinate system, and then according to the position information of each detection frame corresponding to the detection frame information group, Determine the moving speed of the corresponding target between the first moment and the second moment; determine the final moving speed according to the moving speed of at least one target to ensure the accuracy of the final moving speed.

Based on these embodiments, in a possible design, the determining the target camera parameter set according to the qualified second camera parameter set may include: when the number of qualified second camera parameter sets is at least two, Values of corresponding parameters in at least two qualified second camera parameter sets are respectively averaged to obtain a target camera parameter set.

In the above implementation process, the target camera parameter set is obtained by averaging the values of corresponding parameters in at least two qualified second camera parameter sets, so as to improve the accuracy of camera calibration results.

Some other embodiments of the present application provide a camera calibration device, the device may include: a first acquisition unit configured to acquire information of detection frames of areas where at least two targets are located in the first image; wherein, the The first image is the image captured by the camera at the first moment; the information of each detection frame includes: the position and size of the corresponding detection frame in the first image; the first determining unit is configured to be used according to the The information of the detection frames of at least two targets determines a first linear parameter group; wherein, the first linear parameter group is used to characterize the linear relationship between the position and size of the detection frames in the first image; the first The searching unit is configured to find out the first linear parameter group corresponding to the first linear parameter group from the corresponding relationship when the first linear parameter group exists in the predetermined correspondence between the linear parameter group and the camera parameter group. A camera parameter group; wherein, the first camera parameter group includes: the mounting size of the camera, the pitch angle of the camera, and the focal length of the camera; a first target determination unit, configured to parameter set determines the target camera parameter set.

Based on other embodiments above, in a possible design, the device may further include: a camera parameter acquisition unit configured to acquire multiple different camera parameter sets; among the multiple different camera parameter sets The types of parameters corresponding to any two camera parameter groups are the same; the simulation map generation unit is configured to generate a detection frame for each camera parameter group by using the camera parameter group, at least two virtual target size and position information A simulation diagram; wherein, the detection frame simulation diagram includes: the detection frames of the at least two virtual targets; the linear parameter group determination unit is configured to be used for at least two detection frames in the detection frame simulation diagram according to the The detection frame simulates the position and size in the diagram to determine a linear parameter set; the corresponding relationship establishing unit is configured to establish a corresponding relationship between the linear parameter set and the camera parameter set.

Based on other embodiments above, in a possible design, the simulation diagram generation unit may include: a size generation unit configured to generate the The size of at least two virtual targets; the position information generation unit configured to generate the position information of the at least two virtual targets; the simulation map generation subunit configured to, for each camera parameter group, according to the camera parameters group, the size and position information of the at least two virtual objects, and generate the simulation diagram of the detection frame.

Based on the other embodiments above, in a possible design, the camera parameter acquisition unit may be configured to perform discrete values for each camera parameter within a value range based on the target sampling interval to obtain the multiple different camera parameter sets.

Based on some other embodiments above, in a possible design, the device may further include: a second determining unit configured to, when the first linear parameter group does not exist in the corresponding relationship, for the For each linear parameter group in the corresponding relationship, when the difference between the value of the linear parameter group and the value of the corresponding parameter in the first linear parameter group is smaller than the target threshold, the linear parameter group is determined to be the second linear parameter group The second search unit is configured to find out the second camera parameter set corresponding to the second linear parameter set from the corresponding relationship; the second target determination unit is configured to use the second target determination unit according to the second A camera parameter set, for determining the target camera parameter set.

Based on the other embodiments above, in a possible design, when the number of the second linear parameter groups is at least two, the second search unit may be configured to specifically use for each second linear parameter group, from the corresponding relationship, find out the second camera parameter group corresponding to the second linear parameter group;

Correspondingly, the second target determining unit may include:

The second acquiring unit is configured to acquire the information of the detection frame of the area where the target is located in the second image; the second image is an image captured by the camera at a second moment; the first moment and the second The difference between the two moments is less than the target time difference; the speed determination unit is configured to, for each second camera parameter set, based on the second camera parameter set, the information of the detection frame of the target in the first image, and The information of the detection frame of the target in the second image determines the final moving speed of the target captured by the camera from the first moment to the second moment; the screening unit is configured to determine the When the final moving speed is within the normal moving speed range, it is determined that the second camera parameter set is qualified; the second target determination subunit is configured to determine the target camera parameter set according to the qualified second camera parameter set.

Based on other embodiments above, in a possible design, the speed determination unit may include: an information group determination unit configured to obtain the information of the detection frame of the target from the first image, and the second In the information of the detection frame of the target in the two images, at least one detection frame information group of the target is determined; each detection frame information group includes: information of two detection frames corresponding to the target; the position determination unit is configured to be used for each The information of each detection frame in the first detection frame information group, based on the second camera parameter group and the information of the detection frame, determine the position information of the target corresponding to the detection frame in the world coordinate system; the moving speed The determination unit is configured to determine, according to the position information of the target corresponding to each detection frame corresponding to the detection frame information group in the world coordinate system, that the corresponding target The moving speed between the second moments; the speed subunit is configured to obtain the final moving speed according to the moving speed of the at least one target.

Based on other embodiments above, in a possible design, the second target determination unit may be further configured to, when the number of qualified second camera parameter sets is at least two, respectively select at least two qualified The values of the corresponding parameters in the second camera parameter set are averaged to obtain the target camera parameter set.

Still other embodiments of the present application provide an electronic device, including a processor and a memory connected to the processor, where a computer program is stored in the memory, and when the computer program is executed by the processor, the The electronic device executes the method described in the first aspect.

Still other embodiments of the present application provide a computer-readable storage medium, where a computer program is stored in the storage medium, and when the computer program is run on a computer, the computer is made to execute the method described in the first aspect .

Still other embodiments of the present application provide a computer program, where the computer program includes computer readable code, and when the computer readable code is executed by a processor, implements the method provided by some implementations of the present application.

Still other embodiments of the present application provide a computer program product, where the computer program product includes a computer program, and when the computer program is executed by a processor, the method provided by some implementations of the present application is implemented.

Other features and advantages of the present application will be set forth in the ensuing description and, in part, will be apparent from the description, or can be learned by practicing the embodiments of the present application. The objectives and other advantages of the application may be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

Description of drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the accompanying drawings used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, so It should be regarded as a limitation on the scope, and those skilled in the art can also obtain other related drawings based on these drawings without creative work.

FIG. 1 is a schematic flowchart of a camera calibration method provided in an embodiment of the present application.

FIG. 2 is a schematic diagram of a simulation diagram of a detection frame provided by an embodiment of the present application.

FIG. 3 is a schematic structural diagram of a camera calibration device provided by an embodiment of the present application.

FIG. 4 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.

Icons: 300-camera calibration device; 310-first acquisition unit; 320-first determination unit; 330-first search unit; 340-first target determination unit; 400-electronic equipment; 401-processor; 402-memory ; 403-communication interface.

detailed description

The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application.

It should be noted that like numerals and letters denote similar items in the following figures, therefore, once an item is defined in one figure, it does not require further definition and explanation in subsequent figures. Meanwhile, in the description of the present application, the terms "first", "second" and the like are only used to distinguish descriptions, and cannot be understood as indicating or implying relative importance.

As an important branch of artificial intelligence, computer vision is specifically to allow machines to recognize the world. Computer vision technology usually includes face recognition, liveness detection, fingerprint recognition and anti-counterfeiting verification, biometric recognition, face detection, pedestrian detection, target detection, pedestrian detection, etc. Recognition, image processing, image recognition, image semantic understanding, image retrieval, video processing, video content recognition, 3D reconstruction, virtual reality, augmented reality, simultaneous localization and mapping (SLAM), computational photography, robot navigation and positioning and other technologies. With the research and progress of artificial intelligence technology, this technology has been applied in many fields, such as security protection, urban management, traffic management, building management, park management, face access, face attendance, logistics management, warehouse management, Robots, smart marketing, computational photography, mobile imaging, cloud services, smart home, wearable devices, unmanned driving, autonomous driving, smart medical care, face payment, face unlocking, witness verification, cameras, mobile Internet, webcasting, Intelligent temperature measurement and other fields.

Please refer to FIG. 1 . FIG. 1 is a flow chart of a camera calibration method provided by an embodiment of the present application. The process shown in FIG. 1 will be described in detail below, and the method may include steps: S11-S14.

S11: Obtain the information of the detection frames in the area where at least two targets are located in the first image; wherein, the first image is an image captured by the camera at the first moment; the information of each detection frame includes: the corresponding detection frame is in The position and size in the first image.

S12: Determine the first linear parameter group according to the information of the detection frames of the at least two targets; wherein, the first linear parameter group is used to characterize the relationship between the position and size of the detection frames in the first image linear relationship.

S13: When the first linear parameter set exists in the predetermined correspondence between the linear parameter set and the camera parameter set, find the first camera parameter set corresponding to the first linear parameter set from the correspondence; Wherein, the first camera parameter group includes: the mounting height of the camera, the pitch angle of the camera, and the focal length of the camera.

S14: Determine a target camera parameter set according to the first camera parameter set.

The above method is described in detail below.

For the convenience of subsequent understanding, here, the image coordinate system, camera coordinate system and world coordinate system involved in the camera calibration method are first defined.

Specifically, the origin of the image coordinate system can be the center of the image captured by the camera or the vertex of the captured image, one of the x-axis and the y-axis of the image coordinate system is parallel to the upper edge of the image, and the x axis of the image coordinate system is The other of the axis and the y axis is parallel to the lower edge;

The origin of the camera coordinate system is the center of the camera, the z-axis of the camera coordinate system is consistent with the shooting direction of the camera, and the x-axis and y-axis of the camera coordinate system are determined according to the right-hand spiral rule;

The origin of the world coordinate system is the projection point of the center of the camera on the road where the target is located, the z axis of the world coordinate system is perpendicular to the road where the target is located, and the xy axis of the world coordinate system overlaps with the road where the target is located.

Wherein, in this embodiment, the camera can be a camera arranged at any position, and the target can be a pedestrian, an animal, a vehicle, a building, etc.; in this embodiment, the target is a pedestrian, and the detection frame can be a rectangle or a square; The shape of the detection frame can be determined according to the shape of the target, usually a quadrilateral.

The position and size of the detection frame in the first image may be determined according to an image processing algorithm or a target detection model. It can be understood that the position and size of the detection frame in the first image may represent the position and size of its corresponding target in the image coordinate system.

In one example, the position of the detection frame can be determined according to the lower edge of the detection frame. For a detection frame of a specific type of target, the aspect ratio of the detection frame is determined, and the long side, short side, diagonal line, area, etc. of the detection frame can be used as the size of the detection frame.

As an implementation, S11 can be implemented in the following manner: acquire the first image, use an image processing algorithm or a pre-trained target detection model to detect the target in the first image, and determine the area where at least two targets are located Information about the location and size of the detection box in the first image.

Wherein, for the training of the target detection model, reference may be made to related technologies, so details will not be repeated here.

As another implementation manner, the information of the detection frames of the regions where at least two targets are located in the first image may be directly acquired from a third party.

In the actual implementation process, S12 can be implemented in the following manner, for the information of each detection frame in the information of the detection frames of at least two targets, the position and size of the detection frame are input into the linear relationship expression b=k×u In +d, the linear relationship equations are obtained, and the linear relationship equations are solved simultaneously to obtain the first linear parameter group.

S12 can also be implemented in the following manner, according to the information of each detection frame in the information of the detection frames of at least two targets, the linear relationship expression b=k×u+d is linearly fitted to obtain the first linear parameter group .

Wherein, b is the size of the detection frame in the image taken by the camera (in S12, the image is the first image), and u is the image of the detection frame in the camera (in S12, the image is the first image) The position in , k is the slope of the line, and d is the intercept of the line. The first linear parameter group may include a value of parameter k and a value of parameter d; k and d may be collectively referred to as linear parameters.

In one example, u is determined with the center of the image as the coordinate origin of the image coordinate system.

Wherein, the size of the detection frame in the first image conforms to a linear relationship with the position in the first image and is determined as follows:

Obtain the initial relationship expression, where the initial relationship expression represents the position, size and camera parameters of the detection frame corresponding to the target in the image captured by the camera, the position of the target in the world coordinate system, and the relationship between the actual size of the target . When the xy axis of the world coordinate system overlaps with the plane where the target is located (for example, when the target is a pedestrian, the world coordinate system overlaps with the road surface where the pedestrian is located), the following initial expression can be obtained:

c is the actual size of the target, f is the focal length of the camera, α is the pitch angle of the camera in the world coordinate system defined above, h is the height of the camera, and b is the size of the detection frame corresponding to the target in the image captured by the camera , u is the position of the detection frame corresponding to the target in the image captured by the camera.

It is worth mentioning that the camera calibration is to determine the value of the camera parameters, the camera parameters include: the focal length f of the camera, the pitch angle α of the camera, and the erection height h of the camera.

Taking the image size of 1080P as an example, the value range of u is [-540,540]. Within this range, the relationship between b and u can be approximated as a linear relationship from the above initial relationship expression, and the linear parameter used to characterize the linear parameter Groups k and d are associated with camera parameters.

After the first linear parameter set is acquired, S13 is executed.

Wherein, each linear parameter group in the corresponding relationship includes: the value of the parameter k and the value of the parameter d, and each linear parameter group is different; each camera parameter group in the corresponding relationship includes: the value of the parameter f, the value of the parameter α and The value of parameter h is different for each camera parameter group.

In the actual implementation process, S13 can be implemented as follows, the value of the parameter k in the first linear parameter group is compared with each value of the parameter k in the corresponding relationship, and the value of the parameter d in the first linear parameter group Values are compared with each value of the parameter d in the corresponding relationship to determine whether the value of the parameter k is the same as the value of k in the first linear parameter group in the linear parameter group of the corresponding relationship, and the value of the parameter d is the same as that of the first linear parameter group A first linear parameter group with the same value of d in a linear parameter group, when it is determined that there is a first linear parameter group in the corresponding relationship, find the first camera parameter group corresponding to the first linear parameter group from the corresponding relationship .

In an actual implementation process, S14 may be implemented in the following manner, directly determining the first camera parameter set as the target camera parameter set.

The target camera parameter set is the camera parameter set corresponding to the camera that captured the first image.

According to the embodiment of the present application, since it is found that the size and position of the detection frame in the image captured by the camera conforms to a linear relationship, the representative detection frame can be determined according to the size and position of the detection frames of at least two targets in the image captured by the camera The first linear parameter set of the linear relationship between the position and size in the image captured by the camera, and then quickly find out the first linear parameter set from the corresponding relationship between the predetermined linear parameter set and the camera parameter set The corresponding camera parameter group, and determine the target camera parameter group representing the camera parameters according to the searched camera parameter group. The whole process does not need to place a calibration object, nor does it need to control the camera movement. The camera calibration can be completed only by the image taken by the camera itself. The efficiency and accuracy of camera calibration are greatly improved.

As an implementation manner, in order to obtain a predetermined correspondence between a linear parameter set and a camera parameter set, the method may further include steps A1-A4.

A1: multiple different camera parameter sets are acquired; the types of parameters corresponding to any two camera parameter sets in the multiple different camera parameter sets are the same.

Wherein, each camera parameter group in a plurality of different camera parameter groups includes: the value of the focal length f of the camera, the value of the pitch angle α of the camera and the value of the erection height h of the camera, and in each camera parameter group At least one of the focal length f of the camera, the pitch angle α of the camera, and the erection height h of the camera is different from the value of the corresponding parameter in other camera parameter groups.

As an implementation manner, A1 may include: based on the target sampling interval, performing discrete values for each camera parameter within a value range to obtain the plurality of different camera parameter groups.

Among them, the target sampling interval and the value range corresponding to each camera parameter are set according to actual needs, and there is no limitation here. When the value range of each camera parameter is constant, the smaller the sampling interval is, the corresponding After the value is discretely selected within the value range, the number of camera parameter groups obtained is larger.

As another implementation manner, A1 may be implemented in the following manner, directly acquiring multiple different camera parameter sets from a third party, or acquiring multiple different camera parameter sets stored in advance.

After obtaining multiple different camera parameter sets, step A2 is performed.

A2: For each camera parameter group, use the camera parameter group, the size and position information of at least two virtual targets to generate a detection frame simulation diagram; wherein, the detection frame simulation diagram includes: the at least two virtual targets detection frame.

As an implementation manner, A2 may include steps A21-A23.

A21: Generate the sizes of the at least two virtual objects according to the size distribution of real objects of the same type as the virtual objects.

The size of the virtual target may be the size of the virtual target in the world coordinate system.

In the actual implementation process, A21 can be implemented in the following manner, obtain the size distribution of multiple real targets of the same type as the virtual target, and take the values of the minimum size and maximum size according to the minimum size and maximum size in the size distribution The range is randomly sampled to obtain the dimensions of at least two virtual objects.

As an implementation, A21 can be implemented in the following manner: obtain the sizes of multiple real targets of the same type as the virtual target, randomly select the sizes of at least two real targets from the multiple real targets, and select The sizes of the at least two real targets are taken as the sizes of the at least two virtual targets.

As another implementation, A21 can be implemented in the following manner: obtain the sizes of multiple real targets of the same type as the virtual target, determine the average size of multiple real targets according to the sizes of the multiple real targets, and generate at least two The size of the virtual target, wherein the size of each virtual target obeys a normal distribution whose mean is an average size.

Wherein, the number of multiple real targets may be 100, 1000, 2000, etc., which is not limited.

A22: Generate location information of the at least two virtual objects.

During actual implementation, A22 may randomly generate position information of at least two virtual targets in the following manner.

As an implementation manner, A22 may be implemented in the following manner, generating position information of at least two virtual objects according to at least two pieces of predetermined position information.

The position information may be the position of the virtual target in the world coordinate system, or the position of the virtual target in the image coordinate system.

Wherein, the execution order of A21 and A22 is not limited.

A23: For each camera parameter set, generate the detection frame simulation diagram according to the camera parameter set, the size and position information of the at least two virtual objects.

In the actual implementation process, when the position information of the virtual target is the position of the virtual target in the world coordinate system, A23 can be implemented in the following manner, for the size and position information of each virtual target, the virtual target is generated at the corresponding position 3D model, and then according to the 3D model of the virtual target and the camera parameter set, determine the projection of the 3D model of the virtual target in the image coordinate system of the camera corresponding to the camera parameter set (if the projection is determined, it is determined The size and position of the virtual target in the image coordinate system), and then according to the projection of the three-dimensional models of at least two virtual targets, a detection frame simulation diagram is generated; wherein, the detection frame simulation diagram includes at least two detection frames corresponding to the virtual targets box, as shown in Figure 2.

As an implementation, when the position information of the virtual target is the position of the virtual target in the image coordinate system of the camera corresponding to the camera parameter group, A23 can be implemented in the following manner. For the size and position information of each virtual target, use The size and position information of the camera parameter group and the virtual target, determine the size of the detection frame corresponding to the virtual target in the image coordinate system of the camera corresponding to the camera parameter group according to the aforementioned initial relational expression, and then according to at least two The position in the image coordinate system and the size in the image coordinate system of the detection frame corresponding to each virtual target are generated to generate a simulation diagram of the detection frame.

It can be understood that generating a detection frame simulation diagram does not mean that an image must be generated, as long as the size and position information of the detection frame corresponding to the area where the virtual target is located in the image coordinate system is determined, it is considered that the detection frame simulation has been generated picture. Of course, it is also possible to display the pixel values representing the detection frame different from the pixel values of the background to generate an image, so that the user can check whether the size and position of the virtual target are reasonable.

After the detection frame simulation diagram is generated, step A3 is executed.

A3: Determine a linear parameter set according to the positions and sizes of at least two detection frames in the detection frame simulation graph in the detection frame simulation graph.

It can be understood that the coordinate system of the detection frame simulation graph is the image coordinate system of the camera corresponding to the corresponding camera parameter group.

Wherein, for the specific implementation manner of A3, reference may be made to step S12, so details are not repeated here.

After the linear parameter set is determined, step A4 is performed.

A4: Establish the corresponding relationship between the linear parameter set and the camera parameter set.

In the embodiment of the present application, the corresponding relationship between the linear parameter group and the camera parameter group is pre-established by detecting the simulation graph, and the camera parameter group can be determined directly by looking up the linear parameter group table when it is necessary to determine the camera parameter group, which greatly improves the camera parameter estimation efficiency.

As an implementation manner, the method may further include steps B1-B3.

B1: When the first linear parameter group does not exist in the corresponding relationship, for each linear parameter group in the corresponding relationship, between the value of the corresponding parameter in the linear parameter group and the first linear parameter group When the differences between are smaller than the target threshold, the linear parameter set is determined to be the second linear parameter set.

In the actual implementation process, B1 can be implemented in the following manner. When the first linear parameter group does not exist in the corresponding relationship, for each linear parameter group in the corresponding relationship, the parameters in the linear parameter group The value of k is made difference with the value of the parameter k in the first linear parameter group, obtains the first difference; And the value of the parameter d in this linear parameter group is made difference with the value of the parameter d in the first linear parameter group , to obtain the second difference, when it is determined that both the first difference and the second difference are smaller than the target threshold, the linear parameter group is determined to be the second linear parameter group; otherwise, it is determined that the linear parameter group is not the second linear group .

B2: From the corresponding relationship, find out a second camera parameter set corresponding to the second linear parameter set.

When the number of the second linear parameter group is one, B2 can be implemented as follows, for the second linear parameter group, according to the position of the second linear parameter group in the corresponding relationship, from the corresponding relationship, find out the The second camera parameter group corresponding to the second linear parameter group.

As an implementation manner, when the number of the second linear parameter groups is at least two, B2 can be implemented in the following manner, for each second linear parameter group, from the corresponding relationship, find out the The second camera parameter group corresponding to the second linear parameter group.

B3: Determine the target camera parameter set according to the second camera parameter set.

When there is one second camera parameter set, B3 may be implemented in the following manner, directly determining the second camera parameter set as the target camera parameter set.

As an implementation manner, when the number of the second linear parameter sets is at least two, B3 may include steps B31-B34.

B31: Obtain the detection frame information of the area where the target is located in the second image; the second image is the image captured by the camera at the second moment; the difference between the first moment and the second moment less than the target time difference.

Wherein, the first moment may be located before the second moment, or may be located after the second moment.

Wherein, the value range of the target time difference may be 0.3 seconds-2 seconds, and the target time difference is determined according to the average normal moving speed of the target and the shooting range that the camera can cover.

It is worth mentioning that the same target is included in the first image and the second image.

Wherein, the specific implementation manner of B31 is the same as that of S11, and details will not be repeated here. It can be understood that the camera, shooting angle, and size of the second image are the same as those of the first image.

After the information of the detection frame of the area where the target is located in the second image is acquired, step B32 is executed.

B32: For each second camera parameter group, based on the second camera parameter group, the information of the detection frame of the target in the first image, and the information of the detection frame of the target in the second image, determine the The final moving speed of the target captured by the camera from the first moment to the second moment.

As an implementation manner, B32 may include steps B321-B324.

B321: From the information of the detection frame of the target in the first image and the information of the detection frame of the target in the second image, determine at least one detection frame information group of the target; each detection frame information group includes: The information of the two detection boxes corresponding to the target.

In the actual implementation process, B321 can be implemented in the following manner. Regarding the information of the detection frame of the target in the first image, according to the position of the detection frame of the target in the first image, determine the distance between the target and the target from the first image. image, and then according to the image of the target and/or the position of the detection frame of the target in the first image, use target tracking technology to determine whether there is an image of the target in the second image, and if yes, the first image will be The detection frame information of the target in the second image and the detection frame information of the target in the second image are divided into detection frame information groups of the target until at least one detection frame information group of the target is determined.

B322: For the information of each detection frame in each detection frame information group, based on the second camera parameter group and the information of the detection frame, determine the position of the target corresponding to the detection frame in the world coordinate system information.

It can be understood that when the camera parameters are determined, the position information of the target corresponding to the detection frame in the world coordinate system can be uniquely determined according to the position and size of the detection frame in the image coordinate system.

In the actual implementation process, B322 can be implemented in the following manner, for each detection frame in each detection frame information group (each detection frame information group includes the information of two detection frames, and the two detection frames correspond to the first The first image and the second image taken at the first moment and the second moment), based on the second camera parameter set and the position and size of the detection frame in its corresponding image, it is determined that the target corresponding to the detection frame is in the world Position information in a coordinate system.

After determining the position information of the target corresponding to the detection frame information group in the world coordinate system at the first moment and the second moment, step B323 is executed.

B323: According to the position information of the target corresponding to each detection frame corresponding to the detection frame information group in the world coordinate system, determine that the corresponding target is between the first moment and the second moment movement speed between.

In the actual implementation process, B323 can be implemented in the following manner. According to the position information of the target in the world coordinate system corresponding to the two detection frames corresponding to the corresponding target, the moving distance of the corresponding target is determined; according to the first moment and the second moment, the time difference between the first moment and the second moment is determined; then, the moving speed of the corresponding target between the first moment and the second moment is obtained by quotienting the moving distance and the time difference.

After determining the moving speed of at least one target between the first moment and the second moment, step B324 is executed.

B324: Obtain the final moving speed according to the moving speed of the at least one target.

When at least one target includes only one target, B324 may directly determine the moving speed of the target as the final moving speed in the following manner.

When the number of at least one target is at least two, B324 may average the determined moving speeds of the at least two targets between the first moment and the second moment in the following manner to obtain the final moving speed.

After the final moving speed is determined, step B33 is executed.

B33: When it is determined that the final moving speed is within the normal moving speed range, determine that the second camera parameter set is qualified.

Wherein, the normal movement speed range is determined according to the target type. For example, the moving speed of a vehicle type object is 20km/h-120km/h.

In most cases, the moving speed of the target is within the normal moving speed range. If the calculated moving speed of the target is not within the normal moving speed range, there is a high probability that the camera parameters are unreasonable. Therefore, in the actual implementation process, B33 can be implemented in the following manner, determine whether the final moving speed is within the normal moving speed range, if it is determined that the final moving speed is within the normal moving speed range, then determine that the second camera parameter set is qualified; otherwise , it is determined that the second camera parameter set is unqualified.

After the qualified second camera parameter set is determined, step B34 is executed.

B34: Determine the target camera parameter set according to the qualified second camera parameter set.

When the number of qualified second camera parameter sets is one, B34 may be implemented in the following manner, directly determining the qualified second camera parameter set as the final camera parameter set of the camera.

As an implementation manner, when the number of qualified second camera parameter groups is at least two, B34 may be implemented in the following manner, respectively calculating the values of corresponding parameters in at least two qualified second camera parameter groups Averaged to get the target camera parameter set.

Specifically, when the number of qualified second camera parameter groups is at least two, average the values of the parameter f in the at least two qualified second camera parameter groups to obtain the parameter f in the target camera parameter group The value of parameter α in at least two qualified second camera parameter groups is averaged to obtain the value of parameter α in the target camera parameter group, and the parameters in at least two qualified second camera parameter groups The value of h is averaged to obtain the value of parameter h in the target camera parameter group.

In this way, when there are multiple camera parameter sets determined according to the corresponding relationship, the unreasonable camera parameter set can be filtered through the index of target speed, and the final camera parameter set can be determined with a reasonable camera parameter set, which greatly improves the camera parameter set. Estimated accuracy.

Please refer to FIG. 3 . FIG. 3 is a structural block diagram of a camera calibration device 300 provided in an embodiment of the present application. The structural block diagram shown in Figure 3 will be described below, and the shown devices may include:

The first acquiring unit 310 may be configured to acquire information of detection frames of at least two target areas in the first image; wherein, the first image is an image captured by the camera at the first moment; each detection frame The information includes: the position and size of the corresponding detection frame in the first image.

The first determining unit 320 may be configured to determine a first linear parameter group according to the information of the detection frames of the at least two targets; wherein, the first linear parameter group is used to characterize the detection frames in the first A linear relationship between position and size in the image.

The first searching unit 330 may be configured to, when the first linear parameter group exists in the predetermined correspondence between the linear parameter group and the camera parameter group, find out the first linear parameter from the corresponding relationship. The first camera parameter group corresponding to the group; wherein, the first camera parameter group includes: the mounting size of the camera, the pitch angle of the camera, and the focal length of the camera.

The first target determining unit 340 may be configured to determine a target camera parameter set according to the first camera parameter set.

As an implementation manner, the device may further include: a camera parameter acquisition unit configured to acquire multiple different camera parameter sets; any two camera parameter sets in the multiple different camera parameter sets correspond to The types of parameters are the same; the simulation diagram generating unit is configured to use the camera parameter group, the size and position information of at least two virtual targets to generate a detection frame simulation diagram for each camera parameter group; wherein, the detection The frame simulation diagram includes: the detection frames of the at least two virtual targets; a linear parameter group determination unit configured to be used in the detection frame simulation graph according to the at least two detection frames in the detection frame simulation graph The position and size determine the linear parameter set; the corresponding relationship establishing unit is configured to establish the corresponding relationship between the linear parameter set and the camera parameter set.

As an implementation manner, the simulation diagram generating unit may include: a size generating unit configured to generate the sizes of the at least two virtual objects according to the size distribution of real objects of the same type as the virtual objects; The information generation unit is configured to generate the position information of the at least two virtual targets; the simulation map generation subunit is configured to use for each camera parameter group, according to the size and position information of the at least two virtual targets , to generate the simulation diagram of the detection frame.

As an implementation manner, the camera parameter acquisition unit may be configured to perform discrete values of each camera parameter within a value range based on a target sampling interval to obtain the plurality of different camera parameter groups.

As an implementation manner, the device may further include: a second determination unit configured to, when the first linear parameter group does not exist in the corresponding relationship, for each linear parameter group in the corresponding relationship , when the differences between the values of the corresponding parameters in the linear parameter group and the first linear parameter group are smaller than the target threshold, the linear parameter group is determined to be the second linear parameter group; the second search unit is configured to use From the corresponding relationship, find out the second camera parameter set corresponding to the second linear parameter set; the second target determination unit is configured to determine the target camera according to the second camera parameter set parameter group.

As an implementation manner, when the number of the second linear parameter groups is at least two, the second search unit may be configured to specifically for each second linear parameter group, from the corresponding relationship, Find out the second camera parameter group corresponding to the second linear parameter group;

Correspondingly, the second target determining unit may include:

The second acquiring unit is configured to acquire the information of the detection frame of the area where the target is located in the second image; the second image is an image captured by the camera at a second moment; the first moment and the second The difference between the two moments is less than the target time difference; the speed determination unit is configured to, for each second camera parameter set, based on the second camera parameter set, the information of the detection frame of the target in the first image, and The information of the detection frame of the target in the second image determines the final moving speed of the target captured by the camera from the first moment to the second moment; the screening unit is configured to determine the When the final moving speed is not within the normal moving speed range, it is determined that the second camera parameter set is unqualified; otherwise, it is determined that the second camera parameter set is qualified; the second target determination subunit is configured to parameter group, to determine the target camera parameter group.

As an implementation manner, the speed determining unit may include: an information group determining unit configured to obtain the information of the detection frame of the target in the first image and the information of the detection frame of the target in the second image Among them, the detection frame information group of at least one target is determined; each detection frame information group includes: information of two detection frames corresponding to the target; the position determination unit is configured to be used for each detection frame information group The information of the detection frame, based on the second camera parameter group and the information of the detection frame, determines the position information of the target corresponding to the detection frame in the world coordinate system; the moving speed determination unit is configured to use the The position information of the target corresponding to each detection frame corresponding to the detection frame information group in the world coordinate system determines the moving speed of the corresponding target between the first moment and the second moment a speed subunit configured to obtain the final moving speed according to the moving speed of the at least one target.

As an implementation manner, the second target determining unit may also be configured to, when the number of qualified second camera parameter groups is at least two, respectively set corresponding The values of the parameters are averaged to obtain the target camera parameter set.

For the process of realizing the respective functions by each functional unit in this embodiment, please refer to the content represented in the embodiment shown in the foregoing FIGS. 1-2 , which will not be repeated here.

Please refer to FIG. 4. FIG. 4 is a schematic structural diagram of an electronic device 400 provided by an embodiment of the present application. The electronic device 400 may be a personal computer, a tablet computer, a smart phone, a personal digital assistant (personal digital assistant, PDA) and the like.

The electronic device 400 may include: a memory 402, a processor 401, a communication interface 403, and a communication bus, and the communication bus is used to implement connection and communication of these components.

The memory 402 can be used to store various data such as the camera calibration method provided by the embodiment of the present application and the calculation program instructions corresponding to the device, wherein the memory 402 can be, but not limited to, a random access memory, a read-only memory (Read Only Memory, ROM), Programmable Read-Only Memory (Programmable Read-Only Memory, PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (Electric Erasable Programmable Read -Only Memory, EEPROM), etc.

The processor 401 can be used to read and execute the computer program instructions corresponding to the camera calibration method and device stored in the memory, so as to obtain the information of the detection frame of the area where at least two targets are located in the first image; wherein, the first The image is an image captured by the camera at the first moment; the information of each detection frame includes: the position and size of the corresponding detection frame in the first image; according to the information of the detection frames of the at least two targets, determine The first linear parameter group is obtained; wherein, the first linear parameter group is used to characterize the linear relationship between the position and size of the detection frame in the first image; the correspondence between the predetermined linear parameter group and the camera parameter group When the first linear parameter group exists in the relationship, the first camera parameter group corresponding to the first linear parameter group is found from the corresponding relationship; wherein, the first camera parameter group includes: The erection height, the pitch angle of the camera, and the focal length of the camera; determine a target camera parameter set according to the first camera parameter set.

Wherein, the processor 401 may be an integrated circuit chip, which has a signal processing capability. Above-mentioned processor 401 can be general purpose processor, comprises CPU, network processor (Network Processor, NP) etc.; Can also be digital signal processor (DSP), application specific integrated circuit (ASIC), field programmable gate array (FPGA) ) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. Various methods, steps, and logic block diagrams disclosed in the embodiments of the present application may be implemented or executed. A general-purpose processor may be a microprocessor, or the processor may be any conventional processor, or the like.

The communication interface 403 can be used to receive or send data.

In addition, the embodiment of the present application also provides a computer-readable storage medium, in which a computer program is stored, and when the computer program is run on the computer, the computer is made to execute any one of the implementations of the present application. method provided by the method.

To sum up, the camera calibration method, device, electronic equipment, and storage medium proposed by each embodiment of the present application are used to obtain the information of the detection frames of at least two target areas in the first image captured by the camera, and each detection frame The information includes: the position and size of the corresponding detection frame in the first image; according to the information of the detection frames of at least two targets, determine the linear relationship between the position and size of the detection frame in the first image The first linear parameter set, and then from the predetermined correspondence between the linear parameter set and the camera parameter set, quickly find out the first camera parameter set corresponding to the first linear parameter set, and according to the first camera parameter set The target camera parameter set is determined, and the camera calibration can be completed without placing a calibration object or controlling the camera movement during the whole process.

In the embodiments provided in this application, it should be understood that the disclosed devices and methods may also be implemented in other ways. The device embodiments described above are only illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the architecture, functions and possible implementations of devices, methods and computer program products according to multiple embodiments of the present application. operate. In this regard, each block in a flowchart or block diagram may represent a module, program segment, or part of code that includes one or more Executable instructions. It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks in succession may, in fact, be executed substantially concurrently, or they may sometimes be executed in the reverse order, depending upon the functionality involved. It should also be noted that each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations, can be implemented by a dedicated hardware-based device that performs the specified function or action , or may be implemented by a combination of dedicated hardware and computer instructions.

In addition, each functional module in each embodiment of the present application may be integrated to form an independent part, each module may exist independently, or two or more modules may be integrated to form an independent part.

Industrial Applicability

The present application provides a camera calibration method, device, electronic equipment, and storage medium, including: acquiring the detection frame information of at least two target areas in the first image; the first image is the image captured by the camera at the first moment; The information of each detection frame includes the position and size of the corresponding detection frame in the first image; according to the information of the detection frames of at least two targets, a first linear parameter group is determined; the first linear parameter group is used to characterize the detection frame The linear relationship between the position and size in the first image; when the first linear parameter group exists in the corresponding relationship between the predetermined linear parameter group and the camera parameter group, find out the relationship with the first linear parameter from the corresponding relationship The first camera parameter group corresponding to the first camera parameter group; the first camera parameter group includes the erection height of the camera, the pitch angle of the camera and the focal length of the camera, and the target camera parameter group is determined according to the first camera parameter group, without placing calibration objects or controlling camera movement. Can complete camera calibration.

In addition, it can be understood that the camera calibration method, device, electronic device, and storage medium of the present application are reproducible, and can be applied in various applications. For example, the camera calibration method, device, electronic equipment, and storage medium of the present application may be applied in the technical field of image processing and the like.

Claims

A camera calibration method, characterized in that the method comprises:

Acquiring the information of the detection frame of the area where at least two targets are located in the first image; wherein, the first image is an image captured by the camera at the first moment; the information of each detection frame includes: the corresponding detection frame is in the position and size in the first image;

According to the information of the detection frames of the at least two targets, a first linear parameter group is determined; wherein the first linear parameter group is used to characterize the linear relationship between the position and size of the detection frames in the first image ;

When the first linear parameter group exists in the predetermined correspondence between the linear parameter group and the camera parameter group, from the correspondence, find out the first camera parameter group corresponding to the first linear parameter group; wherein, The first camera parameter group includes: the erection height of the camera, the pitch angle of the camera, and the focal length of the camera;

A target camera parameter set is determined according to the first camera parameter set; wherein, the target camera parameter set is a camera parameter set corresponding to the camera that captures the first image.
The method according to claim 1, wherein the correspondence between the predetermined linear parameter set and the camera parameter set is obtained through the following steps:

Acquiring multiple different camera parameter sets; the types of parameters corresponding to any two camera parameter sets in the multiple different camera parameter sets are the same;

For each camera parameter group, use the camera parameter group, the size and position information of at least two virtual targets to generate a detection frame simulation diagram; wherein, the detection frame simulation diagram includes: the detection of the at least two virtual targets frame;

Determine a linear parameter set according to the positions and sizes of at least two detection frames in the detection frame simulation graph in the detection frame simulation graph;

A corresponding relationship between the linear parameter set and the camera parameter set is established.
The method according to claim 2, wherein, for each camera parameter group, using the camera parameter group, the size and position information of at least two virtual targets to generate a detection frame simulation diagram, comprising:

generating sizes of the at least two virtual objects based on a size distribution of real objects of the same type as the virtual objects;

generating location information of the at least two virtual targets;

For each camera parameter set, the detection frame simulation diagram is generated according to the camera parameter set, the size and position information of the at least two virtual objects.
The method according to claim 2 or 3, wherein obtaining a plurality of different camera parameter groups includes:

Based on the target sampling interval, each camera parameter is discretely valued within a value range to obtain the plurality of different camera parameter groups.
The method according to any one of claims 1 to 4, wherein when the first linear parameter group does not exist in the correspondence, the method further comprises:

For each linear parameter group in the corresponding relationship, when the difference between the value of the linear parameter group and the value of the corresponding parameter in the first linear parameter group is smaller than the target threshold, determine that the linear parameter group is the second Linear parameter set;

From the corresponding relationship, find out the second camera parameter group corresponding to the second linear parameter group;

The target camera parameter set is determined according to the second camera parameter set.
The method according to claim 5, wherein when the number of the second linear parameter groups is at least two, the second linear parameter group corresponding to the second linear parameter group is found out from the corresponding relationship. Camera parameter set, including:

For each second linear parameter group, from the corresponding relationship, find out the second camera parameter group corresponding to the second linear parameter group;

Correspondingly, according to the second camera parameter set, determining the target camera parameter set includes:

Acquiring the information of the detection frame of the area where the target is located in the second image; the second image is the image captured by the camera at the second moment; the difference between the first moment and the second moment is smaller than the target Time difference;

For each second camera parameter group, based on the second camera parameter group, the information of the detection frame of the target in the first image, and the information of the detection frame of the target in the second image, it is determined that the camera is in The final moving speed of the target photographed from the first moment to the second moment;

When it is determined that the final moving speed is within the normal moving speed range, it is determined that the second camera parameter set is qualified;

The target camera parameter set is determined according to the qualified second camera parameter set.
The method according to claim 6, characterized in that, based on the second camera parameter group, the information of the detection frame of the target in the first image, and the information of the detection frame of the target in the second image, determine The final moving speed of the target captured by the camera from the first moment to the second moment includes:

From the information of the detection frame of the target in the first image and the information of the detection frame of the target in the second image, at least one detection frame information group of the target is determined; each detection frame information group includes: a corresponding target The information of the two detection boxes of ;

For the information of each detection frame in each detection frame information group, based on the second camera parameter group and the information of the detection frame, determine the position information of the target corresponding to the detection frame in the world coordinate system;

According to the position information of the target corresponding to each detection frame corresponding to the detection frame information group in the world coordinate system, determine the position of the corresponding target between the first moment and the second moment Moving speed;

The final moving speed is obtained according to the moving speed of the at least one target.
The method according to claim 6 or 7, wherein said determining the target camera parameter set according to the qualified second camera parameter set comprises:

When the number of qualified second camera parameter groups is at least two, the values of the corresponding parameters in at least two qualified second camera parameter groups are respectively averaged to obtain the target camera parameter group.
A camera calibration device, characterized in that the camera calibration device comprises:

The first acquiring unit is configured to acquire the information of the detection frames in the area where at least two targets are located in the first image; wherein, the first image is an image captured by the camera at the first moment; the information of each detection frame Including: the position and size of the corresponding detection frame in the first image;

The first determination unit is configured to determine a first linear parameter set according to the information of the detection frames of the at least two targets; wherein, the first linear parameter set is used to characterize the detection frames in the first image The linear relationship between the position and size of ;

The first search unit is configured to, when the first linear parameter set exists in the predetermined correspondence between the linear parameter set and the camera parameter set, find out from the correspondence that the first linear parameter set corresponds to the first linear parameter set. The first camera parameter group; wherein, the first camera parameter group includes: the mounting size of the camera, the pitch angle of the camera and the focal length of the camera;

The first target determination unit is configured to determine a target camera parameter set according to the first camera parameter set.
An electronic device, characterized in that it includes a memory and a processor, and computer program instructions are stored in the memory, and when the computer program instructions are read and executed by the processor, the steps in claims 1 to 8 are executed. any one of the methods described.
A computer-readable storage medium, characterized in that computer program instructions are stored on the storage medium, and when the computer program instructions are read and run by a computer, the computer program described in any one of claims 1 to 8 is executed. described method.
A computer program product, characterized in that the computer program product includes a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 8 is realized.