CN110648368B - Calibration board corner point discrimination method based on edge features - Google Patents
Calibration board corner point discrimination method based on edge features Download PDFInfo
- Publication number
- CN110648368B CN110648368B CN201910817095.9A CN201910817095A CN110648368B CN 110648368 B CN110648368 B CN 110648368B CN 201910817095 A CN201910817095 A CN 201910817095A CN 110648368 B CN110648368 B CN 110648368B
- Authority
- CN
- China
- Prior art keywords
- points
- edge
- feature
- point
- matrix
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- 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
- G06T2207/20112—Image segmentation details
- G06T2207/20164—Salient point detection; Corner detection
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention belongs to the technical field of machine vision, and particularly relates to a calibration board corner point distinguishing method based on edge features, which comprises the following steps: s1, in the calibration plate image, a series of Harris angular points are obtained as feature points, a sampling radius is set for the feature points to be distinguished, and a neighborhood image in the sampling radius is obtained by taking the feature point coordinates as the center; s2, extracting Canny edges of the neighborhood image of the current feature point, supposing that n edge points are obtained, and arranging the coordinates of the n edge points into an n multiplied by 2 matrix called an edge matrix; s3, performing singular value decomposition (i.e., SVD decomposition) on the edge coordinate matrix of the current feature point to obtain an SVD feature value vector of the edge coordinate matrix, and marking as (v1, v2), which is called an edge feature vector. The method can better judge according to the edge shape characteristics of the checkerboard angular points, and can remove abnormal points with similar checkerboard template convolution response values or Harris angular point strength values.
Description
Technical Field
The invention belongs to the technical field of machine vision, and particularly relates to a calibration board corner point distinguishing method based on edge features.
Background
In machine vision, parameter calibration is often required to be performed on a camera, a checkerboard calibration plate is adopted in the current mainstream camera calibration, a plurality of checkerboard calibration plate images are shot by the camera, a checkerboard template convolution or Harris corner detection method is performed on the calibration plate images, a series of feature points are preliminarily obtained, under the condition that the calibration plate background and the environment are complex, the feature points are often not completely checkerboard corner points and are difficult to be directly subjected to fitting calibration, an effective checkerboard corner point in the feature points needs to be screened through a specific judging method, and then camera parameters are calibrated through a fitting mode according to image coordinates and world coordinates of the checkerboard corner points obtained through judgment. Therefore, the method for judging the checkerboard corner points in the calibration board image is one of the key technologies for calibrating the camera parameters.
In the prior art, feature points obtained by performing a checkerboard template convolution or Harris corner detection method on a calibration board image are generally set with a threshold according to a convolution response value or a Harris corner intensity value as a discrimination condition, and a checkerboard corner is considered when the convolution response value or the Harris corner intensity value is greater than the threshold. The prior art is deficient in two aspects: firstly, due to the background interference of a calibration plate, characteristic points of individual abnormity still obtain higher convolution response values or Harris angular point strength values, and the abnormal points cannot be removed through simple threshold value screening, so that abnormal points exist in an output result, and the success rate and the precision of subsequent calibration are influenced. Secondly, in order to eliminate abnormal feature points as much as possible, the threshold value is often set to be high, so that part of the normal checkerboard corner points with low response values are mistakenly eliminated, and the calibration board is failed to detect or calibrate.
Disclosure of Invention
Aiming at the defects in the prior art, a calibration board corner point distinguishing method based on edge features is provided to make up for the defects in the background art.
In order to achieve the purpose, the invention adopts the following technical scheme:
a calibration board angular point distinguishing method based on edge features comprises the following steps:
s1, in the calibration plate image, a series of Harris angular points are obtained as feature points, a sampling radius is set for the feature points to be distinguished, and a neighborhood image in the sampling radius is obtained by taking the feature point coordinates as the center;
s2, extracting Canny edges of the neighborhood image of the current feature point, supposing that n edge points are obtained, and arranging the coordinates of the n edge points into an n multiplied by 2 matrix called an edge matrix;
s3, performing singular value decomposition (namely SVD) on the edge coordinate matrix of the current feature point to obtain an SVD feature value vector of the edge coordinate matrix, and marking the SVD feature value vector as (v1, v2), which is called as an edge feature vector;
s4, calculating the edge feature vectors of all the feature points, and calculating the mean value of all the edge feature vectors
S5, andas a center, calculating edge feature vectors of all feature points toAnd calculating the standard deviation delta of the distance;
s6, using two times of standard deviation, namely 2 delta as the discrimination condition, if the feature point is up toIf the distance is less than 2 delta, the corner points are regarded as the checkerboard corner points, otherwise, the corner points are regarded as abnormal points to be removed. And judging all the characteristic points and outputting the judgment result.
Compared with the prior art, the method has the advantages that: 1. aiming at the problem that only the convolution response value of a checkerboard template or the intensity of a Harris angular point is used as a distinguishing characteristic in the prior art, the method has the advantages that: the method comprises the following steps of extracting edges of adjacent areas of points to be judged, obtaining an edge matrix, and judging whether the points belong to checkerboard angular points or not by taking SVD (singular value decomposition) eigenvectors of the edge matrix as edge eigenvectors, wherein the edge extraction method has the following effects: the judgment can be better carried out according to the edge shape characteristics of the checkerboard angular points, and abnormal points with similar checkerboard template convolution response values or Harris angular point strength values can be eliminated;
2. aiming at the problem that in the prior art, a fixed threshold is set, and whether a checkerboard template convolution response value or a Harris corner strength value is larger than the fixed threshold is only used as a distinguishing mode, the method has the advantages that: the mean value of the edge feature vectors of all points to be judged is taken as the center, the double value of the standard deviation of the distances from the edge feature vectors of all points to be judged to the center is taken as the judgment threshold, the points to be judged, the distances from the edge feature vectors to the center of which are less than the judgment threshold, are taken as the checkerboard corner points, and the checkerboard corner point has the effects that: the discrimination threshold value can be determined in a self-adaptive manner, and the adaptability of the method to different environment images is improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
FIG. 1 is a flow chart of a method in an embodiment of the invention;
Detailed Description
As used in the specification and in the claims, certain terms are used to refer to particular components. As one skilled in the art will appreciate, manufacturers may refer to a component by different names. The specification and the claims do not use the difference of names as the way of distinguishing the components, but use the difference of functions of the components as the criterion of distinguishing, such as a pointer, which can be replaced by an iterator. In the following description and in the claims, the terms "include" and "comprise" are used in an open-ended fashion, and thus should be interpreted to mean "include, but not limited to. "substantially" means within an acceptable error range, within which a person skilled in the art can solve the technical problem to substantially achieve the technical result. Further, it is not necessary for certain coefficients or thresholds in the specification and claims to be specific values, but rather values are generally appropriate and some increase or decrease may be possible.
As shown in fig. 1, a method for distinguishing corner points of a checkerboard of a punctuation board based on edge features comprises the following steps:
step 1, performing Harris angular point detection on a checkerboard calibration board image to obtain m characteristic points;
step 2, setting a sampling radius r for a certain characteristic point i in the step 1, and sampling by taking the image coordinate of the characteristic point i as the center and a rectangular sampling window with the side length of 2 x r +1 in a checkerboard image to obtain a neighborhood image of the characteristic point i;
step 3, Canny edge detection is carried out on the neighborhood image of the current feature point i, and n detected edge point coordinates form an nx2 matrix M;
step 4, carrying out SVD on the matrix M to obtain an eigenvalue vector (v 1)i,v2i);
Step 5, repeating the steps 2 to 4, and obtaining corresponding eigenvalue vectors (v 1) of the m eigenvalues obtained in the step 11,v21),(v12,v22),(v13,v23)...(v1m,v2m);
Step 6, solving the mean value of all the characteristic value vectors according to the following equation to obtain
Step 7, calculating the eigenvalue vectors of all the characteristic points according to the following equationDistance d of1,d2,d3...dm:
Step 8, solving the eigenvalue vectors of all the characteristic points according to the following equationThe standard deviation δ of the distance of (a):
step 9, judging whether the distance of each feature point is less than 2 delta, if so, regarding the feature point as a checkerboard angular point, and if not, rejecting the feature point;
and step 10, outputting a judgment result. The judgment can be better carried out according to the edge shape characteristics of the checkerboard angular points, and abnormal points with similar checkerboard template convolution response values or Harris angular point strength values can be eliminated; the mean value of the edge feature vectors of all points to be judged is taken as the center, the double value of the standard deviation of the distances from the edge feature vectors of all points to be judged to the center is taken as the judgment threshold, the points to be judged, the distances from the edge feature vectors to the center of which are less than the judgment threshold, are taken as the checkerboard corner points, and the checkerboard corner point has the effects that: the discrimination threshold value can be determined in a self-adaptive manner, and the adaptability of the method to different environment images is improved.
While the foregoing specification illustrates and describes several embodiments of the invention, it is to be understood, as noted above, that the invention is not limited to the forms disclosed herein, but is not intended to be exhaustive of other embodiments, and is capable of use in various other combinations, modifications, and environments and is capable of changes within the scope of the inventive concept as described herein, commensurate with the above teachings, or the skill or knowledge of the relevant art. And that modifications and variations may be effected by those skilled in the art without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (1)
1. A calibration board angular point distinguishing method based on edge features is characterized by comprising the following steps:
s1, in the calibration plate image, a series of Harris angular points are obtained as feature points, a sampling radius is set for the feature points to be judged, and a neighborhood image within the sampling radius is obtained by taking the feature point coordinates as the center;
s2, extracting Canny edges of the neighborhood image of the current feature point, supposing that n edge points are obtained, and arranging the coordinates of the n edge points into an nx2 matrix which is called an edge matrix;
s3, performing singular value decomposition (namely SVD) on the edge coordinate matrix of the current feature point to obtain an SVD feature value vector of the edge coordinate matrix, and marking the SVD feature value vector as (v1, v2), which is called as an edge feature vector;
s4, calculating the edge feature vectors of all the feature points, and calculating the mean value of all the edge feature vectors
S5, andas a center, calculating edge feature vectors of all feature points toAnd calculating a standard deviation δ of the distance as follows:
Step 2: calculating the edge feature vector of all feature points according to the following equationThe standard deviation δ of the distance of (a):
s6, using two times of standard deviation, namely 2 delta as the discrimination condition, if the feature point is up toIs less than 2 delta, the chess is regarded as the chessAnd (4) checking the angular points, otherwise, rejecting the angular points as abnormal points, judging all the characteristic points, and outputting the judgment result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910817095.9A CN110648368B (en) | 2019-08-30 | 2019-08-30 | Calibration board corner point discrimination method based on edge features |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910817095.9A CN110648368B (en) | 2019-08-30 | 2019-08-30 | Calibration board corner point discrimination method based on edge features |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110648368A CN110648368A (en) | 2020-01-03 |
CN110648368B true CN110648368B (en) | 2022-05-17 |
Family
ID=69009953
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910817095.9A Active CN110648368B (en) | 2019-08-30 | 2019-08-30 | Calibration board corner point discrimination method based on edge features |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110648368B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112614146B (en) * | 2020-12-21 | 2022-05-13 | 广东奥普特科技股份有限公司 | Method and device for judging chessboard calibration corner points and computer readable storage medium |
CN113379814B (en) * | 2021-06-09 | 2024-04-09 | 北京超图软件股份有限公司 | Three-dimensional space relation judging method and device |
CN115222825B (en) * | 2022-09-15 | 2022-12-16 | 湖南视比特机器人有限公司 | Calibration method, computer storage medium and calibration system |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103745449A (en) * | 2013-12-24 | 2014-04-23 | 南京理工大学 | Rapid and automatic mosaic technology of aerial video in search and tracking system |
CN106447708A (en) * | 2016-10-10 | 2017-02-22 | 吉林大学 | OCT eye fundus image data registration method |
CN106546233A (en) * | 2016-10-31 | 2017-03-29 | 西北工业大学 | A kind of monocular visual positioning method towards cooperative target |
CN106570887A (en) * | 2016-11-04 | 2017-04-19 | 天津大学 | Adaptive Mean Shift target tracking method based on LBP features |
CN109241981A (en) * | 2018-09-03 | 2019-01-18 | 哈尔滨工业大学 | A kind of characteristic detection method based on sparse coding |
CN109509200A (en) * | 2018-12-26 | 2019-03-22 | 深圳市繁维医疗科技有限公司 | Checkerboard angle point detection process, device and computer readable storage medium based on contours extract |
-
2019
- 2019-08-30 CN CN201910817095.9A patent/CN110648368B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103745449A (en) * | 2013-12-24 | 2014-04-23 | 南京理工大学 | Rapid and automatic mosaic technology of aerial video in search and tracking system |
CN106447708A (en) * | 2016-10-10 | 2017-02-22 | 吉林大学 | OCT eye fundus image data registration method |
CN106546233A (en) * | 2016-10-31 | 2017-03-29 | 西北工业大学 | A kind of monocular visual positioning method towards cooperative target |
CN106570887A (en) * | 2016-11-04 | 2017-04-19 | 天津大学 | Adaptive Mean Shift target tracking method based on LBP features |
CN109241981A (en) * | 2018-09-03 | 2019-01-18 | 哈尔滨工业大学 | A kind of characteristic detection method based on sparse coding |
CN109509200A (en) * | 2018-12-26 | 2019-03-22 | 深圳市繁维医疗科技有限公司 | Checkerboard angle point detection process, device and computer readable storage medium based on contours extract |
Non-Patent Citations (4)
Title |
---|
"Accurate Corner Detection Methods using Two Step Approach";Nitin Bhatia et al.;《GLOBAL JOURNAL OF COMPUTER SCIENCE & TECHNOLOGY》;20110430;第11卷(第6期);全文 * |
"The coener matching based on improved singular value decomposition for motion detection";Kang Meng et al.;《IEEE》;20121224;全文 * |
"基于SUSAN算子和角点判别因子的目标边缘检测";吴一全 等;《中国科学院大学学报》;20160131;第33卷(第1期);全文 * |
"基于SUSAN算子的特征检测方法研究";王冠群;《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑》;20160315(第03期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN110648368A (en) | 2020-01-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110648368B (en) | Calibration board corner point discrimination method based on edge features | |
US7746505B2 (en) | Image quality improving apparatus and method using detected edges | |
US10038862B2 (en) | Methods and apparatus for automated noise and texture optimization of digital image sensors | |
US9811899B2 (en) | Method, image processing system and computer-readable recording medium for item defect inspection | |
US20090040343A1 (en) | Methods and apparatuses for defective pixel detection and correction | |
US20060271226A1 (en) | Inspection standard setting device, inspection standard setting method and process inspection device | |
US10348989B2 (en) | Image processing device, image processing method, and image processing system | |
US20070189606A1 (en) | Automatic detection and correction of non-red eye flash defects | |
KR101243363B1 (en) | Apparatus and method for concealing a dead pixel | |
KR101665137B1 (en) | Apparatus and method for reducing noise from image sensor | |
US9077873B2 (en) | Color filter array defect correction | |
JP5660861B2 (en) | Foreign matter inspection method and foreign matter inspection apparatus on substrate | |
CN111445502B (en) | Integrated circuit image matching method and system | |
EP3119080B1 (en) | Method and system for dead pixel correction of digital image | |
US8594416B2 (en) | Image processing apparatus, image processing method, and computer program | |
WO2017185507A1 (en) | Method and device for identifying multiple fingerprints | |
KR101583423B1 (en) | Method for calibrating distortion of image in camera | |
US8687921B2 (en) | Image processing apparatus, image processing method, and image processing program | |
CN110634146B (en) | Circle center sub-pixel precision positioning method | |
KR100827906B1 (en) | Substrate inspection device | |
CN111403318B (en) | Method and device for detecting wafer state in process chamber | |
US8345965B2 (en) | Method and apparatus for positioning edges of photograph | |
US8866903B2 (en) | Image processing apparatus, image processing method, and computer program | |
US20130142441A1 (en) | Information processing apparatus and method | |
JP4560434B2 (en) | Change region extraction method and program of the method |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |