CN114897990A - Camera distortion calibration method and system based on neural network and storage medium - Google Patents
Camera distortion calibration method and system based on neural network and storage medium Download PDFInfo
- Publication number
- CN114897990A CN114897990A CN202210342340.7A CN202210342340A CN114897990A CN 114897990 A CN114897990 A CN 114897990A CN 202210342340 A CN202210342340 A CN 202210342340A CN 114897990 A CN114897990 A CN 114897990A
- Authority
- CN
- China
- Prior art keywords
- camera
- neural network
- coordinates
- calibration
- calibration plate
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
- G06T7/85—Stereo camera calibration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
-
- 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
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30244—Camera pose
Abstract
The invention discloses a camera distortion calibration method, a system and a storage medium based on a neural network, and the method comprises the steps of firstly obtaining calibration plate maps of a plurality of cameras to be calibrated; then extracting characteristic points in the calibration plate diagram; calculating world coordinates of the feature points, and calibrating internal and external parameters of the camera by using a Zhangyingyou calibration method; obtaining ideal camera coordinates of the feature points according to the world coordinates of the feature points and the transformation of the external parameters; then extracting pixel coordinates of the feature points, and obtaining distorted camera coordinates according to inverse transformation of the internal parameters; the ideal camera coordinates correspond to the distorted camera coordinates one by one, a neural network is constructed, in addition, the thought of an extreme learning machine is added in the neural network parameter initialization process, and the initial parameters of the hidden layer and the output layer are obtained through least squares.
Description
Technical Field
The invention belongs to the technical field of image processing, and particularly relates to a camera distortion calibration method, system and storage medium based on a neural network.
Background
One of the basic tasks of computer vision is to calculate the geometric information of an object in a three-dimensional space based on the image information acquired by a camera, and to reconstruct and identify the object accordingly, and the correlation between the three-dimensional geometric position of a point on the surface of the object in the space and the corresponding point in the image is determined by the geometric model imaged by the camera, and the parameters of the geometric model are the parameters of the camera. Under most conditions, these parameters must be obtained through experimentation and calculation. In image measurement or machine vision application, calibration of camera parameters is a very critical link, and the accuracy of a calibration result and the stability of an algorithm directly influence the accuracy of a result generated by the operation of a camera. Therefore, the camera calibration is a precondition for subsequent work, and the improvement of the calibration precision is a key point of scientific research.
The existing camera calibration method can be divided into a traditional camera calibration method, an active vision camera calibration method, a camera self-calibration method and the like in a large direction. The traditional camera calibration method is suitable for any camera model, and the calibration precision is relatively high; the disadvantage is that the calibration reference is needed, and the size of the calibration reference is known and is difficult to realize in some applications. The active vision camera calibration method is simple in algorithm, can often obtain a linear solution, and is high in robustness; however, the system has high implementation cost, expensive experimental equipment and high requirement on experimental conditions. The self-calibration method has strong flexibility and can perform online calibration on the camera; but has the disadvantage of poor robustness.
Two different types of distortion mainly exist in the imaging process of a video camera, one is the distortion caused by the shape of a lens, namely mirror image distortion, and the other is the distortion caused by the condition that the lens and an imaging surface cannot be strictly parallel in the camera assembling process, namely tangential distortion. It is common practice to construct mathematical distortion models for these two distortions, respectively, wherein one of the commonly used radial distortion models is:
one of the commonly used tangential distortion models is:
in the traditional method, mathematical models of formula (1) and formula (2) are only approximate simulations of the main distortion condition of the camera, and in the actual condition, the distortion generated during camera imaging is caused by various factors, and the two formulas cannot completely describe the actual distortion condition. In some cases, the distortion condition does not exhibit the distribution of the above two equations. Therefore, a large deviation occurs when the distortion correction of the camera is performed using the distortion parameter obtained by the solution.
In view of the above, it is necessary to develop a method for calibrating camera distortion based on a neural network to solve the above technical problems.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a camera distortion calibration method, a system and a storage medium based on a neural network. The method solves the problems that the existing camera calibration method can generate larger deviation when the distortion correction of the camera is carried out.
The invention aims to provide a camera distortion calibration method based on a neural network.
A camera distortion calibration method based on a neural network comprises the following steps:
s1, obtaining calibration plate diagrams of a plurality of cameras to be calibrated;
s2, extracting characteristic points in the calibration plate diagram;
s3, calculating world coordinates of the feature points by taking the plane of the calibration board diagram as a zero plane, and calibrating internal and external parameters of the camera by using a Zhang Zhengyou calibration method;
s4, obtaining ideal camera coordinates of the feature points according to the world coordinates of the feature points and the transformation of the external parameters; then extracting pixel coordinates of the feature points, and obtaining distorted camera coordinates according to inverse transformation of the internal parameters; corresponding the ideal camera coordinates to the distorted camera coordinates one by one, and constructing a neural network; and obtaining a distortion correction model through neural network training, and carrying out distortion correction on the camera through the distortion correction model.
Further, in step S1, the calibration plate maps are obtained from multiple angle shots, and the number of the calibration plate maps is greater than 3.
Further, in step S2, the specific method for extracting the feature points in the calibration plate map is as follows:
s21, searching an area in the calibration plate diagram through template matching;
s22, extracting all circular areas in the calibration plate map according to the area growing method;
and S23, finding the circular boundary points of all circular areas, and performing circular fitting to obtain the circle center, namely the characteristic point of the calibration plate chart.
Further, in step S3, the specific method for calibrating the inside and outside parameters of the camera by using the zhangnyou calibration method is as follows: solving a homography matrix from the world coordinates of the feature points to the image coordinates; calculating camera internal parameters according to the pinhole imaging model; and solving the corresponding camera external parameters according to the camera internal parameters.
Further, in step S4, the neural network is constructed as follows:
1) dividing a neural network into three layers, namely an input layer, a hidden layer and an output layer;
2) let W be { W ═ as the parameter of the input layer and the hidden layer i，j }; i is 1, …, m; j is 1, …, l, and the parameters of the hidden layer and the output layer are α ═ { α ═ α i，j }; 1, …, l; j is 1, …, n, where m, l, n are number of input nodes, number of hidden nodes, number of output nodes respectively; randomly initializing W, wherein the hidden layer is obtained as H ═ f (P · W + b), wherein b is a bias, and f (·) is an activation function;
3) the output A is H alpha, H is a hidden layer, A is the output of a training sample, and alpha is obtained through a least square method based on the calculation mode of an extreme learning machine, so that the initial value of the neural network is obtained;
4) and after initialization, training and optimizing the neural network in an error feedback mode.
Further, in step S4, the method for establishing the distortion correction model is as follows: inputting the distorted camera coordinates as a neural network, taking the ideal camera coordinates as a learning standard, and training the neural network in an error feedback manner to obtain a distortion correction neural network model; the aberration correction is specifically as follows: and (3) obtaining distorted camera coordinates through internal reference inverse transformation of each pixel coordinate in the calibration plate map, using the distorted camera coordinates as input of a neural network, and giving coordinate pixel values at output points to coordinates at input points after the correction of the neural network, namely completing the distortion correction.
The invention also provides a camera distortion calibration system based on the neural network, which comprises the following components:
a calibration board diagram acquisition module of the camera to be calibrated: acquiring calibration plate images of a plurality of cameras to be calibrated;
a characteristic point extraction module in the calibration plate diagram: extracting characteristic points in the calibration plate diagram;
a camera internal and external parameter calibration module: calculating world coordinates of the characteristic points by taking the plane of the calibration plate diagram as a zero plane, and calibrating internal and external parameters of the camera by using a Zhang-Zhengyou calibration method;
a camera distortion correction module: obtaining ideal camera coordinates of the feature points according to the world coordinates of the feature points and the transformation of the external parameters; then extracting pixel coordinates of the feature points, and obtaining distorted camera coordinates according to inverse transformation of the internal parameters; corresponding the ideal camera coordinates to the distorted camera coordinates one by one, and constructing a neural network; and obtaining a distortion correction model through neural network training, and carrying out distortion correction on the camera through the distortion correction model.
The invention finally provides a storage medium for storing a computer program to be loaded by a processor for performing the detection method of any of the above.
Compared with the prior art, the invention has the following advantages:
firstly, obtaining calibration plate diagrams of a plurality of cameras to be calibrated; then extracting characteristic points in the calibration plate graph; calculating world coordinates of the feature points, and calibrating internal and external parameters of the camera by using a Zhangyingyou calibration method; obtaining ideal camera coordinates of the feature points according to the world coordinates of the feature points and the transformation of the external parameters; then extracting pixel coordinates of the feature points, and obtaining distorted camera coordinates according to inverse transformation of the internal parameters; the ideal camera coordinates correspond to the distorted camera coordinates one by one, and a neural network is constructed, so that the neural network can better simulate the distribution relation of input and output through self-adaptive learning, thereby improving the precision and reducing the reprojection error; in addition, the thought of an extreme learning machine is added in parameter initialization, and initial parameters of the hidden layer and the output layer are obtained through least squares.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a flow chart of a method for calibrating camera distortion based on a neural network according to the present invention;
FIG. 2 is a block diagram of a neural network-based camera distortion calibration system according to the present invention;
FIG. 3 is a diagram of a neural network architecture according to the present invention;
fig. 4 is a diagram of the reprojection error for three different camera calibrations according to the present invention, where the abscissa represents the pixel error and the ordinate represents the number of pixels.
Detailed Description
The technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. It is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments, and all other embodiments obtained by those skilled in the art without any inventive work are within the scope of the present invention.
Referring to fig. 1, a method for calibrating camera distortion based on a neural network includes the following steps:
s1, obtaining calibration plate diagrams of a plurality of cameras to be calibrated;
s2, extracting characteristic points in the calibration plate diagram;
s3, calculating world coordinates of the feature points by taking the plane of the calibration board diagram as a zero plane, and calibrating internal and external parameters of the camera by using a Zhang Zhengyou calibration method;
s4, obtaining ideal camera coordinates of the feature points according to the world coordinates of the feature points and the transformation of the external parameters; then extracting pixel coordinates of the feature points, and obtaining distorted camera coordinates according to inverse transformation of the internal parameters; corresponding the ideal camera coordinates to the distorted camera coordinates one by one, and constructing a neural network; and obtaining a distortion correction model through neural network training, and carrying out distortion correction on the camera through the distortion correction model.
Specifically, in step S1, the calibration plate maps are obtained by shooting from multiple angles, and the number of the calibration plate maps is greater than 3.
In step S2, the specific method for extracting the feature points in the calibration plate map is as follows:
s21, searching an area in the calibration plate diagram through template matching;
s22, extracting all circular areas in the calibration plate map according to the area growing method;
and S23, finding the circular boundary points of all circular areas, and performing circular fitting to obtain the circle center, namely the characteristic point of the calibration plate chart.
In step S3, the specific method for calibrating the internal and external parameters of the camera by using the Zhangyingyou calibration method is as follows: solving a homography matrix from the world coordinates of the feature points to the image coordinates; calculating camera internal parameters according to the pinhole imaging model; and solving the corresponding camera external parameters according to the camera internal parameters.
In step S4, the neural network is constructed as follows:
1) dividing the neural network into three layers, as shown in fig. 3, an input layer, a hidden layer and an output layer;
2) let W be { W ═ as the parameter of the input layer and the hidden layer i，j }; i is 1, …, m; j is 1, …, l, and the parameters of the hidden layer and the output layer are α ═ { α ═ α i，j }; 1, …, l; j is 1, …, n, where m, l, n are the number of input nodes, implicit node number, output node number respectively; randomly initializing W, wherein the hidden layer is obtained as H ═ f (P · W + b), wherein b is a bias, and f (·) is an activation function;
3) the output A is H alpha, H is a hidden layer, A is the output of a training sample, and alpha is obtained through a least square method based on the calculation mode of an extreme learning machine, so that the initial value of the neural network is obtained;
4) and after initialization, training and optimizing the neural network in an error feedback mode.
In step S4, the distortion correction model is established as follows: taking the distorted camera coordinates as a neural network for input, taking the ideal camera coordinates as a learning standard, and training the neural network in an error feedback mode to obtain a distortion correction neural network model; the aberration correction is specifically as follows: and (3) obtaining distorted camera coordinates through internal reference inverse transformation of each pixel coordinate in the calibration plate map, using the distorted camera coordinates as input of a neural network, and giving coordinate pixel values at output points to coordinates at input points after the correction of the neural network, namely completing the distortion correction.
50 images of the calibration plate are shot by a camera from different angles, calibration effect comparison is carried out by respectively applying a direct linear transformation method (DLT), a Zhang calibration method and a text method, the model of the camera is Adimec, the pixel resolution of the camera is 4096X 3072, and the visual field size of a single-channel black-and-white camera is about 45mm X60 mm; the results are shown in FIG. 4.
As can be seen from the figure, when the pixel deviation of the re-projection is less than 0.09 pixel, the number of pixels in the method is the largest, and the next is the zhang-shi calibration method, and the DLT method is the smallest; when the reprojection error is larger than 0.15 pixel, the number of pixels in the method is the least, and then the Zhang calibration method is adopted, and the number of DLT methods is the most. It can be concluded that after the distortion model of the text is corrected, the points with the smallest errors are the most, and the points with the largest errors are the least, so that the result obtained by the text model is the optimal.
Referring to fig. 2, a calibration system for camera distortion based on neural network includes: a calibration board diagram acquisition module 101 of a camera to be calibrated, a characteristic point extraction module 102 in the calibration board diagram, a calibration module 103 of internal and external parameters of the camera and a camera distortion correction module 104;
the calibration plate diagram obtaining module 101 of the camera to be calibrated is used for obtaining a plurality of calibration plate diagrams of the camera to be calibrated;
the feature point extraction module 102 in the calibration plate map is configured to extract feature points in the calibration plate map;
the camera internal and external parameter calibration module 103 is used for calculating world coordinates of feature points by taking the plane where the calibration plate diagram is located as a zero plane, and calibrating the camera internal and external parameters by using a Zhangyingyou calibration method;
the camera distortion correction module 104 is configured to obtain ideal camera coordinates of the feature points according to the world coordinates of the feature points and the transformation of the external parameters; then extracting pixel coordinates of the feature points, and obtaining distorted camera coordinates according to inverse transformation of the internal parameters; corresponding the ideal camera coordinates to the distorted camera coordinates one by one, and constructing a neural network; and obtaining a distortion correction model through neural network training, and carrying out distortion correction on the camera through the distortion correction model.
The invention also provides a storage medium for storing a computer program which is loaded by a processor to perform the detection method of any one of the above. For example, the computer program, when loaded by the processor, may perform the steps of:
extracting characteristic points in the calibration plate graph; calculating world coordinates of the characteristic points by taking the plane of the calibration plate diagram as a zero plane, and calibrating internal and external parameters of the camera by using a Zhang-Zhengyou calibration method; calculating world coordinates of the characteristic points by taking the plane of the calibration plate diagram as a zero plane, and calibrating internal and external parameters of the camera by using a Zhang-Zhengyou calibration method; obtaining ideal camera coordinates of the feature points according to the world coordinates of the feature points and the transformation of the external parameters; then extracting pixel coordinates of the feature points, and obtaining distorted camera coordinates according to inverse transformation of the internal parameters; corresponding the ideal camera coordinates to the distorted camera coordinates one by one, and constructing a neural network; and obtaining a distortion correction model through neural network training, and carrying out distortion correction on the camera through the distortion correction model.
The storage medium may be an internal storage unit of the image processing apparatus of the foregoing embodiment, such as a hard disk or a memory of the image processing apparatus. The storage medium may also be an external storage device of the image processing apparatus, such as a plug-in hard disk provided on the image processing apparatus, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash card (FlashCard), or the like.
Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art will understand that: those skilled in the art can still make modifications or changes to the embodiments described in the foregoing embodiments, or make equivalent substitutions for some features, within the scope of the disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the embodiments of the present invention, and they should be construed as being included therein.
Claims (8)
1. A camera distortion calibration method based on a neural network is characterized by comprising the following steps:
s1, obtaining calibration plate diagrams of a plurality of cameras to be calibrated;
s2, extracting characteristic points in the calibration plate diagram;
s3, calculating world coordinates of the feature points by taking the plane of the calibration board diagram as a zero plane, and calibrating internal and external parameters of the camera by using a Zhang Zhengyou calibration method;
s4, obtaining ideal camera coordinates of the feature points according to the world coordinates of the feature points and the transformation of the external parameters; then extracting pixel coordinates of the feature points, and obtaining distorted camera coordinates according to inverse transformation of the internal parameters; corresponding the ideal camera coordinates to the distorted camera coordinates one by one, and constructing a neural network; and obtaining a distortion correction model through neural network training, and carrying out distortion correction on the camera through the distortion correction model.
2. The neural network-based camera distortion calibration method according to claim 1, wherein in step S1, the calibration plate maps are obtained from multiple angle shots, and the number of the calibration plate maps is greater than 3.
3. The method for calibrating camera distortion based on neural network as claimed in claim 1, wherein in step S2, the specific method for extracting the feature points in the calibration plate map is as follows:
s21, searching an area in the calibration plate diagram through template matching;
s22, extracting all circular areas in the calibration plate map according to the area growing method;
and S23, finding the circular boundary points of all circular areas, and performing circular fitting to obtain the circle center, namely the characteristic point of the calibration plate chart.
4. The method for calibrating camera distortion based on neural network as claimed in claim 1, wherein in step S3, the specific method for calibrating the internal and external parameters of the camera by Zhang Zhengyou calibration method is as follows: solving a homography matrix from the world coordinates of the characteristic points to the image coordinates; calculating camera internal parameters according to the pinhole imaging model; and solving the corresponding external camera parameters according to the internal camera parameters.
5. The method for calibrating camera distortion based on neural network as claimed in claim 1, wherein in step S4, the neural network is constructed as follows:
1) dividing a neural network into three layers, namely an input layer, a hidden layer and an output layer;
2) let W be { W ═ as the parameter of the input layer and the hidden layer i，j }; 1,. m; j 1.. times.l, the parameter of the hidden layer and the output layer is α ═ { α ═ α · i，j }；i＝1，...，l；j＝1，...，nWherein m, l and n are the number of input nodes, the number of implicit nodes and the number of output nodes respectively; randomly initializing W, and obtaining H ═ f (P × W + b) for the hidden layer, where b is the bias and f (·) is the activation function;
3) the output A is H alpha, H is a hidden layer, A is the output of a training sample, and alpha is obtained through a least square method based on the calculation mode of an extreme learning machine, so that the initial value of the neural network is obtained;
4) and after initialization, training and optimizing the neural network in an error feedback mode.
6. The method for calibrating distortion of a camera based on neural network as claimed in claim 1, wherein in step S4, the method for establishing the distortion correction model is as follows: taking the distorted camera coordinates as a neural network for input, taking the ideal camera coordinates as a learning standard, and training the neural network in an error feedback mode to obtain a distortion correction neural network model; the aberration correction is specifically as follows: and (3) obtaining distorted camera coordinates through internal reference inverse transformation of each pixel coordinate in the calibration plate map, using the distorted camera coordinates as input of a neural network, and giving coordinate pixel values at output points to coordinates at input points after the correction of the neural network, namely completing the distortion correction.
7. A neural network-based camera distortion calibration system, the calibration system comprising:
a calibration board diagram acquisition module of the camera to be calibrated: acquiring calibration plate images of a plurality of cameras to be calibrated;
a characteristic point extraction module in the calibration plate diagram: extracting characteristic points in the calibration plate diagram;
a camera internal and external parameter calibration module: calculating world coordinates of the characteristic points by taking the plane of the calibration plate diagram as a zero plane, and calibrating internal and external parameters of the camera by using a Zhang-Zhengyou calibration method;
a camera distortion correction module: obtaining ideal camera coordinates of the feature points according to the world coordinates of the feature points and the transformation of the external parameters; then extracting pixel coordinates of the feature points, and obtaining distorted camera coordinates according to inverse transformation of the internal parameters; corresponding the ideal camera coordinates to the distorted camera coordinates one by one, and constructing a neural network; and obtaining a distortion correction model through neural network training, and carrying out distortion correction on the camera through the distortion correction model.
8. A storage medium for storing a computer program which is loaded by a processor to perform the detection method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210342340.7A CN114897990A (en) | 2022-03-31 | 2022-03-31 | Camera distortion calibration method and system based on neural network and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210342340.7A CN114897990A (en) | 2022-03-31 | 2022-03-31 | Camera distortion calibration method and system based on neural network and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114897990A true CN114897990A (en) | 2022-08-12 |
Family
ID=82716467
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210342340.7A Pending CN114897990A (en) | 2022-03-31 | 2022-03-31 | Camera distortion calibration method and system based on neural network and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114897990A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116297496A (en) * | 2023-05-11 | 2023-06-23 | 征图新视(江苏)科技股份有限公司 | Multispectral intelligent defect detection method |
-
2022
- 2022-03-31 CN CN202210342340.7A patent/CN114897990A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116297496A (en) * | 2023-05-11 | 2023-06-23 | 征图新视(江苏)科技股份有限公司 | Multispectral intelligent defect detection method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Tang et al. | A precision analysis of camera distortion models | |
| US6437823B1 (en) | Method and system for calibrating digital cameras | |
| CN107633536A (en) | A kind of camera calibration method and system based on two-dimensional planar template | |
| CN110969668A (en) | Stereoscopic calibration algorithm of long-focus binocular camera | |
| CN111127422A (en) | Image annotation method, device, system and host | |
| Zhou et al. | Line-based camera calibration with lens distortion correction from a single image | |
| CN111833237A (en) | Image registration method based on convolutional neural network and local homography transformation | |
| CN111709980A (en) | Multi-scale image registration method and device based on deep learning | |
| CN113920205B (en) | Calibration method of non-coaxial camera | |
| CN114549652A (en) | Camera calibration method, device, equipment and computer readable medium | |
| CN111462246B (en) | Equipment calibration method of structured light measurement system | |
| Perdigoto et al. | Calibration of mirror position and extrinsic parameters in axial non-central catadioptric systems | |
| CN114299156A (en) | Method for calibrating and unifying coordinates of multiple cameras in non-overlapping area | |
| CN115147499A (en) | Calibration parameter determination method, hybrid calibration plate, device, equipment and medium | |
| CN114332183A (en) | Image registration method and device, computer equipment and storage medium | |
| CN107067441B (en) | Camera calibration method and device | |
| CN115830135A (en) | Image processing method and device and electronic equipment | |
| CN115187612A (en) | Plane area measuring method, device and system based on machine vision | |
| CN114897990A (en) | Camera distortion calibration method and system based on neural network and storage medium | |
| CN114998448A (en) | Method for calibrating multi-constraint binocular fisheye camera and positioning space point | |
| CN110728296B (en) | Two-step random sampling consistency method and system for accelerating feature point matching | |
| CN115965697A (en) | Projector calibration method, calibration system and device based on Samm's law | |
| Deglint et al. | Auto‐calibration of a projector–camera stereo system for projection mapping | |
| Tola | Multiview 3D Reconstruction of a scene containing independently moving objects | |
| CN114663533A (en) | Multi-view camera calibration method and system based on nonlinear optimization |
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 |