CN110378963A - Camera parameter scaling method and device - Google Patents
Camera parameter scaling method and device Download PDFInfo
- Publication number
- CN110378963A CN110378963A CN201811477402.5A CN201811477402A CN110378963A CN 110378963 A CN110378963 A CN 110378963A CN 201811477402 A CN201811477402 A CN 201811477402A CN 110378963 A CN110378963 A CN 110378963A
- Authority
- CN
- China
- Prior art keywords
- camera
- model
- parameter
- coordinate system
- calibration
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Studio Devices (AREA)
Abstract
Present disclose provides a kind of camera parameter scaling methods, comprising: has trained obtained optimal model parameters to obtain the corresponding original model parameter of this camera based on other same type cameras;Sample set is obtained, the sample set includes nominal data of multiple calibration points about this camera;Default imaging model is trained based on the sample set and the original model parameter, obtains the corresponding optimal model parameters of this camera;The calibrating parameters of this camera are obtained based on the corresponding optimal model parameters of this camera.The disclosure additionally provides a kind of camera parameter caliberating device, a kind of computer equipment and a kind of computer readable storage medium.
Description
Technical field
This disclosure relates to field of computer technology, more particularly, to a kind of camera parameter scaling method and device.
Background technique
The camera parameter scaling scheme of the prior art independently carries out parameter calibration to each camera, demarcates in camera parameter
In the process it is generally necessary to which a large amount of nominal data is as support, due to either offline or online acquisition different perspectives calibration
Data need individually to carry out each camera, are all to lead to camera parameter calibration process inefficiency than relatively time-consuming, unfavorable
In the deployment of mass product.
Summary of the invention
In view of this, present disclose provides a kind of camera parameter calibration high-efficient, low to the quantitative requirement of nominal data
Method and apparatus.
An aspect of this disclosure provides a kind of camera parameter scaling method, comprising: based on other same type cameras
The optimal model parameters that training obtains obtain the corresponding original model parameter of this camera;Sample set is obtained, which includes more
Nominal data of a calibration point about this camera;Default imaging model is trained based on sample set and original model parameter,
Obtain the corresponding optimal model parameters of this camera;The calibration ginseng of this camera is obtained based on the corresponding optimal model parameters of this camera
Number.
In accordance with an embodiment of the present disclosure, above-mentioned to have trained obtained optimal model parameters to obtain based on other same type cameras
The corresponding original model parameter of this camera include: other any same type cameras have been trained obtained optimal model parameters as
The corresponding original model parameter of this camera;Alternatively, other multiple same type cameras have been trained obtained optimal model parameters into
Row averagely obtains mean value model parameter, using the mean value model parameter as the corresponding original model parameter of this camera.
In accordance with an embodiment of the present disclosure, each calibration point in sample set includes: the mark about the nominal data of this camera
Pinpoint the location information of location information and the corresponding picture point of the calibration point in pixel coordinate system in world coordinate system.
Default imaging model table levies the transformation model from world coordinate system to pixel coordinate system.
In accordance with an embodiment of the present disclosure, above-mentioned acquisition sample set includes: to establish world coordinate system and pixel coordinate system;For
Any calibration point obtains location information of the calibration point in world coordinate system;The image for shooting the calibration point, from described image
The corresponding characteristic point of middle extraction calibration point obtains location information of this feature point in pixel coordinate system;The calibration point is existed
Location information of the location information characteristic point corresponding with the calibration point in pixel coordinate system in world coordinate system is as the mark
Pinpoint the nominal data about this camera;Multiple calibration points constitute the sample set about the nominal data of this camera.
In accordance with an embodiment of the present disclosure, above-mentioned that default imaging model is trained based on sample set and original model parameter
It include: the initial model that the default imaging model is constituted using original model parameter;Based on least square method, the sample is utilized
This collection is iterated optimization to the initial model, obtains optimal model parameters.
In accordance with an embodiment of the present disclosure, above-mentioned that default imaging model is instructed based on sample set and original model parameter
Practice, it includes training set and test set that obtain the corresponding optimal model parameters of this camera, which include: sample set,;Based on training set and initially
Model parameter is trained default imaging model;It is verified using the model that test set obtains each training, works as verifying
When as a result meeting preset condition, using acquired model as optimal models, using the parameter of the model as optimal model parameters, when
Verification result is unsatisfactory for preset condition, continues to be trained obtained model based on the training set.
In accordance with an embodiment of the present disclosure, the above-mentioned model obtained using the test set to each training carries out verifying packet
It includes: calculating the re-projection error for the model that training obtains every time using test set, when re-projection error is less than preset threshold, really
Determine verification result and meet preset condition, when re-projection error is not less than preset threshold, determines that verification result is unsatisfactory for default item
Part.
In accordance with an embodiment of the present disclosure, above-mentioned to obtain the calibration ginseng of this camera based on the corresponding optimal model parameters of this camera
Number includes: the intrinsic parameter of calibration and calibration distortion parameter that this camera is obtained from the corresponding optimal model parameters of this camera.
Another aspect of the disclosure provides a kind of camera parameter caliberating device, including the first acquisition module, second obtains
Modulus block, training module and demarcating module.First acquisition module for trained based on other same type cameras obtain it is optimal
Model parameter obtains the corresponding original model parameter of this camera.For obtaining sample set, which includes second acquisition module
Nominal data of multiple calibration points about this camera.Training module is used for based on sample set and original model parameter to default imaging
Model is trained, and obtains the corresponding optimal model parameters of this camera.Demarcating module is used to be based on the corresponding optimal mould of this camera
Shape parameter obtains the calibrating parameters of this camera.
In accordance with an embodiment of the present disclosure, the first acquisition module has trained obtained optimal models based on other same type cameras
It includes: the first acquisition module for having instructed other any same type cameras that parameter, which obtains the corresponding original model parameter of this camera,
The optimal model parameters got are as the corresponding original model parameter of this camera;Alternatively, for other multiple same type phases
Machine has trained obtained optimal model parameters averagely to be obtained mean value model parameter, using the mean value model parameter as this phase
The corresponding original model parameter of machine.
In accordance with an embodiment of the present disclosure, each calibration point in sample set includes: the mark about the nominal data of this camera
Pinpoint the location information of location information and the corresponding picture point of the calibration point in pixel coordinate system in world coordinate system.
Default imaging model table levies the transformation model from world coordinate system to pixel coordinate system.
In accordance with an embodiment of the present disclosure, it includes: the second acquisition module for establishing generation that the second acquisition module, which obtains sample set,
Boundary's coordinate system and pixel coordinate system;For any calibration point, location information of the calibration point in world coordinate system is obtained;Shooting
The image of the calibration point extracts the corresponding characteristic point of the calibration point from described image, obtains this feature point in pixel coordinate system
In location information;Location information of the calibration point in world coordinate system characteristic point corresponding with the calibration point is sat in pixel
Nominal data of the location information as the calibration point about this camera in mark system;Calibration number of multiple calibration points about this camera
According to the composition sample set.
In accordance with an embodiment of the present disclosure, training module is based on the sample set and the original model parameter to default imaging
It includes: training module for constituting the initial model of the default imaging model using original model parameter that model, which is trained,;
Based on least square method, optimization is iterated to the initial model using the sample set, obtains optimal model parameters.
In accordance with an embodiment of the present disclosure, sample set includes training set and test set.Training module be based on the sample set and
The original model parameter is trained default imaging model, and obtaining the corresponding optimal model parameters of this camera includes: training
Module is used to be trained default imaging model based on the training set and the original model parameter;Utilize the test set
The model obtained to each training is verified, when verification result meets preset condition, using acquired model as optimal mould
Type continues to be trained obtained model based on the training set when verification result is unsatisfactory for preset condition.
In accordance with an embodiment of the present disclosure, training module is verified using the model that the test set obtains each training
It include: the re-projection error that training module is used to calculate the model that training obtains every time using the test set, when re-projection misses
When difference is less than preset threshold, determine that verification result meets preset condition, when re-projection error is not less than preset threshold, determination is tested
Card result is unsatisfactory for preset condition.
In accordance with an embodiment of the present disclosure, demarcating module obtains the mark of this camera based on the corresponding optimal model parameters of this camera
Determine parameter include: demarcating module be used for obtained from the corresponding optimal model parameters of this camera this camera the intrinsic parameter of calibration and
Demarcate distortion parameter.
Another aspect of the present disclosure provides a kind of computer equipment, including memory, processor and is stored in memory
Computer program that is upper and can running on a processor, the processor realize method as described above when executing described program.
Another aspect of the present disclosure provides a kind of computer readable storage medium, is stored with computer executable instructions,
Described instruction is when executed for realizing method as described above.
Another aspect of the present disclosure provides a kind of computer program, and the computer program, which includes that computer is executable, to be referred to
It enables, described instruction is when executed for realizing method as described above.
In accordance with an embodiment of the present disclosure, can at least be partially solved/mitigation/inhibit/or even avoid existing camera parameter
The problem that nominal data needed for scaling scheme is more, the nominal time is long, and therefore may be implemented to carry out in the camera to same type
When parameter calibration, the parameter optimization training of each camera is no longer wasted time based on the training result of other cameras to do it
The optimization process that his camera has been completed, it is high-efficient, it is low to the quantitative requirement of the nominal data as sample, it is suitable for extensive
Camera calibration deployment beneficial effect.
Detailed description of the invention
By referring to the drawings to the description of the embodiment of the present disclosure, the above-mentioned and other purposes of the disclosure, feature and
Advantage will be apparent from, in the accompanying drawings:
Fig. 1 diagrammatically illustrates the example that can apply camera parameter scaling method and device according to the embodiment of the present disclosure
Property system architecture;
Fig. 2 diagrammatically illustrates the flow chart of camera parameter scaling method according to an embodiment of the present disclosure;
Fig. 3 A diagrammatically illustrates the flow chart of camera parameter scaling method according to another embodiment of the present disclosure;
Fig. 3 B diagrammatically illustrates the schematic diagram of pixel coordinate system and image coordinate system according to an embodiment of the present disclosure;
Fig. 3 C diagrammatically illustrates the schematic diagram of the pin-hole imaging model of camera according to an embodiment of the present disclosure;
Fig. 3 D diagrammatically illustrates the camera of camera parameter scaling method and the prior art according to an embodiment of the present disclosure
The optimization process comparison diagram of parameter calibration method;
Fig. 4 diagrammatically illustrates the block diagram of camera parameter caliberating device according to an embodiment of the present disclosure;And
Fig. 5 diagrammatically illustrates the block diagram of computer equipment according to an embodiment of the present disclosure.
Specific embodiment
Hereinafter, will be described with reference to the accompanying drawings embodiment of the disclosure.However, it should be understood that these descriptions are only exemplary
, and it is not intended to limit the scope of the present disclosure.In the following detailed description, to elaborate many specific thin convenient for explaining
Section is to provide the comprehensive understanding to the embodiment of the present disclosure.It may be evident, however, that one or more embodiments are not having these specific thin
It can also be carried out in the case where section.In addition, in the following description, descriptions of well-known structures and technologies are omitted, to avoid
Unnecessarily obscure the concept of the disclosure.
Term as used herein is not intended to limit the disclosure just for the sake of description specific embodiment.It uses herein
The terms "include", "comprise" etc. show the presence of the feature, step, operation and/or component, but it is not excluded that in the presence of
Or add other one or more features, step, operation or component.
There are all terms (including technical and scientific term) as used herein those skilled in the art to be generally understood
Meaning, unless otherwise defined.It should be noted that term used herein should be interpreted that with consistent with the context of this specification
Meaning, without that should be explained with idealization or excessively mechanical mode.
It, in general should be according to this using statement as " at least one in A, B and C etc. " is similar to
Field technical staff is generally understood the meaning of the statement to make an explanation (for example, " system at least one in A, B and C "
Should include but is not limited to individually with A, individually with B, individually with C, with A and B, with A and C, have B and C, and/or
System etc. with A, B, C).Using statement as " at least one in A, B or C etc. " is similar to, generally come
Saying be generally understood the meaning of the statement according to those skilled in the art to make an explanation (for example, " having in A, B or C at least
One system " should include but is not limited to individually with A, individually with B, individually with C, with A and B, have A and C, have
B and C, and/or the system with A, B, C etc.).
Embodiment of the disclosure provides a kind of camera parameter scaling method and device.This method includes original model parameter
Acquisition process, sample acquisition process, training process and calibrating parameters determination process.In original model parameter acquisition process, base
It has trained obtained optimal model parameters to obtain the corresponding original model parameter of this camera in other same type cameras, has been obtained in sample
During taking, the sample set being made of multiple calibration points about the nominal data of this camera is obtained, is then based on initial model ginseng
Several and sample set is trained default imaging model to obtain optimal model parameters, finally in calibrating parameters determination process, base
The calibrating parameters of this camera are obtained in obtained optimal model parameters.
Fig. 1 diagrammatically illustrates the example that can apply camera parameter scaling method and device according to the embodiment of the present disclosure
Property system architecture 100.It should be noted that be only the example that can apply the system architecture of the embodiment of the present disclosure shown in Fig. 1, with
The technology contents of the disclosure are helped skilled in the art to understand, but are not meant to that the embodiment of the present disclosure may not be usable for other
Equipment, system, environment or scene.
As shown in Figure 1, system architecture 100 may include camera 101 and calibration object 104 according to this embodiment.Camera 101
In camera parameter calibration process, 101 pairs of calibration objects 102 of camera are shot, to obtain according to calibration object 102 and shooting
The relevant information of corresponding picture carry out the parameter of calibration for cameras 101.Camera 101 can pass through various sides with other same type cameras
Formula interacts, then camera 101 can get the calibrated parameter information of camera and be used for reference use.
Camera 101 can be various types of cameras, the camera components being also possible in various types of electronic equipments, only
If the camera with shooting function can be with herein with no restrictions.
It should be understood that the number of the camera in Fig. 1 is only schematical.According to needs are realized, arbitrary number can have
Purpose camera, to realize large-scale camera parameter calibration.
Fig. 2 diagrammatically illustrates the flow chart of camera parameter scaling method according to an embodiment of the present disclosure.
As shown in Fig. 2, this method is included in operation S201, obtained optimal models have been trained based on other same type cameras
Parameter obtains the corresponding original model parameter of this camera.
In this operation, same type camera can be the camera of identical type selecting, can be the camera of same model, i.e. hardware is matched
It sets close so that the camera that camera parameter is closer to.
Then, in operation S202, sample set is obtained, which includes calibration number of multiple calibration points about this camera
According to.
Then, in operation S203, default imaging model is trained based on sample set and original model parameter, obtains this
The corresponding optimal model parameters of camera.
In operation S204, the calibrating parameters of this camera are obtained based on the corresponding optimal model parameters of this camera.
As it can be seen that the calibrating parameters that the camera of same type is utilized in method shown in Fig. 2 have difference but difference is lesser
Feature, when carrying out parameter calibration to camera, other cameras with this camera same type trained obtain it is optimal
It is trained on the basis of model parameter, obtains the corresponding optimal model parameters of this camera, for any camera, the camera
Corresponding optimal model parameters are able to reflect out the optimal calibrating parameters of the camera, and this programme is in other same type
Optimization obtains the optimal calibrating parameters of this camera on the basis of the optimal calibrating parameters of camera, and this camera is obtained excellent
The calibrating parameters of change can also be used as optimization basis of the camera of other same types when carrying out parameter calibration.Foundation this programme,
When the camera to same type carries out parameter calibration, the parameter optimization training of each camera is using the training result of other cameras as base
Plinth no longer wastes time to do the optimization process that other cameras have been completed, high-efficient, to the number of the nominal data as sample
Amount requires low calibration deployment suitable for large-scale camera.
In one embodiment of the present disclosure, the operation S201 of method shown in Fig. 2 has been trained based on other same type cameras
It includes: to have trained other any same type cameras that obtained optimal model parameters, which obtain the corresponding original model parameter of this camera,
Obtained optimal model parameters are as the corresponding original model parameter of this camera;Alternatively, having been instructed to other multiple same type cameras
The optimal model parameters got are averagely obtained mean value model parameter, corresponding using the mean value model parameter as this camera
Original model parameter.In the present embodiment, can choose other any same type cameras has trained obtained optimal models to join
Number is used as the corresponding original model parameter of this camera, other multiple same type cameras can also have been trained the equal of obtained parameter
It is worth the original model parameter as this camera, the second way can make selected original model parameter from the statistical significance
Bigger probability further increases the parameter calibration efficiency of camera close to the parameter of this camera.
In one embodiment of the present disclosure, for this camera of progress parameter calibration, world coordinate system is established, with the world
The spatial position of camera and subject can be described on the basis of coordinate system, and, establish pixel coordinate system, pixel coordinate system
The arranging situation of pixel in the imaging sensor of this camera is reflected, the picture that can describe subject is being taken in picture
Position.It include calibration number of multiple calibration points about this camera in sample set acquired in the operation S202 of method shown in Fig. 2
According to each calibration point includes: location information of the calibration point in world coordinate system about the nominal data of this camera, and should
Location information of the corresponding picture point of calibration point in pixel coordinate system, that is to say, that each sample in sample set reflects one
The corresponding relationship of object point and picture point of a calibration point under the shooting of this camera, and the parameter of camera exactly influences the corresponding relationship
Most important factor.And the default imaging model table for being trained is levied from world coordinate system to pixel coordinate system
Transformation model, the then optimal models being trained based on sample set and original model parameter to default imaging model are established
The object and its accurate transformation model for being formed by image on the image sensor being taken, in optimal models
Parameter is acquired optimal model parameters, reflects the parameter of this accurate camera.
Specifically, as an optional embodiment, it is above-mentioned establish world coordinate system and pixel coordinate system after, it is above-mentioned
Operating S202 and obtaining the process of sample set may include: to obtain the calibration point in world coordinate system for any calibration point
Location information shoots the image of the calibration point, and the corresponding characteristic point of the calibration point is extracted from the image, obtains this feature point and exists
Location information in pixel coordinate system, by location information of the calibration point in world coordinate system feature corresponding with the calibration point
Location information nominal data as the calibration point about this camera of the point in pixel coordinate system.One calibration point is about this phase
The nominal data of machine constitutes sample set about the nominal data of this camera as a sample, multiple calibration points.
In one embodiment of the present disclosure, the operation S203 of method shown in Fig. 2 is based on sample set and original model parameter
Be trained to default imaging model includes: the initial model that the default imaging model is constituted using original model parameter, base
In least square method, optimization is iterated to the initial model using the sample set, obtains optimal model parameters.This implementation
Example is iterated optimization based on least square method during model training, more meets the camera calibration aspect of model, helps to mention
High optimization efficiency.
In one embodiment of the present disclosure, the operation S203 of method shown in Fig. 2 is based on sample set and original model parameter
Default imaging model is trained, it includes training set and survey that obtain the corresponding optimal model parameters of this camera, which include: sample set,
Examination collection, is trained default imaging model based on training set and original model parameter, using the test set to each training
Obtained model is verified, when verification result meets preset condition, using acquired model as optimal models, by the model
Parameter as optimal model parameters, when verification result is unsatisfactory for preset condition, continue based on the training set to obtained
Model is trained.As it can be seen that the model training process of the present embodiment is using the training set in sample set with having supervision to first
Beginning model parameter constantly optimizes, and is verified using the model that the test set in sample set obtains every suboptimization, directly
Meet preset condition to being optimized to, reach convergence, does not continue to optimize, obtain optimal models, obtain optimal model parameters,
To obtain optimal camera calibration parameter.
Specifically, as an optional embodiment, the above-mentioned model that each training is obtained using the test set into
Row verifying includes: the re-projection error that the model that training obtains every time is calculated using the test set, and re-projection error indicates mark
The theoretical picture point of fixed point and the error of the measurement point on image work as re-projection error for the evaluation criterion as calibration effect
It when less than preset threshold, determines that verification result meets preset condition, when re-projection error is not less than preset threshold, determines verifying
As a result it is unsatisfactory for preset condition.The present embodiment when whether the obtained model of verifying training is the corresponding optimal models of this camera,
It is judgment criteria using the re-projection error of test set computation model, specifically can use the calibration point in test set in the world
Location information of the theoretical picture point that positional information calculation in coordinate system is obtained by the model conversion in pixel coordinate system, then
It is sat with corresponding theoretical picture point in pixel using location information of the corresponding picture point of calibration point in test set in pixel coordinate system
Location information in mark system is compared, and obtains re-projection error, and the program meets the camera calibration aspect of model, is facilitated very fast
Find the optimal models for meeting demand.
In one embodiment of the present disclosure, the operation S204 of method shown in Fig. 2 is based on the corresponding optimal models of this camera
Parameter obtain this camera calibrating parameters include: obtained from the corresponding optimal model parameters of this camera this camera calibration it is intrinsic
Parameter and calibration distortion parameter.Wherein, the calibration intrinsic parameter characterization of camera some build-in attributes of camera, such as camera
Focal length, principal point for camera (main shaft of camera and the intersection point as plane), the pixel of camera and size of true environment etc., phase
The calibration distortion parameter of machine characterizes the distortion parameter of camera, such as radial distortion, tangential distortion etc., demarcate intrinsic parameter and
The characterized inner parameter of camera of distortion parameter is demarcated, although cannot be complete with the corresponding inner parameter of actual camera
It is identical, but by the optimization of this programme can obtain it is accurate as far as possible, close to actual nominal data.
Below with reference to Fig. 3 A~Fig. 3 D, method shown in Fig. 2 is described further in conjunction with specific embodiments.
Fig. 3 A diagrammatically illustrates the flow chart of camera parameter scaling method according to another embodiment of the present disclosure.
As shown in Figure 3A, this method includes operation S301~operation S308.
In operation S301, the default imaging model of the corresponding original model parameter of this camera and camera is obtained.
Wherein, in order to improve the efficiency of camera parameter calibration, nominal data needed for reduction quantity can by it is any its
His same type camera has trained obtained optimal model parameters as the corresponding original model parameter of this camera, utilizes same type phase
The feature of the parameter approximation of machine, it is subsequent to be trained on the basis of same type camera training.
Since the essence of the process of camera shooting object generation image is will to be located at the subject in world coordinate system to turn
It is changed to the picture being located in pixel coordinate system, world coordinate system is a three-dimensional cartesian coordinate system, and phase can be described on the basis of it
The relative tertiary location of machine and subject, pixel coordinate system are a two-dimensional Cartesian coordinate systems, reflect camera CCD/CMOS
The arranging situation of pixel in chip, therefore, the default imaging model table of camera have been levied from world coordinate system to pixel coordinate system
Conversion process.
Conversion from world coordinate system to pixel coordinate system needs first to be converted to camera coordinates system from world coordinate system, then from phase
Machine coordinate system is converted to pixel coordinate system, introduces in detail below:
Establish world coordinate system (O-XwYwZw), camera coordinates system (C-XYZ) and pixel coordinate system (o-uv).Camera coordinates
System is also a three-dimensional cartesian coordinate system, and origin is the optical center C of camera at, and X-axis and Y-axis are divided than to be parallel with the both sides of image planes, Z
Axis is the optical axis of camera.
World coordinate system is now converted into camera coordinates system:
The wherein spin matrix that R is 3 × 3, the translation vector that t is 3 × 1, (X, Y, Z, 1)TFor the odd times of camera coordinates system
Coordinate, (Xw, Yw, Zw, 1)TFor the odd times coordinate of world coordinate system.
Camera coordinates system is converted into pixel coordinate system again:
The first step, pixel coordinate system are unfavorable for coordinate transform, it is therefore desirable to establish image coordinate system (p-xy).
Fig. 3 B diagrammatically illustrates the schematic diagram of pixel coordinate system and image coordinate system according to an embodiment of the present disclosure.
As shown in Figure 3B, the origin o of pixel coordinate system is located at the upper left corner of image, u axis and v axis respectively with the both sides of image planes
In parallel, the unit of reference axis is pixel in pixel coordinate system.The origin p of image coordinate system is the intersection point of camera optical axis and image planes,
That is the principal point of camera, positioned at the center of image, x-axis and y-axis are parallel with u axis and v axis respectively, the list of reference axis in image coordinate system
Position is rice or millimeter etc..Pixel coordinate system and image coordinate system are substantially translation relation, can be converted by translation,
Image coordinate system is then converted into pixel coordinate system:
Wherein, dx, dy are respectively physical size of the pixel on x, y-axis direction, (u0, v0) indicate principal point coordinate.
Camera coordinates system is converted to image coordinate system by second step:
Fig. 3 C diagrammatically illustrates the schematic diagram of the pin-hole imaging model of camera according to an embodiment of the present disclosure.
As shown in Figure 3 C, the pin-hole imaging model of camera reflects camera coordinates system (C-XYZ) and image coordinate system (p-
Xy transformation relation), for example, any point M corresponds to the picture point in image coordinate system (p-xy) in camera coordinates system (C-XYZ)
The line of m, M and camera photocentre C are CM, and the intersection point of CM and image planes is picture point m, and m is the throwing of spatial point P on the image plane
Shadow.Camera coordinates system is then converted into the process that image coordinate system corresponds to perspective projection, following matrix indicates:
Wherein, s is 0 scale factor, and f is the effective focal length (distance of optical center to the plane of delineation) of camera, (X, Y,
Z, 1)TFor the odd times coordinate of camera coordinates system, (x, y, 1)TFor the odd times coordinate of image coordinate system.
The transformation of transformation relation, camera coordinates system based on above-mentioned world coordinate system to camera coordinates system to image coordinate system
The transformation relation of relationship and image coordinate system to pixel coordinate system, the change of available world coordinate system to pixel coordinate system
Change relationship:
Wherein, M is the transition matrix that world coordinate system arrives pixel coordinate system, include in the matrix camera external parameter with
The inner parameter of camera, can be using the transformation relation as the default imaging model of camera, the parameter pair of the default imaging model
It is to be based on nominal data this is trained to be preset to that the process for obtaining calibrating parameters should be trained in the various parameters of camera, this programme
As model obtains the process of optimal model parameters.In other embodiments, the default imaging model of camera can also reflect camera
Distortion, model parameter further comprises the distortion parameter of camera.It is trained to preset imaging model to this, nominal data
It is made of the location information of the corresponding picture point in the location information and pixel coordinate system of the calibration point in world coordinate system, so that mould
Type optimizes towards the direction of re-projection error reduction.It then continues back in method shown in Fig. 3 A, carries out obtaining for nominal data
It takes.
In operation S302, the image of calibration point is obtained.
Such as can using with chessboard grid pattern object as demarcate object, using the angle point on gridiron pattern as calibration point,
Utilize the image of this camera shooting calibration point to be calibrated.
In operation S303, the corresponding characteristic point of calibration point is extracted from the image of calibration point.
This operation can use various feature extraction algorithms and extract the corresponding characteristic point of calibration point from the image of calibration point.
Specifically, when the image of this camera shooting is RGB image, RGB image is converted into gray level image, is extracted from gray level image
The corresponding characteristic point of calibration point.
In operation S304, the nominal data of calibration point is obtained.
In this operation, location information of the calibration point in world coordinate system is obtained, the corresponding spy of the calibration point is obtained
Positional relationship of the sign point in pixel coordinate system, the two form a nominal data.
In operation S305, judge whether the quantity of nominal data meets required sample size, is to execute operation S306, it is no
Then execute operation S308.
Scheme provided by the disclosure can greatly reduce the required sample size of training, can preset required sample
Quantity collects nominal data according to the preset sample size, and the mistake of nominal data collection is completed in a manner of optimum efficiency
Journey.If required sample size, further progress calibration acquisition, if calibration number has not been reached yet in the total quantity of nominal data
According to total quantity reach required sample size, then can enter training process.
In operation S306, it is based on least-squares estimation, using multiple nominal datas and original model parameter to default imaging
Model is trained, and obtains optimal model parameters.
In this operation, optimization aim is based on depending on least-squares estimation, and original model parameter has obtained, and presets imaging model
Also known, original model parameter is substituting in default imaging model and obtains initial model, no longer needed to generate introductory die at random
Nominal data as sample data substitution initial model is constantly iterated optimization, finally obtains and meet least square by type
The optimal model parameters of estimation.
In operation S307, the calibrating parameters of this camera are extracted from optimal model parameters.
Above it was mentioned that the invention that may include external parameter, inner parameter, distortion parameter of camera etc. in model parameter, because
This can be extracted from optimal model parameters needed for this camera calibrating parameters.
In operation S308, delay scheduled time interval, redo S302.
This operation is carried out when aforesaid operations S305 determines the quantity of nominal data not enough, twice between calibration
Delay scheduled time interval is wanted, such as 2 seconds, and the nominal data of different calibration acquisitions needs to reflect the calibration of different angle
The corresponding nominal data of point.
It can be seen that the characteristics of scheme that the disclosure provides makes full use of same type camera internal parameter similar, similar
When the new camera parameter calibration of type, with the parameter of Optimized model of a certain camera of the same type demarcated for camera to be calibrated
Initial parameter, it is only necessary to a small amount of camera calibration data can to the practical internal reference of camera carry out least-squares estimation, due to
The initial value of camera parameter estimation is the Optimized model parameter of same type a certain camera, therefore relatively camera to be calibrated
Actual parameter, a small amount of nominal data can quickly estimate the parameter of camera.
Fig. 3 D diagrammatically illustrates the camera of camera parameter scaling method and the prior art according to an embodiment of the present disclosure
The optimization process comparison diagram of parameter calibration method.
As shown in Figure 3D, the model parameter that the camera parameter scaling method of the disclosure has optimized in same type camera
On the basis of be iterated optimization, it is only necessary to a small amount of nominal data and relatively short iteration step length can reach minimum re-projection
Error.And the original model parameter that the camera parameter scaling method of the prior art can not used for reference, Optimized Iterative from the beginning,
It needs a large amount of nominal data and longer iteration step length just to can reach minimum re-projection error, related personnel is caused to need to put into
A large amount of time and efforts carries out the acquisition and training of nominal data, is not suitable for the camera parameter calibration of large scale deployment.
Fig. 4 diagrammatically illustrates the block diagram of camera parameter caliberating device according to an embodiment of the present disclosure.
As shown in figure 4, camera parameter caliberating device 400 includes that the first acquisition module 410, second obtains module 420, training
Module 430 and demarcating module 440.
First acquisition module 410 is used to train obtained optimal model parameters to obtain this phase based on other same type cameras
The corresponding original model parameter of machine.
For second acquisition module 420 for obtaining sample set, which includes calibration of multiple calibration points about this camera
Data.
Training module 430 is used to be trained default imaging model based on sample set and original model parameter, obtains this
The corresponding optimal model parameters of camera.
Demarcating module 440 is used to obtain the calibrating parameters of this camera based on the corresponding optimal model parameters of this camera.
In one embodiment of the present disclosure, the first acquisition module 410, which has been trained based on other same type cameras, obtains
Optimal model parameters obtain the corresponding original model parameter of this camera include: the first acquisition module 410 be used for will it is any other together
Types of cameras has trained obtained optimal model parameters as the corresponding original model parameter of this camera;Alternatively, for multiple
Other same type cameras have trained obtained optimal model parameters averagely to be obtained mean value model parameter, by the mean value model
Parameter is as the corresponding original model parameter of this camera.
In one embodiment of the present disclosure, each calibration point in sample set includes: about the nominal data of this camera
Position letter of location information and the calibration point corresponding picture point of the calibration point in world coordinate system in pixel coordinate system
Breath.Default imaging model table levies the transformation model from world coordinate system to pixel coordinate system.
Wherein, as an optional embodiment, it includes: the second acquisition module that the second acquisition module 420, which obtains sample set,
420, for establishing world coordinate system and pixel coordinate system, for any calibration point, obtain the calibration point in world coordinate system
Location information, shoots the image of the calibration point, and the corresponding characteristic point of the calibration point is extracted from described image, obtains this feature point
Location information in pixel coordinate system, by location information of the calibration point in world coordinate system spy corresponding with the calibration point
Location information nominal data as the calibration point about this camera of the sign point in pixel coordinate system.Multiple calibration points are about this
The nominal data of camera constitutes sample set.
In one embodiment of the present disclosure, training module 430 is based on sample set and original model parameter to default imaging
It includes: training module 430 for constituting the introductory die of the default imaging model using original model parameter that model, which is trained,
Type;Based on least square method, optimization is iterated to the initial model using the sample set, obtains optimal model parameters.
In one embodiment of the present disclosure, sample set includes training set and test set.Training module 430 is based on the sample
This collection and the original model parameter are trained default imaging model, obtain the corresponding optimal model parameters packet of this camera
Include: training module 430 is used to be trained default imaging model based on the training set and the original model parameter;It utilizes
The model that the test set obtains each training is verified, when verification result meets preset condition, by acquired model
As optimal models, when verification result is unsatisfactory for preset condition, continue to be trained obtained model based on training set.
Wherein, the mould each training obtained using the test set as an optional embodiment, training module 430
It includes: that training module 430 is used to miss using the re-projection of the test set calculating model that training obtains every time that type, which carries out verifying,
Difference determines that verification result meets preset condition when re-projection error is less than preset threshold, when re-projection error is not less than default
When threshold value, determine that verification result is unsatisfactory for preset condition.
In one embodiment of the present disclosure, demarcating module 440 is based on the corresponding optimal model parameters of this camera and obtains this
The calibrating parameters of camera include: demarcating module 440 for obtaining consolidating for this camera from the corresponding optimal model parameters of this camera
There are parameter and distortion parameter.
It should be noted that in device section Example each module/unit/subelement etc. embodiment, the skill of solution
Art problem, the function of realization and the technical effect reached respectively with the implementation of corresponding step each in method section Example
Mode, the technical issues of solving, the function of realization and the technical effect that reaches are same or like, and details are not described herein.
It is module according to an embodiment of the present disclosure, submodule, unit, any number of or in which any more in subelement
A at least partly function can be realized in a module.It is single according to the module of the embodiment of the present disclosure, submodule, unit, son
Any one or more in member can be split into multiple modules to realize.According to the module of the embodiment of the present disclosure, submodule,
Any one or more in unit, subelement can at least be implemented partly as hardware circuit, such as field programmable gate
Array (FPGA), programmable logic array (PLA), system on chip, the system on substrate, the system in encapsulation, dedicated integrated electricity
Road (ASIC), or can be by the hardware or firmware for any other rational method for integrate or encapsulate to circuit come real
Show, or with any one in three kinds of software, hardware and firmware implementations or with wherein any several appropriately combined next reality
It is existing.Alternatively, can be at least by part according to one or more of the module of the embodiment of the present disclosure, submodule, unit, subelement
Ground is embodied as computer program module, when the computer program module is run, can execute corresponding function.
For example, first obtains in the acquisition of module 410, second module 420, training module 430 and demarcating module 440
Any number of may be incorporated in a module is realized or any one module therein can be split into multiple modules.
Alternatively, at least partly function of one or more modules in these modules can mutually be tied at least partly function of other modules
It closes, and is realized in a module.In accordance with an embodiment of the present disclosure, identification signals sending module 410, identification signals
At least one of receiving module 420, identification module 430 and information signal transceiver module 440 can be at least by parts
Ground is embodied as hardware circuit, such as field programmable gate array (FPGA), programmable logic array (PLA), system on chip, substrate
On system, the system in encapsulation, specific integrated circuit (ASIC), or can be by carrying out integrated to circuit or encapsulating any
The hardware such as other rational methods or firmware realize, or with any one in three kinds of software, hardware and firmware implementations
Or it several appropriately combined is realized with wherein any.Alternatively, first obtains the acquisition of module 410, second module 420, training mould
At least one of block 430 and demarcating module 440 can at least be implemented partly as computer program module, when the meter
When calculation machine program module is run, corresponding function can be executed.
Fig. 5 is diagrammatically illustrated according to the computer equipment for being adapted for carrying out method as described above of the embodiment of the present disclosure
Block diagram.Computer equipment shown in Fig. 5 is only an example, should not function to the embodiment of the present disclosure and use scope bring
Any restrictions.
As shown in figure 5, include processor 501 according to the computer equipment 500 of the embodiment of the present disclosure, it can be according to storage
It is loaded into random access storage device (RAM) 503 in the program in read-only memory (ROM) 502 or from storage section 508
Program and execute various movements appropriate and processing.Processor 501 for example may include general purpose microprocessor (such as CPU), refer to
Enable set processor and/or related chip group and/or special microprocessor (for example, specific integrated circuit (ASIC)), etc..Processing
Device 501 can also include the onboard storage device for caching purposes.Processor 501 may include for executing according to disclosure reality
Apply single treatment unit either multiple processing units of the different movements of the method flow of example.
In RAM 503, it is stored with computer equipment 500 and operates required various programs and data.Processor 501, ROM
502 and RAM 503 is connected with each other by bus 504.Processor 501 is by executing the journey in ROM 502 and/or RAM 503
Sequence executes the various operations of the method flow according to the embodiment of the present disclosure.It is being removed it is noted that described program also can store
In one or more memories other than ROM 502 and RAM 503.Processor 501 can also be stored in described one by executing
Program in a or multiple memories executes the various operations of the method flow according to the embodiment of the present disclosure.
In accordance with an embodiment of the present disclosure, computer equipment system 500 can also include input/output (I/O) interface 505, defeated
Enter/export (I/O) interface 505 and is also connected to bus 504.System 500 can also include be connected to I/O interface 505 with lower part
It is one or more in part: the importation 506 including keyboard, mouse etc.;Including such as cathode-ray tube (CRT), liquid crystal
Show the output par, c 507 of device (LCD) etc. and loudspeaker etc.;Storage section 508 including hard disk etc.;And including such as LAN
The communications portion 509 of the network interface card of card, modem etc..Communications portion 509 is executed via the network of such as internet
Communication process.Driver 510 is also connected to I/O interface 505 as needed.Detachable media 511, such as disk, CD, magneto-optic
Disk, semiconductor memory etc. are mounted on as needed on driver 510, in order to from the computer program root read thereon
According to needing to be mounted into storage section 508.
In accordance with an embodiment of the present disclosure, computer software journey may be implemented as according to the method flow of the embodiment of the present disclosure
Sequence.For example, embodiment of the disclosure includes a kind of computer program product comprising be carried on computer readable storage medium
Computer program, which includes the program code for method shown in execution flow chart.In such implementation
In example, which can be downloaded and installed from network by communications portion 509, and/or from detachable media 611
It is mounted.When the computer program is executed by processor 501, the above-mentioned function limited in the system of the embodiment of the present disclosure is executed
Energy.In accordance with an embodiment of the present disclosure, system as described above, unit, module, unit etc. can pass through computer program
Module is realized.
The disclosure additionally provides a kind of computer readable storage medium, which can be above-mentioned reality
It applies included in equipment/device/system described in example;Be also possible to individualism, and without be incorporated the equipment/device/
In system.Above-mentioned computer readable storage medium carries one or more program, when said one or multiple program quilts
When execution, the method according to the embodiment of the present disclosure is realized.
In accordance with an embodiment of the present disclosure, computer readable storage medium can be non-volatile computer-readable storage medium
Matter, such as can include but is not limited to: portable computer diskette, hard disk, random access storage device (RAM), read-only memory
(ROM), erasable programmable read only memory (EPROM or flash memory), portable compact disc read-only memory (CD-ROM), light
Memory device, magnetic memory device or above-mentioned any appropriate combination.In the disclosure, computer readable storage medium can
With to be any include or the tangible medium of storage program, the program can be commanded execution system, device or device use or
Person is in connection.For example, in accordance with an embodiment of the present disclosure, computer readable storage medium may include above-described
One or more memories other than ROM 502 and/or RAM 503 and/or ROM 502 and RAM 503.
Flow chart and block diagram in attached drawing are illustrated according to the system of the various embodiments of the disclosure, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, program segment or code of table, a part of above-mentioned module, program segment or code include one or more
Executable instruction for implementing the specified logical function.It should also be noted that in some implementations as replacements, institute in box
The function of mark can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are practical
On can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it wants
It is noted that the combination of each box in block diagram or flow chart and the box in block diagram or flow chart, can use and execute rule
The dedicated hardware based systems of fixed functions or operations is realized, or can use the group of specialized hardware and computer instruction
It closes to realize.
It will be understood by those skilled in the art that the feature recorded in each embodiment and/or claim of the disclosure can
To carry out multiple combinations and/or combination, even if such combination or combination are not expressly recited in the disclosure.Particularly, exist
In the case where not departing from disclosure spirit or teaching, the feature recorded in each embodiment and/or claim of the disclosure can
To carry out multiple combinations and/or combination.All these combinations and/or combination each fall within the scope of the present disclosure.
Embodiment of the disclosure is described above.But the purpose that these embodiments are merely to illustrate that, and
It is not intended to limit the scope of the present disclosure.Although respectively describing each embodiment above, but it is not intended that each reality
Use cannot be advantageously combined by applying the measure in example.The scope of the present disclosure is defined by the appended claims and the equivalents thereof.It does not take off
From the scope of the present disclosure, those skilled in the art can make a variety of alternatives and modifications, these alternatives and modifications should all fall in this
Within scope of disclosure.
Claims (18)
1. a kind of camera parameter scaling method, comprising:
Obtained optimal model parameters have been trained to obtain the corresponding original model parameter of this camera based on other same type cameras;
Sample set is obtained, the sample set includes nominal data of multiple calibration points about this camera;
Default imaging model is trained based on the sample set and the original model parameter, it is corresponding most to obtain this camera
Excellent model parameter;
The calibrating parameters of this camera are obtained based on the corresponding optimal model parameters of this camera.
2. described to have trained obtained optimal models based on other same type cameras according to the method described in claim 1, wherein
Parameter obtains the corresponding original model parameter of this camera
Train obtained optimal model parameters as the corresponding original model parameter of this camera in other any same type cameras;
Or
Obtained optimal model parameters have been trained averagely to be obtained mean value model parameter in other multiple same type cameras, by institute
Mean value model parameter is stated as the corresponding original model parameter of this camera.
3. according to the method described in claim 1, wherein:
Each calibration point in the sample set includes: the calibration point in world coordinate system about the nominal data of this camera
The location information of location information and the corresponding picture point of the calibration point in pixel coordinate system;
The default imaging model table levies the transformation model from world coordinate system to pixel coordinate system.
4. according to the method described in claim 3, wherein, the acquisition sample set includes:
Establish world coordinate system and pixel coordinate system;
For any calibration point, location information of the calibration point in world coordinate system is obtained;
The image for shooting the calibration point extracts the corresponding characteristic point of the calibration point from described image, obtains this feature point in picture
Location information in plain coordinate system;
By location information of the calibration point in world coordinate system characteristic point corresponding with the calibration point in pixel coordinate system
Nominal data of the location information as the calibration point about this camera;
Multiple calibration points constitute the sample set about the nominal data of this camera.
5. according to the method described in claim 1, wherein, the sample set and the original model parameter of being based on is to default
Imaging model, which is trained, includes:
The initial model of the default imaging model is constituted using original model parameter;
Based on least square method, optimization is iterated to the initial model using the sample set, obtains optimal model parameters.
6. according to the method described in claim 1, wherein, the sample set and the original model parameter of being based on is to default
Imaging model is trained, and is obtained the corresponding optimal model parameters of this camera and is included:
The sample set includes training set and test set;
Default imaging model is trained based on the training set and the original model parameter;
It is verified using the model that the test set obtains each training, when verification result meets preset condition, by institute
Model is obtained as optimal models, using the parameter of the model as optimal model parameters, when verification result is unsatisfactory for preset condition,
Continue to be trained obtained model based on the training set.
7. according to the method described in claim 6, wherein, being verified using the model that the test set obtains each training
Include:
The re-projection error that the model that training obtains every time is calculated using the test set, when re-projection error is less than preset threshold
When, it determines that verification result meets preset condition, when re-projection error is not less than preset threshold, it is pre- to determine that verification result is unsatisfactory for
If condition.
8. according to the method described in claim 1, wherein, the corresponding optimal model parameters of this camera that are based on obtain this camera
Calibrating parameters include:
The intrinsic parameter of calibration and calibration distortion parameter of this camera are obtained from the corresponding optimal model parameters of this camera.
9. a kind of camera parameter caliberating device, comprising:
First obtains module, corresponding for having trained obtained optimal model parameters to obtain this camera based on other same type cameras
Original model parameter;
Second obtains module, and for obtaining sample set, the sample set includes nominal data of multiple calibration points about this camera;
Training module is obtained for being trained based on the sample set and the original model parameter to default imaging model
The corresponding optimal model parameters of this camera;
Demarcating module, for obtaining the calibrating parameters of this camera based on the corresponding optimal model parameters of this camera.
10. device according to claim 9, wherein the first acquisition module is based on other same type cameras and has trained
Obtained optimal model parameters obtain the corresponding original model parameter of this camera
Described first obtains module, for having trained obtained optimal model parameters as this phase in other any same type cameras
The corresponding original model parameter of machine;Alternatively, for other multiple same type cameras have been trained obtained optimal model parameters into
Row averagely obtains mean value model parameter, using the mean value model parameter as the corresponding original model parameter of this camera.
11. device according to claim 9, in which:
Each calibration point in the sample set includes: the calibration point in world coordinate system about the nominal data of this camera
The location information of location information and the corresponding picture point of the calibration point in pixel coordinate system;
The default imaging model table levies the transformation model from world coordinate system to pixel coordinate system.
12. device according to claim 11, wherein the second acquisition module obtains sample set and includes:
Described second obtains module, for establishing world coordinate system and pixel coordinate system;For any calibration point, the calibration is obtained
Location information of the point in world coordinate system;It is corresponding to extract the calibration point from described image for the image for shooting the calibration point
Characteristic point obtains location information of this feature point in pixel coordinate system;By position letter of the calibration point in world coordinate system
Cease location information calibration as the calibration point about this camera of the characteristic point corresponding with the calibration point in pixel coordinate system
Data;Multiple calibration points constitute the sample set about the nominal data of this camera.
13. device according to claim 9, wherein the training module is based on the sample set and the initial model
Parameter is trained default imaging model
The training module, for constituting the initial model of the default imaging model using original model parameter;Based on minimum
Square law is iterated optimization to the initial model using the sample set, obtains optimal model parameters.
14. device according to claim 9, in which:
The sample set includes training set and test set;
The training module is based on the sample set and the original model parameter is trained default imaging model, obtains this
The corresponding optimal model parameters of camera include: the training module, for being based on the training set and the original model parameter
Default imaging model is trained;It is verified using the model that the test set obtains each training, works as verification result
When meeting preset condition, using acquired model as optimal models, when verification result is unsatisfactory for preset condition, continue based on described
Training set is trained obtained model.
15. device according to claim 14, wherein the training module obtains each training using the test set
Model carry out verifying include:
The training module, for calculating the re-projection error of model that training obtains every time using the test set, when throwing again
When shadow error is less than preset threshold, determine that verification result meets preset condition, when re-projection error is not less than preset threshold, really
Determine verification result and is unsatisfactory for preset condition.
16. device according to claim 9, wherein the demarcating module is based on the corresponding optimal model parameters of this camera
The calibrating parameters for obtaining this camera include:
The demarcating module, for obtaining the intrinsic parameter of calibration and mark of this camera from the corresponding optimal model parameters of this camera
Determine distortion parameter.
17. a kind of computer equipment including memory, processor and stores the meter that can be run on a memory and on a processor
Calculation machine program, the processor realize that camera parameter according to any one of claims 1 to 8 such as is demarcated when executing described program
Method.
18. a kind of computer readable storage medium, is stored thereon with executable instruction, which makes to handle when being executed by processor
Device executes such as camera parameter scaling method according to any one of claims 1 to 8.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811477402.5A CN110378963A (en) | 2018-12-04 | 2018-12-04 | Camera parameter scaling method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811477402.5A CN110378963A (en) | 2018-12-04 | 2018-12-04 | Camera parameter scaling method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110378963A true CN110378963A (en) | 2019-10-25 |
Family
ID=68243151
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811477402.5A Pending CN110378963A (en) | 2018-12-04 | 2018-12-04 | Camera parameter scaling method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110378963A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111369632A (en) * | 2020-03-06 | 2020-07-03 | 北京百度网讯科技有限公司 | Method and device for acquiring internal parameters in camera calibration |
CN112894154A (en) * | 2021-05-07 | 2021-06-04 | 视睿(杭州)信息科技有限公司 | Laser marking method and device |
CN114170323A (en) * | 2021-12-07 | 2022-03-11 | 合肥英睿系统技术有限公司 | Parameter calibration method and equipment for image shooting device and storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050099501A1 (en) * | 2000-06-12 | 2005-05-12 | Microsoft Corporation | Self-calibration for a catadioptric camera |
CN102075785A (en) * | 2010-12-28 | 2011-05-25 | 武汉大学 | Method for correcting wide-angle camera lens distortion of automatic teller machine (ATM) |
CN102208108A (en) * | 2011-04-01 | 2011-10-05 | 哈尔滨工业大学 | Large-field high-precision quick field overall calibration method for camera |
CN103035005A (en) * | 2012-12-13 | 2013-04-10 | 广州致远电子股份有限公司 | Panorama parking calibration method and device, and automatic calibration method |
CN105678740A (en) * | 2015-12-30 | 2016-06-15 | 完美幻境(北京)科技有限公司 | Camera geometrical calibration processing method and apparatus |
CN108734738A (en) * | 2017-04-18 | 2018-11-02 | 杭州海康威视数字技术股份有限公司 | Camera calibration method and device |
-
2018
- 2018-12-04 CN CN201811477402.5A patent/CN110378963A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050099501A1 (en) * | 2000-06-12 | 2005-05-12 | Microsoft Corporation | Self-calibration for a catadioptric camera |
CN102075785A (en) * | 2010-12-28 | 2011-05-25 | 武汉大学 | Method for correcting wide-angle camera lens distortion of automatic teller machine (ATM) |
CN102208108A (en) * | 2011-04-01 | 2011-10-05 | 哈尔滨工业大学 | Large-field high-precision quick field overall calibration method for camera |
CN103035005A (en) * | 2012-12-13 | 2013-04-10 | 广州致远电子股份有限公司 | Panorama parking calibration method and device, and automatic calibration method |
CN105678740A (en) * | 2015-12-30 | 2016-06-15 | 完美幻境(北京)科技有限公司 | Camera geometrical calibration processing method and apparatus |
CN108734738A (en) * | 2017-04-18 | 2018-11-02 | 杭州海康威视数字技术股份有限公司 | Camera calibration method and device |
Non-Patent Citations (2)
Title |
---|
熊有伦等: "《机器人学建模、控制和视觉》", 华中科技大学出版社 * |
赵越 等: "基于最小二乘法的摄像机标定参数非线性优化", 《仪器仪表学报》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111369632A (en) * | 2020-03-06 | 2020-07-03 | 北京百度网讯科技有限公司 | Method and device for acquiring internal parameters in camera calibration |
CN112894154A (en) * | 2021-05-07 | 2021-06-04 | 视睿(杭州)信息科技有限公司 | Laser marking method and device |
CN112894154B (en) * | 2021-05-07 | 2021-09-03 | 视睿(杭州)信息科技有限公司 | Laser marking method and device |
CN114170323A (en) * | 2021-12-07 | 2022-03-11 | 合肥英睿系统技术有限公司 | Parameter calibration method and equipment for image shooting device and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110427917B (en) | Method and device for detecting key points | |
AU2011312140B2 (en) | Rapid 3D modeling | |
JP2024041815A (en) | Scene reconstruction system and method for solving entanglement of light and material field | |
EP2111530B1 (en) | Automatic stereo measurement of a point of interest in a scene | |
US8098958B2 (en) | Processing architecture for automatic image registration | |
CN108198145A (en) | For the method and apparatus of point cloud data reparation | |
CN107146254A (en) | The Camera extrinsic number scaling method of multicamera system | |
CN111383285B (en) | Sensor fusion calibration method and system based on millimeter wave radar and camera | |
CN109118545A (en) | 3-D imaging system scaling method and system based on rotary shaft and binocular camera | |
WO2021140886A1 (en) | Three-dimensional model generation method, information processing device, and program | |
CN106780618A (en) | 3 D information obtaining method and its device based on isomery depth camera | |
CN109523597A (en) | The scaling method and device of Camera extrinsic | |
CN110378963A (en) | Camera parameter scaling method and device | |
CN105654547B (en) | Three-dimensional rebuilding method | |
CN107679537A (en) | A kind of texture-free spatial target posture algorithm for estimating based on profile point ORB characteristic matchings | |
CN109559349A (en) | A kind of method and apparatus for calibration | |
CN103017676B (en) | Three-dimensional scanner and 3-D scanning method | |
CN105989603A (en) | Machine vision image sensor calibration | |
CN106643492B (en) | A kind of aero-engine damaged blade 3-dimensional digital speckle formative method | |
US20190082173A1 (en) | Apparatus and method for generating a camera model for an imaging system | |
CN112270719B (en) | Camera calibration method, device and system | |
CN106705849A (en) | Calibration method of linear-structure optical sensor | |
CN108761271A (en) | A kind of power grid screen of trees detection method and system | |
CN110376570A (en) | Method, system and the equipment that scanner coordinate system and IMU coordinate system are demarcated | |
Fritsch et al. | Multi-sensors and multiray reconstruction for digital preservation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |