CN112200785B - Improved digital image correlation method based on random scattered point relation topology matching function - Google Patents
Improved digital image correlation method based on random scattered point relation topology matching function Download PDFInfo
- Publication number
- CN112200785B CN112200785B CN202011098896.3A CN202011098896A CN112200785B CN 112200785 B CN112200785 B CN 112200785B CN 202011098896 A CN202011098896 A CN 202011098896A CN 112200785 B CN112200785 B CN 112200785B
- Authority
- CN
- China
- Prior art keywords
- point
- points
- deformation
- random
- 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
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000006073 displacement reaction Methods 0.000 claims abstract description 5
- 239000011159 matrix material Substances 0.000 claims description 56
- 238000005314 correlation function Methods 0.000 claims description 20
- 238000010586 diagram Methods 0.000 claims description 5
- 238000005259 measurement Methods 0.000 abstract description 7
- 238000004364 calculation method Methods 0.000 description 4
- 230000006978 adaptation Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Classifications
-
- 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
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention provides an improved digital image correlation method based on a random scattered point relation topology matching function, and belongs to the technical field of deformation measurement. The method comprises the steps of firstly obtaining two random scatter digital images before and after deformation, and marking the random scatter digital images as img1 and img2. Each point in the scatter images is then matched one by one in img1 and img2 until each point in the last random scatter plot is matched. According to the matched result, the pixel offset condition of the two images, namely pixel displacement data, can be correspondingly calculated, and finally deformation information before and after deformation is obtained. The method uses a "randomly distributed speckle pattern" that is easy to arrange as a characteristic speckle for correlation matching. Then, the 'topological position relation' of the discrete points which are randomly distributed before and after deformation is calculated, so that the images before and after deformation are subjected to correlation matching, and finally the whole-field deformation information is obtained.
Description
Technical Field
The invention relates to the technical field of deformation measurement, in particular to an improved digital image correlation method based on a random scattered point relation topology matching function.
Background
In recent years, a non-contact deformation measurement technology based on a digital image correlation method is widely applied to the fields of optometry mechanics, deformation measurement and various engineering measurement. The digital image correlation method is at the earliest mountain mouth Yilang from japan in the beginning of the 80 th century [1] And Peter and Ranson et al, university of south Carolina, U.S [1-2] It is proposed independently of each other that the position of the sub-region before deformation in the image after deformation is determined by comparing the gray levels of the sub-region in the digital image before and after deformation.
The digital image correlation method uses a pattern of features called speckle patterns, which are captured by an industrial camera as digital speckle patterns. Fig. 1 (b) is a conventional digital speckle pattern (referred to as an image in a computer), also referred to as a random speckle pattern, characterized by: the gray level of the spots in the image (i.e., the black shade) and the positions and sizes of the spots are randomly distributed, and the digital speckle pattern is herein referred to as a "digital image" of the digital image correlation method. The correlation of the two digital speckle patterns is then typically calculated using a conventional least squares distance sum function (SSD, ZNSSD) or cross correlation function (CC, ZNCC), the function used being the "correlation function".
The existing conventional "random speckle digital image" and conventional "correlation function" (ZNCC, etc.) have the following problems in practical applications:
1. random speckle pattern placement requires skill and is difficult to master quickly. The conventional random speckle pattern requires both random spot size and random spot location.
2. The speckle pattern correlation operation takes longer. As can be seen from practical use, the mathematical operation of the conventional correlation function is huge, and the required calculation time is long.
3. There are some cases where the related function fails. Firstly, the deformation of the measured object is too large, so that the related search of the DIC method fails, and the related function of the DIC method fails, so that the measurement fails. The above-described problems are referred to as large deformation problems or decorrelation problems, which have become a common problem for 2D/3D DIC. The study by the learner found that the traditional correlation function only can realize the large deformation measurement of up to about 20% strain at most.
Reference is made to:
[1]Yamaguchi I.A laser-speckle strain gauge[J].Journal of Physics E:Scientific Instruments.1981,14:1270～1273.
[2]W.H.Peters,W.F.Ranson.Digital Imaging Techniques in Experimental Stress Analysis[J].Optical Engineering.1981，21:427～431.
disclosure of Invention
The technical problem to be solved by the invention is to provide an improved digital image correlation method based on a random scattered point relation topological matching function, which uses a random distributed scattered point pattern easy to arrange as characteristic speckle of correlation matching. Then, calculating the 'topological position relation' of the discrete points which are randomly distributed before and after deformation, so as to perform correlation matching on the images before and after deformation, and finally obtaining full-field deformation information. The "random speckle pattern" is improved using a "random distribution speckle pattern", and the traditional "correlation function" method is improved using a matching function that calculates the "topological positional relationship". Finally, the improved method is realized by relying on random distribution scattered points capable of resisting large deformation, and the problem of decorrelation caused by the large deformation in calculation is avoided, so that the large deformation correlation matching operation without replacing the reference frame is realized. The method has the characteristics of high calculation speed and suitability for other special deformation conditions such as large deformation.
The method specifically comprises the following steps:
s1: acquiring two random scatter digital images before and after deformation, wherein the random scatter digital images are marked as img1 and img2, img1 represents an original image, and img2 represents a deformed image; then each point in the scattered point image is matched in img1 and img2 one by one, and the specific process is as follows:
s11: setting the size of a relation point set as r, taking a key point P1 to be matched in img1, taking r points (at least 5 points, 5-10 points are optimal, and more points are determined according to matching correlation) around the P1 point as the relation point set of the P1 point, and using an adjacency matrix solving algorithm of a topology improvement correlation function (an adjacency matrix representing the property of the key point is formed according to Euclidean distance between the key point and all points of surrounding relation points): calculating Euclidean distances among all points including the P1 point and the relation points thereof to form an adjacent matrix representing the property of the point set, and marking the adjacent matrix as M1;
s12: in img2, taking the pixel coordinate of the original P1 point as the center, and taking the pixel coordinate as an initial matching position in img 2;
s13: setting the size of a search point set as s, searching s points (about 5 points, generally 3-8 points, about 20% of small deformation and more points are required to be determined according to the deformation) around an initial matching position of img2 as a search point set, marking the search point set as P2_1-s, then taking each point in the point set and searching r points around the point set as a relation point set of the point;
s14: calculating an adjacency matrix of P2_1-s points by using an adjacency matrix solving algorithm of a topology improvement correlation function, and marking the adjacency matrix as M2_1-s;
s15: respectively making differences between the matrixes M2-1-S in the S14 and the matrix M1 obtained in the S11, then calculating eigenvalues of the matrixes after the differences, and finally forming a row matrix Eige_1xS of 1xS;
s16: searching the minimum value position of the row matrix Eige_1xS, and if not, increasing the number r of the relation point sets and the size s of the search point sets; if so, correspondingly finding an M2_ matrix corresponding to the minimum value position, and thus positioning P2_ points in the search point set;
s2: repeating S11-S16 to match all points in img1 and img2 until each point in the random scatter diagram is matched finally;
s3: and correspondingly calculating pixel offset conditions of the two images, namely pixel displacement data, according to the matched result, and finally obtaining deformation information before and after deformation.
Wherein, the shape of the scattered points in the random scattered point digital image in the S1 is a two-dimensional shape with a fixed size, including a circle, a rectangle, a triangle and the like.
The P2_ point in the set of locating search points in S15 is specifically: firstly, the matrix is subjected to difference, then, matrix eigenvalues after the difference are obtained, all eigenvalues are formed into a row matrix, the minimum value of the row matrix is obtained, and the corresponding position of the minimum value is the P2X point.
The technical scheme of the invention has the following beneficial effects:
in the scheme, the random scatter pattern with randomly distributed positions and consistent sizes is used, the arrangement difficulty is low, and only the positions of the scattered points need to be controlled. Unlike conventional random speckle patterns, it is necessary to pay attention to both the random size of the speckle and the inability of the speckle to coincide, i.e., the random location. The invention designs a topology improvement correlation function method for performing correlation matching on random scattered points in images before and after deformation. The method uses the connectivity concept and the adjacency matrix definition concept in the topology theory to judge the similarity of the scattered points by using the adjacency matrix and the characteristic value of each point in the random scattered points, thereby realizing the image correlation matching before and after deformation. The invention provides an effective improvement for a digital image correlation method, provides random scattered point patterns which are convenient to arrange, and provides a topological improvement correlation function for matching the random scattered point patterns before and after deformation. The development of non-contact material/component deformation detection technology based on a digital image correlation method can be effectively promoted.
Drawings
FIG. 1 is a comparison of the form of a random scattergram used in the present invention with a conventional random speckle pattern, wherein (a) is a random scattergram used in the present invention and (b) is a conventional random speckle pattern;
FIG. 2 is a schematic diagram of a relational topology improvement correlation function method for performing correlation matching on random scattered points used in the invention;
FIG. 3 is a flow chart of a method of the improved digital image correlation method based on a random scattered point relationship topology matching function of the present invention.
Detailed Description
In order to make the technical problems, technical solutions and advantages to be solved more apparent, the following detailed description will be given with reference to the accompanying drawings and specific embodiments.
The invention provides an improved digital image correlation method based on a random scattered point relation topology matching function.
The "random distribution scatter pattern" used in the present invention is shown in fig. 1 (a). In comparison to the random speckle pattern (fig. 1 (b)), the difference is that the size of individual feature points of the speckle pattern is fixed, and the same points are that the positions of the speckle points are randomly distributed. Then, by means of the topological relation feature definition method, the feature point positions which are randomly distributed can provide the topological relation feature which is specific to each feature point relative to other feature points, so that the similarity of the feature points is judged.
The method for improving the correlation function by the relation topology for carrying out correlation matching on random scattered points in the images before and after deformation is the core of the method provided by the invention. General topology is a principle study in mathematics of the nature of topological space and its related concepts, where connectivity and compactness are important concepts in topology. First, the present invention uses the concept of connectivity to identify connectivity and non-connectivity to the random scatter plot of FIG. 1 (a), thereby forming a topological relationship property. The present invention then defines that a range or number of other points around the key matching point (e.g., point P3 in fig. 2) have connectivity with the feature point, forming a set of points (e.g., points P1-P5 in fig. 2), and then using the euclidean distance between all points in the set of points to form a adjacency matrix that characterizes the nature of the set of points. Finally, the similarity of the point set is judged by comparing the minimum value of the characteristic values after the adjacent matrix is differenced, and the point which is matched finally is the smallest difference, so that the matching process between the characteristic points is obtained. The calculation is the core of the method, and the relation topology of the random scatter diagram is used for improving the digital image correlation method of the correlation function.
As shown in fig. 3, the method of the present invention specifically includes the following steps:
s1: acquiring two random scatter digital images before and after deformation, wherein the random scatter digital images are marked as img1 and img2, img1 represents an original image, and img2 represents a deformed image; then each point in the scattered point image is matched in img1 and img2 one by one, and the specific process is as follows:
s11: setting the size of a relation point set as r, taking a key point P1 to be matched in img1, taking r points (at least 5 points, 5-10 points are optimal, and more points are determined according to matching correlation) around the P1 point as the relation point set of the P1 point, and using an adjacency matrix solving algorithm of a topology improvement correlation function (an adjacency matrix representing the property of the key point is formed according to Euclidean distance between the key point and all points of surrounding relation points): calculating Euclidean distances among all points including the P1 point and the relation points thereof to form an adjacent matrix representing the property of the point set, and marking the adjacent matrix as M1;
s12: in img2, taking the pixel coordinate of the original P1 point as the center, and taking the pixel coordinate as an initial matching position in img 2;
s13: setting the size of a search point set as s, searching s points (small deformation, about 5 points and large deformation of 5-20 points, more points are determined according to the deformation amount) around the initial matching position of img2 as the search point set, marking the search point set as P2_1-s, and then taking each point in the point set and searching r points around the point set as a relation point set of the point;
s14: calculating an adjacency matrix of P2_1-s points by using an adjacency matrix solving algorithm of a topology improvement correlation function, and marking the adjacency matrix as M2_1-s;
s15: respectively making differences between the matrixes M2-1-S in the S14 and the matrix M1 obtained in the S11, then calculating eigenvalues of the matrixes after the differences, and finally forming a row matrix Eige_1xS of 1xS;
s16: searching the minimum value position of the row matrix Eige_1xS, and if not, increasing the number r of the relation point sets and the size s of the search point sets; if so, correspondingly finding an M2_ matrix corresponding to the minimum value position, and thus positioning P2_ points in the search point set;
s2: repeating S11-S16 to match all points in img1 and img2 until each point in the random scatter diagram is matched finally;
s3: and correspondingly calculating pixel offset conditions of the two images, namely pixel displacement data, according to the matched result, and finally obtaining deformation information before and after deformation.
The following description is made in connection with the specific implementation.
Step 1, two random scatter digital images before deformation and after deformation are obtained, and img1 represents an original image and img2 deformation images. Then, each point in the scatter images is matched one by one in img1 and img2, taking matching of one scatter as an example.
And 2, setting the size of the relation point set as r. In img1, a key point to be matched is taken, wherein P1 is taken as the key point to be matched, and r points around the P1 point are taken as a relation point set of the P1 points. The topological improvement correlation function provided by the invention is used for calculating the Euclidean distance between the P1 point and all points including the relation point thereof to form an adjacency matrix representing the property of the point set, and the adjacency matrix is marked as M1.
And 3, in img2, taking the pixel coordinate where the original P1 point is as the center, and taking the pixel coordinate as an initial matching position in img2.
And 4, setting the size of the search point set as s. Searching s points around the initial matching position of img2 as a searching point set, namely P2_1-s, then taking each point in the point set and searching r points around the point set as a relation point set of the point, such as P2-2 points and r points around P2-2 points.
And 5, calculating an adjacency matrix of the P2_1-s point by using the topology improvement correlation function provided by the invention, and marking the adjacency matrix as M2_1-s.
Step 6: and (3) respectively differencing the matrixes M2-1-s obtained in the step (5) with the matrix M1 obtained in the step (2), then calculating the eigenvalue of the matrix after the differencing, and finally forming a row matrix Eige_1xS of 1xS.
Step 7: searching the minimum value position of the row matrix eige_1xs, and correspondingly finding the M2 x matrix corresponding to the minimum value position, so as to locate the P2 x point in the search point set, for example, the P2 x point corresponds to the minimum value of the row matrix, and then determining that the P2 x 2 point in img2 is the matching point of the P1 point in img 1.
Step 8: repeating steps 2-7 to match all points in img1 and img2 until each point in the final random scatter plot is matched.
Step 9: according to the matched result, the pixel offset condition of the two images, namely pixel displacement data, can be correspondingly calculated, and finally deformation information before and after deformation is obtained.
While the foregoing is directed to the preferred embodiments of the present invention, it will be appreciated by those skilled in the art that various modifications and adaptations can be made without departing from the principles of the present invention, and such modifications and adaptations are intended to be comprehended within the scope of the present invention.
Claims (5)
1. An improved digital image correlation method based on a random scattered point relation topology matching function is characterized by comprising the following steps of: firstly, using a random distribution scatter pattern as a relevant matched characteristic speckle, and then calculating the topological position relation of random distribution discrete points before and after deformation, so as to perform relevant matching on images before and after deformation, and finally obtaining full-field deformation information;
the method specifically comprises the following steps:
s1: acquiring two random scatter digital images before and after deformation, wherein the random scatter digital images are marked as img1 and img2, img1 represents an original image, and img2 represents a deformed image; then each point in the scattered point image is matched in img1 and img2 one by one, and the specific process is as follows:
s11: setting the size of a relation point set as r, taking a key point P1 to be matched in img1, taking r points around the P1 point as the relation point set of the P1 point, and using an adjacency matrix solving algorithm of a topology improvement correlation function: calculating Euclidean distances among all points including the P1 point and the relation points thereof to form an adjacent matrix representing the property of the point set, and marking the adjacent matrix as M1;
s12: in img2, taking the pixel coordinate of the original P1 point as the center, and taking the pixel coordinate as an initial matching position in img 2;
s13: setting the size of a search point set as s, searching s points around an initial matching position of img2 as the search point set, marking as P2_1-s, then taking each point in the point set and searching r points around the point set as a relation point set of the point;
s14: calculating an adjacency matrix of P2_1-s points by using an adjacency matrix solving algorithm of a topology improvement correlation function, and marking the adjacency matrix as M2_1-s;
s15: respectively making differences between the matrixes M2-1-S in the S14 and the matrix M1 obtained in the S11, then calculating eigenvalues of the matrixes after the differences, and finally forming a row matrix Eige_1xS of 1xS;
s16: searching the minimum value position of the row matrix Eige_1xS, and if not, increasing the number r of the relation point sets and the size s of the search point sets; if so, correspondingly finding an M2_ matrix corresponding to the minimum value position, and thus positioning P2_ points in the search point set;
s2: repeating S11-S16 to match all points in img1 and img2 until each point in the random scatter diagram is matched finally;
s3: and correspondingly calculating pixel offset conditions of the two images, namely pixel displacement data, according to the matched result, and finally obtaining deformation information before and after deformation.
2. The improved digital image correlation method based on a random speckle relationship topology matching function of claim 1, wherein: the shape of the scattered points in the random scattered point digital image in the S1 is a two-dimensional shape with a fixed size, including a circle, a rectangle and a triangle.
3. The improved digital image correlation method based on a random speckle relationship topology matching function of claim 1, wherein: the adjacency matrix solving algorithm of the topology improvement correlation function used in S11 is to form an adjacency matrix representing the property of the key point according to the euclidean distance between the key point and all points of the surrounding relationship points.
4. The improved digital image correlation method based on a random speckle relationship topology matching function of claim 1, wherein: the P2_ point in the positioning search point set in S15 is specifically: firstly, the matrix is subjected to difference, then, matrix eigenvalues after the difference are obtained, all eigenvalues are formed into a row matrix, the minimum value of the row matrix is obtained, and the corresponding position of the minimum value is the P2X point.
5. The improved digital image correlation method based on a random speckle relationship topology matching function of claim 1, wherein: r is more than or equal to 5; s is 3 to 8 when the deformation is less than or equal to 20 percent; when the deformation is more than 20%, s is 5-20.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011098896.3A CN112200785B (en) | 2020-10-14 | 2020-10-14 | Improved digital image correlation method based on random scattered point relation topology matching function |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011098896.3A CN112200785B (en) | 2020-10-14 | 2020-10-14 | Improved digital image correlation method based on random scattered point relation topology matching function |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112200785A CN112200785A (en) | 2021-01-08 |
| CN112200785B true CN112200785B (en) | 2023-12-29 |
Family
ID=74008965
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011098896.3A Active CN112200785B (en) | 2020-10-14 | 2020-10-14 | Improved digital image correlation method based on random scattered point relation topology matching function |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112200785B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113516622B (en) * | 2021-04-23 | 2023-02-17 | 天津大学 | Visual detection pattern and method integrating digital speckles, grid lines and coding mark points |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH01320411A (en) * | 1988-06-22 | 1989-12-26 | Hamamatsu Photonics Kk | Measuring method for deformation |
| CN101540061A (en) * | 2009-04-10 | 2009-09-23 | 西北工业大学 | Topological and ordering matching method for disordered images based on simulated annealing |
| CN101608908A (en) * | 2009-07-20 | 2009-12-23 | 杭州先临三维科技股份有限公司 | The three-dimension digital imaging method that digital speckle projection and phase measuring profilometer combine |
| CN102221341A (en) * | 2011-03-16 | 2011-10-19 | 中国人民解放军国防科学技术大学 | Quick digital image correlation measurement method based on stochastic parallel gradient descent optimization technology |
| CN110443261A (en) * | 2019-08-15 | 2019-11-12 | 南京邮电大学 | A kind of more figure matching process restored based on low-rank tensor |
| CN110456010A (en) * | 2019-07-04 | 2019-11-15 | 南开大学 | A kind of method of lossless quick detection egg freshness |
| CN110779454A (en) * | 2019-10-10 | 2020-02-11 | 北京科技大学 | Improved digital image correlation method based on double-channel structure speckle cross-correlation algorithm |
| CN111156917A (en) * | 2020-01-17 | 2020-05-15 | 北京林业大学 | Deformation measurement method based on gray level mark points |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6990254B2 (en) * | 2001-08-06 | 2006-01-24 | Mitutoyo Corporation | Systems and methods for correlating images in an image correlation system with reduced computational loads |
| US8600147B2 (en) * | 2009-06-03 | 2013-12-03 | The United States of America as represented by the Secreatary of the Navy | System and method for remote measurement of displacement and strain fields |
-
2020
- 2020-10-14 CN CN202011098896.3A patent/CN112200785B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH01320411A (en) * | 1988-06-22 | 1989-12-26 | Hamamatsu Photonics Kk | Measuring method for deformation |
| CN101540061A (en) * | 2009-04-10 | 2009-09-23 | 西北工业大学 | Topological and ordering matching method for disordered images based on simulated annealing |
| CN101608908A (en) * | 2009-07-20 | 2009-12-23 | 杭州先临三维科技股份有限公司 | The three-dimension digital imaging method that digital speckle projection and phase measuring profilometer combine |
| CN102221341A (en) * | 2011-03-16 | 2011-10-19 | 中国人民解放军国防科学技术大学 | Quick digital image correlation measurement method based on stochastic parallel gradient descent optimization technology |
| CN110456010A (en) * | 2019-07-04 | 2019-11-15 | 南开大学 | A kind of method of lossless quick detection egg freshness |
| CN110443261A (en) * | 2019-08-15 | 2019-11-12 | 南京邮电大学 | A kind of more figure matching process restored based on low-rank tensor |
| CN110779454A (en) * | 2019-10-10 | 2020-02-11 | 北京科技大学 | Improved digital image correlation method based on double-channel structure speckle cross-correlation algorithm |
| CN111156917A (en) * | 2020-01-17 | 2020-05-15 | 北京林业大学 | Deformation measurement method based on gray level mark points |
Non-Patent Citations (2)
| Title |
|---|
| "数字图像相关法散斑图质量评价方法";俞海，刘云鹏，郭荣鑫，夏海廷，颜峰;《激光技术》;237-243 * |
| Ma, XD ; Yang, JH and Li, L." Design of the optical Structure of a LED Light of Airfield used on the Taxiway Centerline of Bend".《 INTERNATIONAL SYMPOSIUM ON PHOTONICS AND OPTOELECTRONICS 2014》.2014,全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112200785A (en) | 2021-01-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11551341B2 (en) | Method and device for automatically drawing structural cracks and precisely measuring widths thereof | |
| Kassim et al. | A comparative study of efficient generalised Hough transform techniques | |
| CN109544599B (en) | Three-dimensional point cloud registration method based on camera pose estimation | |
| CN101639947A (en) | Image-based plant three-dimensional shape measurement and reconstruction method and system | |
| CN106548462A (en) | Non-linear SAR image geometric correction method based on thin-plate spline interpolation | |
| CN112200785B (en) | Improved digital image correlation method based on random scattered point relation topology matching function | |
| CN111256607B (en) | Deformation measurement method based on three-channel mark points | |
| CN112184783A (en) | Three-dimensional point cloud registration method combined with image information | |
| US20210358133A1 (en) | Method and device for estimating background motion of infrared image sequences and storage medium | |
| CN114494265A (en) | Method for identifying packaging printing quality in cosmetic production field and artificial intelligence system | |
| WO2018133027A1 (en) | Grayscale constraint-based method and apparatus for integer-pixel search for three-dimensional digital speckle pattern | |
| CN112050752B (en) | Projector calibration method based on secondary projection | |
| CN112233104B (en) | Real-time displacement field and strain field detection method, system, device and storage medium | |
| CN114266899A (en) | Image target parallel detection method based on multi-core DSP | |
| US11348261B2 (en) | Method for processing three-dimensional point cloud data | |
| CN101082987B (en) | Column diagram comparability measurement method based on average difference between windows | |
| Yang et al. | Center extraction algorithm of linear structured light stripe based on improved gray barycenter method | |
| CN114511631A (en) | Method and device for measuring height of visual object of camera and computer readable storage medium | |
| CN111401385A (en) | Similarity calculation method for image local topological structure feature descriptors | |
| Gonzalez-Huitron et al. | Jaccard distance as similarity measure for disparity map estimation | |
| CN115830340B (en) | Point cloud target identification method and device and storage medium | |
| CN111462199A (en) | Rapid speckle image matching method based on GPU | |
| Ma et al. | Improving KAZE feature matching algorithm with alternative image gray method | |
| Wang | Design and Implementation of Artificial Intelligence Image Detection System Based on Plane Features | |
| CN114858094B (en) | Learning type single-frame Moire profile measurement method for dynamic object |
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 | ||
| CB03 | Change of inventor or designer information |
Inventor after: Liu Fujia Inventor after: Yang Jianhong Inventor before: Yang Jianhong Inventor before: Liu Fujia |
|
| CB03 | Change of inventor or designer information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |