CN111754462A - Visual detection method and system for three-dimensional bent pipe - Google Patents
Visual detection method and system for three-dimensional bent pipe Download PDFInfo
- Publication number
- CN111754462A CN111754462A CN202010470623.0A CN202010470623A CN111754462A CN 111754462 A CN111754462 A CN 111754462A CN 202010470623 A CN202010470623 A CN 202010470623A CN 111754462 A CN111754462 A CN 111754462A
- Authority
- CN
- China
- Prior art keywords
- bent pipe
- point cloud
- dimensional
- dimensional reconstruction
- cloud data
- 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
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T15/00—3D [Three Dimensional] image rendering
- G06T15/50—Lighting effects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T19/00—Manipulating 3D models or images for computer graphics
- G06T19/20—Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
Abstract
A visual detection method and a system of a three-dimensional bent pipe are disclosed, wherein a three-dimensional reconstruction module is used for acquiring an image of a bent pipe to be detected and obtaining single bent pipe point cloud data, and a visual positioning module is used for visually positioning the current spatial position and posture of the three-dimensional reconstruction module; fusing single bent pipe point cloud data according to the visual positioning result to form complete bent pipe point cloud data; and identifying the cylindrical section and the circular arc section of the bent pipe according to the complete bent pipe point cloud data, and determining the parameters of the cylindrical section and the circular arc section of the bent pipe. Therefore, the three-dimensional bent pipe workpiece detection device can realize rapid and accurate detection of the three-dimensional bent pipe workpiece, and has the advantages of convenience and rapidness in operation, high flexibility and accurate measurement.
Description
Technical Field
The invention relates to bent pipe detection, in particular to a visual detection method and system for a three-dimensional bent pipe.
Background
The main traditional methods for industrial elbow test include: gauge detection (by a mold method), visual detection based on a two-dimensional image, forming measurement, fork measurement, and the like. In the existing detection methods, the detection method of the detection tool is time-consuming in operation, cannot realize full detection, has large mould volume, is inconvenient to store, has poor result traceability and is difficult to quantify; the visual detection method based on the two-dimensional image has higher requirements on detection scenes, and part of detection equipment has larger volume and cannot adapt to occasions where bent pipes are difficult to disassemble; the efficiency of the forming measurement method is not high, and the fork type measurement mode is only suitable for low-precision measurement occasions; high-precision machinery such as a joint arm and a mechanical arm is used in forming measurement and fork type measurement, the measurement range is limited for measurement of large-size bent pipes, and secondly, mechanical equipment such as the joint arm is not portable enough, so that certain requirements are met for measurement scenes.
Disclosure of Invention
The present invention is directed to overcoming at least one of the above-mentioned technical drawbacks, and providing a method and a system for visually inspecting a three-dimensional bent pipe.
In order to achieve the purpose, the invention adopts the following technical scheme:
a visual inspection method of a three-dimensional bent pipe comprises the following steps:
A. acquiring an image of a bent pipe to be detected through a three-dimensional reconstruction module, and performing three-dimensional reconstruction to obtain single bent pipe point cloud data;
B. performing visual positioning on the current spatial position and posture of the three-dimensional reconstruction module through a visual positioning module;
C. fusing the single bent pipe point cloud data according to the visual positioning result to form complete bent pipe point cloud data;
D. identifying a cylindrical section and an arc section of the bent pipe according to the complete bent pipe point cloud data;
E. and determining parameters of the cylindrical section and the circular arc section of the bent pipe according to the identification results of the cylindrical section and the circular arc section of the bent pipe.
Further:
in step a, the three-dimensional reconstruction module is based on a line structured light principle, and step a specifically includes the following steps:
a1, shooting the structural light stripe projected onto the bent pipe by using the three-dimensional reconstruction module, and performing sub-pixel extraction on the stripe center of the structural light stripe;
and A2, performing three-dimensional reconstruction on the extracted pixel position according to the triangulation method of the line structured light to obtain the single bent pipe point cloud data.
In the step B, the visual positioning module is based on a binocular visual positioning principle.
In the step C, according to the current spatial position and the current attitude of the three-dimensional reconstruction module, the single-width bent pipe point cloud data is translated and rotated to obtain the complete bent pipe point cloud data;
wherein the translation and rotation of the single bent-tube point cloud data uses the following transformation relationship:
Pw＝Rcw·Pc+Tcw(1)
wherein P iswRepresenting point cloud coordinates in the world coordinate system, PcRepresenting the coordinates of the point cloud in the coordinate system of the three-dimensional reconstruction module, RcwRepresenting a rotation matrix from the three-dimensional reconstruction module coordinate system to the world coordinate system, TcwRepresenting a translation matrix from the three-dimensional reconstructed coordinate system to the world coordinate system.
The rotation matrix RcwAnd the translation matrix TcwThree-dimensional coordinate P of coding point on the three-dimensional reconstruction module in the world coordinate systemcodewAnd a coordinate P in the three-dimensional reconstruction module coordinate systemcodecDetermining;
wherein the rotation matrix R is determinedcwAnd the translation matrix TcwThe process comprises the following steps:
and (3) performing decentralized processing on the coordinates of the two groups of points:
wherein
And
are respectively PcodewAnd PcodecThe coordinates of the center point of (a);
p is obtained according to the following formulacodewAnd PcodecDecentralized coordinate matrix
And
performing singular value decomposition on the H matrix to obtain a left singular matrix U, a right singular matrix V and a singular value matrix S;
f＝[1,1,det(V·UT)](8)
m＝diag(f) (9)
Rcw＝V·m·UT(10)
obtaining a rotation matrix R from the self coordinate system of the three-dimensional reconstruction module to the world coordinate system through calculationcw；
Calculating a translation matrix T from the three-dimensional reconstruction module coordinate system to the world coordinate system by a formula (11)cw。
D, according to the complete bent pipe point cloud data, calculating and judging a cylindrical section and an arc section of the bent pipe by using least square fitting; and/or judging the point cloud of the cylindrical section according to the normal vector direction characteristic of the point cloud.
And step E, according to the identification result in the step D, calculating parameters of the cylindrical section and the circular arc section of the bent pipe by using a least square method and an iterative optimization method respectively.
After the step E, the following steps are also included:
according to the parameter calculation result in the step E, aligning the bent pipe with the standard three-dimensional bent pipe model by using an iterative optimization method, wherein according to the requirement of the detection process on the priority of the reference part of the bent pipe, corresponding weights are set aiming at a target equation corresponding to the reference part so as to ensure that the deviation value of the reference part and the corresponding part of the standard three-dimensional bent pipe model after iterative optimization does not exceed the standard;
and according to the aligned data, carrying out one or more of full deviation detection and form and position tolerance detection on the bent pipe.
Before the step A, the method also comprises the following steps:
calibrating a visual positioning module, wherein a calibration object is arranged in a visual positioning view field, global calibration is carried out on internal parameters and external parameters of a camera in the visual positioning module, and the calibration object comprises annular coding points and circular non-coding points;
and calibrating the three-dimensional reconstruction module, wherein a calibration object is arranged in a view field of the three-dimensional reconstruction module, and internal and external parameters of the three-dimensional reconstruction module are calibrated, and the calibration object comprises an annular coding point and a circular non-coding point.
A visual inspection system for three-dimensional bent pipes, comprising:
the three-dimensional reconstruction module is configured to acquire an image of the bent pipe to be detected so as to perform three-dimensional reconstruction to obtain single bent pipe point cloud data;
the visual positioning module is configured to perform visual positioning on the current spatial position and posture of the three-dimensional reconstruction module;
the processing module is configured to fuse the single bent pipe point cloud data according to a visual positioning result to form complete bent pipe point cloud data; identifying a cylindrical section and an arc section of the bent pipe according to the complete bent pipe point cloud data; and determining parameters of the cylindrical section and the circular arc section of the bent pipe according to the identification results of the cylindrical section and the circular arc section of the bent pipe.
The invention has the following beneficial effects:
the invention provides a visual detection method and a visual detection system for a three-dimensional bent pipe, wherein a three-dimensional reconstruction module is used for acquiring an image of the bent pipe to be detected and obtaining single bent pipe point cloud data, and a visual positioning module is used for carrying out visual positioning on the current spatial position and posture of the three-dimensional reconstruction module; fusing single bent pipe point cloud data according to the visual positioning result to form complete bent pipe point cloud data; according to the method, the cylinder section and the arc section of the bent pipe are identified according to the complete bent pipe point cloud data, and the parameters of the cylinder section and the arc section of the bent pipe are determined, so that the method can realize the rapid and accurate detection of the three-dimensional bent pipe workpiece, and has the advantages of convenience and rapidness in operation, high flexibility and accuracy in measurement.
The invention has the following advantages:
(1) according to the invention, the position and the posture of the three-dimensional reconstruction module are measured by a visual positioning technology, so that the three-dimensional reconstruction module can perform mobile measurement at any position in a view field, and the method has the characteristic of high operation flexibility, can reconstruct more complete point cloud data to a great extent, and meets the point cloud acquisition requirement of large-size industrial bent pipes;
(2) according to the invention, the position and the attitude of the three-dimensional reconstruction module are measured by a visual positioning technology, so that the splicing alignment of a single point cloud does not depend on a coded mark point and a non-coded mark point in the measurement process, so that the point pasting operation on the surface of an object is not required, and the non-contact measurement can be realized to the maximum extent;
(3) the method is based on the visual measurement, so the method has the advantages of non-contact, simple and convenient operation and high detection efficiency.
In a word, the method can quickly and accurately detect the three-dimensional bent pipe workpiece, is simple and convenient to operate, can quantify the detection data, has low requirement on the measurement environment, and can be used for (but not limited to) measuring large-size industrial bent pipes.
Drawings
FIG. 1 is a flow chart of a three-dimensional elbow visual inspection method according to an embodiment of the invention;
FIG. 2 is a schematic diagram of point cloud reconstruction by a three-dimensional elbow visual inspection method according to an embodiment of the invention;
FIG. 3 is a schematic diagram of calibration of a three-dimensional reconstruction module according to a three-dimensional elbow visual inspection method of an embodiment of the invention;
FIG. 4 shows the identification and determination results of the cylindrical segment and the circular segment of the elbow according to an embodiment of the present invention;
FIG. 5 shows the calculation results of the parameters of the elbow according to one embodiment of the present invention;
FIG. 6 is a diagram illustrating the effect of elbow weight alignment according to an embodiment of the present invention;
fig. 7 is a diagram illustrating an effect of detecting the shape of the bent pipe according to an embodiment of the invention.
Detailed Description
The embodiments of the present invention will be described in detail below. It should be emphasized that the following description is merely exemplary in nature and is not intended to limit the scope of the invention or its application.
The flow of the three-dimensional elbow visual inspection method according to an embodiment of the invention is shown in fig. 1.
Referring to fig. 1, an embodiment of the present invention provides a visual inspection method for a three-dimensional bent pipe, including the following steps:
step S1, calibrating the visual positioning module: arranging a calibration object in a visual positioning view field, and carrying out global calibration on internal parameters and external parameters of a camera in a positioning module, wherein the calibration object comprises annular coding points and circular non-coding points;
step S2, calibrating the three-dimensional reconstruction module: arranging a calibration object in a view field of a three-dimensional reconstruction module, calibrating internal and external parameters of reconstruction equipment such as a camera in the three-dimensional reconstruction module, wherein the calibration object comprises an annular coding point and a circular non-coding point;
step S3, collecting the point cloud data of the bent pipe workpiece: according to a three-dimensional reconstruction principle, single bent pipe point cloud data are obtained through calculation, and the single bent pipe point cloud data are fused according to a visual positioning result to form complete bent pipe point cloud data;
step S4, identifying the bent pipe cylindrical section: according to the acquired complete bent pipe point cloud data, a cylindrical section and an arc section of the bent pipe are calculated and judged by using least square fitting;
step S5, calculating the parameters of the bent pipe: according to the judgment result in the S4, calculating parameters of the cylindrical section and the circular arc section of the bent pipe by using a least square method and an iterative optimization method respectively;
step S6, elbow weight alignment: aligning the bent pipe with a standard three-dimensional bent pipe model by using an iterative optimization method according to the parameter calculation result in the S5 and the parameter calculation result in the step S5, wherein corresponding weights are set for a target equation corresponding to a reference part according to the requirement of the reference part on the priority of the bent pipe in the detection process so as to ensure that the deviation value of the reference part and the corresponding part of the standard three-dimensional bent pipe model does not exceed the standard after iterative optimization;
step S7, bend pipe deviation detection: and detecting the bent pipe fitting according to the data aligned in the step S6, wherein the detection content includes full deviation detection, form and position tolerance detection and the like.
In a typical embodiment, the specific steps include:
step S1, calibrating the visual positioning module: and arranging a calibration object in the visual positioning view field, and carrying out global calibration on internal parameters and external parameters of the camera in the positioning module, wherein the calibration object comprises annular coding points and circular non-coding points.
Taking binocular vision positioning as an example, accurate calibration of internal and external parameters of the binocular positioning module is realized by arranging a plurality of calibration objects in a measurement view field, the arrangement of the binocular vision positioning module is shown in fig. 2, and the calibration objects comprise annular coding points and circular non-coding points.
In one embodiment, the step S1 includes the following steps:
step S101: measuring three-dimensional space data of all the markers by using a close-range photogrammetry system;
step S102: and calculating to obtain the inside and outside orientation parameters of the plurality of cameras in the visual positioning module according to the collected calibration images by utilizing the photogrammetric space rear intersection principle.
It should be noted that the calibration of the binocular vision positioning module may adopt the prior art, such as the multi-camera calibration method proposed in the document "global calibration of large-field multi-camera video measurement system" (huhao, lianjin, tang orthodox, etc.. optical precision engineering, 2012). Of course, other calibration methods for binocular vision measurement systems may be used.
Step S2, calibrating the three-dimensional reconstruction module: and arranging a calibration object in a view field of the three-dimensional reconstruction module, and calibrating internal and external parameters of reconstruction equipment such as a camera in the three-dimensional reconstruction module.
In a specific embodiment, the precise calibration of the internal and external parameters of the line structured light three-dimensional reconstruction module can be realized by arranging a calibration object in the scanning range, the arrangement of the line structured light three-dimensional reconstruction module is shown in fig. 3, and the calibration object comprises an annular coding point and a circular non-coding point.
Step S3, collecting the point cloud data of the bent pipe workpiece: and calculating to obtain single bent pipe point cloud data according to a three-dimensional reconstruction principle, and fusing the single bent pipe point cloud data according to a visual positioning result to form complete bent pipe point cloud data.
Specifically, the step S3 may include the following steps:
step S301: extracting structural light stripes shot by a three-dimensional reconstruction module according to a reconstruction principle of line structured light, and performing sub-pixel extraction on the centers of the stripes;
step S302: according to the triangulation method of the line structured light, three-dimensional reconstruction is carried out on the pixel position extracted in the S301, and single-frame point cloud data are obtained;
step S303: the method comprises the following steps that when a three-dimensional reconstruction module collects images, a visual positioning module shoots to obtain the space position and the posture of the current three-dimensional reconstruction module through a binocular visual positioning principle;
step S304: according to the spatial position and the posture of the three-dimensional reconstruction module obtained in the step S303, the single point cloud data obtained in the step S302 is translated and rotated to finally obtain a complete point cloud;
the conversion relation from the single point cloud coordinate to the world coordinate system is as follows:
Pw＝Rcw·Pc+Tcw(1)
wherein R iswRepresenting point cloud coordinates in the world coordinate system, PcRepresenting the coordinates of the point cloud in the coordinate system of the three-dimensional reconstruction module, RcwRepresenting a rotation matrix from the three-dimensional reconstruction module coordinate system to the world coordinate system, TcwRepresenting a translation matrix from the three-dimensional reconstructed coordinate system to the world coordinate system.
In a specific embodiment, RcwAnd TcwCan be represented by three-dimensional coordinates P of coding points on the three-dimensional reconstruction module in a world coordinate systemcodewAnd coordinates P in its own coordinate systemcodecThe solution is performed, in this embodiment, R is solved by using singular value decompositioncwRotating the matrix to obtain TcwAnd (3) translating the matrix, and calculating as follows:
firstly, two groups of point coordinates are processed by decentralization:
wherein
And
are respectively PcodewAnd PcodecCentral point coordinates of (2), for convenience of representation, the following
And
each represents a number of rows N, each row being
And
a matrix of N rows and 3 columns;
here, the
And
respectively represent PcodewAnd PcodecRemoving the centralized coordinate matrix;
[U,S,V]＝svd(H) (7
performing singular value decomposition on the H matrix to obtain a left singular matrix U, a right singular matrix V and a singular value matrix S;
f＝[1,1,det(V·UT)](8)
m＝diag(f) (9)
Rcw＝V·m·UT(10)
obtaining a rotation matrix R from the self coordinate system of the three-dimensional reconstruction module to the world coordinate system through calculationcw；
Here, the translation matrix T from the three-dimensional reconstruction module's own coordinate system to the world coordinate system can be calculated by the formula (11)cw。
Step S4, identifying the bent pipe cylindrical section: and according to the acquired complete bent pipe point cloud data, calculating and judging the cylindrical section and the circular arc section of the bent pipe by using least square fitting.
Specifically, in an embodiment, when the cylindrical segment and the circular arc segment of the bent pipe are determined, for step S4, the point cloud of the cylindrical segment may be effectively determined according to the normal vector direction characteristic of the point cloud, so as to provide a good initial value for the calculation of the bent pipe parameters, and the bent pipe cylinder identification effect in the specific embodiment is as shown in fig. 4.
Step S5, calculating the parameters of the bent pipe: according to the judgment result in the step S4, the parameters of the cylindrical section and the circular arc section of the bent pipe are calculated by using a least square method and an iterative optimization method, and the calculation result of the parameters of the bent pipe in the specific embodiment is shown in fig. 5.
Step S6, elbow weight alignment: and aligning the bent pipe with the standard three-dimensional bent pipe model by using an iterative optimization method according to the parameter calculation result in the step S5, wherein according to the requirement of the detection process on the priority of the reference part of the bent pipe, a corresponding weight is set for a target equation corresponding to the reference part so as to ensure that the deviation value of the reference part and the corresponding part of the standard three-dimensional bent pipe model after iterative optimization does not exceed the standard, and the deviation value is usually close to zero.
Specifically, in an embodiment, a user has a higher requirement on the alignment accuracy of the end point, the inflection point, the cylindrical axis, and the like of the bent pipe workpiece, and a higher weight may be set for these portions to obtain a corresponding effect, and the alignment effect of this embodiment is shown in fig. 6.
Step S7, bend pipe deviation detection: and detecting the bent pipe fitting according to the data aligned in the step S6, wherein the detection contents comprise full deviation detection, form and position tolerance detection and the like of the appearance of the bent pipe fitting.
Specifically, the visualization effect of the full detection of the shape of the bent pipe in this embodiment is shown in fig. 7.
In addition, the embodiment of the invention also provides a three-dimensional bent pipe visual detection system, and the system is used for implementing the detection method. The three-dimensional bent pipe visual detection system comprises a three-dimensional reconstruction module, a visual positioning module and a processing module, wherein the three-dimensional reconstruction module acquires an image of a bent pipe to be detected so as to carry out three-dimensional reconstruction to obtain single bent pipe point cloud data; the visual positioning module carries out visual positioning on the current space position and the current space posture of the three-dimensional reconstruction module; the processing module fuses the single-width bent pipe point cloud data according to the visual positioning result to form complete bent pipe point cloud data; identifying a cylindrical section and an arc section of the bent pipe according to the complete bent pipe point cloud data; and determining parameters of the cylindrical section and the circular arc section of the bent pipe according to the identification results of the cylindrical section and the circular arc section of the bent pipe.
The background of the present invention may contain background information related to the problem or environment of the present invention and does not necessarily describe the prior art. Accordingly, the inclusion in the background section is not an admission of prior art by the applicant.
The foregoing is a more detailed description of the invention in connection with specific/preferred embodiments and is not intended to limit the practice of the invention to those descriptions. It will be apparent to those skilled in the art that various substitutions and modifications can be made to the described embodiments without departing from the spirit of the invention, and these substitutions and modifications should be considered to fall within the scope of the invention. In the description herein, references to the description of the term "one embodiment," "some embodiments," "preferred embodiments," "an example," "a specific example," or "some examples" or the like are intended to mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Various embodiments or examples and features of various embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction. Although embodiments of the present invention and their advantages have been described in detail, it should be understood that various changes, substitutions and alterations can be made herein without departing from the scope of the claims.
Claims (10)
1. A visual inspection method of a three-dimensional bent pipe is characterized by comprising the following steps:
A. acquiring an image of a bent pipe to be detected through a three-dimensional reconstruction module, and performing three-dimensional reconstruction to obtain single bent pipe point cloud data;
B. performing visual positioning on the current spatial position and posture of the three-dimensional reconstruction module through a visual positioning module;
C. fusing the single bent pipe point cloud data according to the visual positioning result to form complete bent pipe point cloud data;
D. identifying a cylindrical section and an arc section of the bent pipe according to the complete bent pipe point cloud data;
E. and determining parameters of the cylindrical section and the circular arc section of the bent pipe according to the identification results of the cylindrical section and the circular arc section of the bent pipe.
2. The visual inspection method of a three-dimensional bent pipe according to claim 1, wherein in step a, the three-dimensional reconstruction module is a three-dimensional reconstruction module based on a line structured light principle, and step a specifically includes the following steps:
a1, shooting the structural light stripe projected onto the bent pipe by using the three-dimensional reconstruction module, and performing sub-pixel extraction on the stripe center of the structural light stripe;
and A2, performing three-dimensional reconstruction on the extracted pixel position according to the triangulation method of the line structured light to obtain the single bent pipe point cloud data.
3. The visual inspection method of the three-dimensional bent pipe according to claim 1 or 2, wherein in the step B, the visual positioning module is a visual positioning module based on a binocular visual positioning principle.
4. The visual inspection method of a three-dimensional bent pipe according to any one of claims 1 to 3, wherein in step C, the single bent pipe point cloud data is translated and rotated according to the current spatial position and posture of the three-dimensional reconstruction module to obtain the complete bent pipe point cloud data;
wherein the translation and rotation of the single bent-tube point cloud data uses the following transformation relationship:
Pw＝Rcw·Pc+Tcw(1)
wherein P iswRepresenting point cloud coordinates in the world coordinate system, PvRepresenting the coordinates of the point cloud in the coordinate system of the three-dimensional reconstruction module, RvwRepresenting a rotation matrix from the three-dimensional reconstruction module coordinate system to the world coordinate system, TcwRepresenting a translation matrix from the three-dimensional reconstructed coordinate system to the world coordinate system.
5. The method according to claim 4, wherein the rotation matrix R is a matrix of rotationcwAnd the translation matrix TcwThree-dimensional coordinate P of coding point on the three-dimensional reconstruction module in the world coordinate systemcodewAnd a coordinate P in the three-dimensional reconstruction module coordinate systemcodecDetermining;
wherein the rotation matrix R is determinedcwAnd the translation matrix TcwThe process comprises the following steps:
and (3) performing decentralized processing on the coordinates of the two groups of points:
wherein
And
are respectively PcodewAnd PcodecThe coordinates of the center point of (a);
p is obtained according to the following formulacodewAnd PcodecDecentralized coordinate matrix
And
[U,S,V]＝svd(H) (7)
performing singular value decomposition on the H matrix to obtain a left singular matrix U, a right singular matrix V and a singular value matrix S;
f＝[1,1,det(V·UT)](8)
m＝diag(f) (9)
Rcw＝V·m·UT(10)
obtaining a rotation matrix R from the self coordinate system of the three-dimensional reconstruction module to the world coordinate system through calculationcw；
Calculating a translation matrix T from the three-dimensional reconstruction module coordinate system to the world coordinate system by a formula (11)cw。
6. The visual inspection method of a three-dimensional bent pipe according to any one of claims 1 to 5, wherein in step D, the cylindrical section and the circular arc section of the bent pipe are calculated and judged by using least square method according to the complete bent pipe point cloud data; and/or judging the point cloud of the cylindrical section according to the normal vector direction characteristic of the point cloud.
7. The method according to any one of claims 1 to 6, wherein in step E, the parameters of the cylindrical segment and the circular arc segment of the bent pipe are calculated by using a least square method and an iterative optimization method according to the identification result in step D.
8. The method according to any one of claims 1 to 6, further comprising the following steps after step E:
according to the parameter calculation result in the step E, aligning the bent pipe with the standard three-dimensional bent pipe model by using an iterative optimization method, wherein according to the requirement of the detection process on the priority of the reference part of the bent pipe, corresponding weights are set aiming at a target equation corresponding to the reference part so as to ensure that the deviation value of the reference part and the corresponding part of the standard three-dimensional bent pipe model after iterative optimization does not exceed the standard;
and according to the aligned data, carrying out one or more of full deviation detection and form and position tolerance detection on the bent pipe.
9. The method according to any one of claims 1 to 6, further comprising, before step A, the steps of:
calibrating a visual positioning module, wherein a calibration object is arranged in a visual positioning view field, global calibration is carried out on internal parameters and external parameters of a camera in the visual positioning module, and the calibration object comprises annular coding points and circular non-coding points;
and calibrating the three-dimensional reconstruction module, wherein a calibration object is arranged in a view field of the three-dimensional reconstruction module, and internal and external parameters of the three-dimensional reconstruction module are calibrated, and the calibration object comprises an annular coding point and a circular non-coding point.
10. A visual inspection system for three-dimensional bent pipes, comprising:
the three-dimensional reconstruction module is configured to acquire an image of the bent pipe to be detected so as to perform three-dimensional reconstruction to obtain single bent pipe point cloud data;
the visual positioning module is configured to perform visual positioning on the current spatial position and posture of the three-dimensional reconstruction module;
the processing module is configured to fuse the single bent pipe point cloud data according to a visual positioning result to form complete bent pipe point cloud data; identifying a cylindrical section and an arc section of the bent pipe according to the complete bent pipe point cloud data; and determining parameters of the cylindrical section and the circular arc section of the bent pipe according to the identification results of the cylindrical section and the circular arc section of the bent pipe.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010470623.0A CN111754462A (en) | 2020-05-28 | 2020-05-28 | Visual detection method and system for three-dimensional bent pipe |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010470623.0A CN111754462A (en) | 2020-05-28 | 2020-05-28 | Visual detection method and system for three-dimensional bent pipe |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111754462A true CN111754462A (en) | 2020-10-09 |
Family
ID=72673647
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010470623.0A Pending CN111754462A (en) | 2020-05-28 | 2020-05-28 | Visual detection method and system for three-dimensional bent pipe |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111754462A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112381847A (en) * | 2020-10-27 | 2021-02-19 | 新拓三维技术(深圳)有限公司 | Pipeline end head space pose measuring method and system |
| CN112487576A (en) * | 2020-11-26 | 2021-03-12 | 新拓三维技术(深圳)有限公司 | Pipeline reverse modeling method |
| CN112648934A (en) * | 2020-12-07 | 2021-04-13 | 新拓三维技术(深圳)有限公司 | Automatic elbow geometric form detection method |
| CN113066335A (en) * | 2021-04-01 | 2021-07-02 | 中核核电运行管理有限公司 | Bent pipe operation guidance system |
| CN113118266A (en) * | 2021-04-13 | 2021-07-16 | 江阴市宏业机械制造有限公司 | Numerical control pipe bender capable of dynamically correcting hole positions through visual detection and pipe bending method |
| CN113205086A (en) * | 2021-07-05 | 2021-08-03 | 武汉瀚迈科技有限公司 | Feature parameter identification algorithm for circular-section bent pipe parts based on ellipse fitting |
| CN117525763A (en) * | 2023-11-08 | 2024-02-06 | 金寨国轩新能源有限公司 | Layered pole lug of soft-package battery core and bending detection method thereof |
-
2020
- 2020-05-28 CN CN202010470623.0A patent/CN111754462A/en active Pending
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112381847A (en) * | 2020-10-27 | 2021-02-19 | 新拓三维技术(深圳)有限公司 | Pipeline end head space pose measuring method and system |
| CN112381847B (en) * | 2020-10-27 | 2024-02-13 | 新拓三维技术(深圳)有限公司 | Pipeline end space pose measurement method and system |
| CN112487576A (en) * | 2020-11-26 | 2021-03-12 | 新拓三维技术(深圳)有限公司 | Pipeline reverse modeling method |
| CN112487576B (en) * | 2020-11-26 | 2024-04-02 | 新拓三维技术(深圳)有限公司 | Pipeline reverse modeling method |
| CN112648934A (en) * | 2020-12-07 | 2021-04-13 | 新拓三维技术(深圳)有限公司 | Automatic elbow geometric form detection method |
| CN112648934B (en) * | 2020-12-07 | 2022-07-01 | 新拓三维技术(深圳)有限公司 | Automatic elbow geometric form detection method |
| CN113066335A (en) * | 2021-04-01 | 2021-07-02 | 中核核电运行管理有限公司 | Bent pipe operation guidance system |
| CN113118266A (en) * | 2021-04-13 | 2021-07-16 | 江阴市宏业机械制造有限公司 | Numerical control pipe bender capable of dynamically correcting hole positions through visual detection and pipe bending method |
| CN113118266B (en) * | 2021-04-13 | 2023-03-07 | 江阴市宏业机械制造有限公司 | Numerical control pipe bender capable of dynamically correcting hole positions through visual detection and pipe bending method |
| CN113205086A (en) * | 2021-07-05 | 2021-08-03 | 武汉瀚迈科技有限公司 | Feature parameter identification algorithm for circular-section bent pipe parts based on ellipse fitting |
| CN117525763A (en) * | 2023-11-08 | 2024-02-06 | 金寨国轩新能源有限公司 | Layered pole lug of soft-package battery core and bending detection method thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111754462A (en) | Visual detection method and system for three-dimensional bent pipe | |
| CN109029257B (en) | Large-scale workpiece pose measurement system and method based on stereoscopic vision and structured light vision | |
| CN105046715B (en) | A kind of line-scan digital camera scaling method based on interspace analytic geometry | |
| CN105115560B (en) | A kind of non-contact measurement method of cabin volume of compartment | |
| CN110672020A (en) | Stand tree height measuring method based on monocular vision | |
| CN109579695B (en) | Part measuring method based on heterogeneous stereoscopic vision | |
| CN108648242B (en) | Two-camera calibration method and device without public view field based on assistance of laser range finder | |
| CN103759670A (en) | Object three-dimensional information acquisition method based on digital close range photography | |
| CN103278138A (en) | Method for measuring three-dimensional position and posture of thin component with complex structure | |
| CN111220126A (en) | Space object pose measurement method based on point features and monocular camera | |
| CN112184830B (en) | Camera internal parameter and external parameter calibration method and device, computer equipment and storage medium | |
| Toschi et al. | On the evaluation of photogrammetric methods for dense 3D surface reconstruction in a metrological context | |
| CN109544642B (en) | N-type target-based TDI-CCD camera parameter calibration method | |
| Feng et al. | Inertial measurement unit aided extrinsic parameters calibration for stereo vision systems | |
| Zhang et al. | A novel 3D optical method for measuring and evaluating springback in sheet metal forming process | |
| CN112013830A (en) | Accurate positioning method for unmanned aerial vehicle inspection image detection defects of power transmission line | |
| CN108871190B (en) | Handheld spherical target for binocular stereo vision measurement and measurement method | |
| CN116051659B (en) | Linear array camera and 2D laser scanner combined calibration method | |
| Jiang et al. | Combined shape measurement based on locating and tracking of an optical scanner | |
| CN114459345A (en) | System and method for detecting position and attitude of airplane body based on visual space positioning | |
| CN112508885B (en) | Method and system for detecting three-dimensional central axis of bent pipe | |
| CN112697044B (en) | Static rigid object vision measurement method based on unmanned aerial vehicle platform | |
| Fu et al. | A flexible approach to light pen calibration for a monocular-vision-based coordinate measuring system | |
| CN112734842B (en) | Auxiliary positioning method and system for centering installation of large ship equipment | |
| Jing et al. | The technologies of close-range photogrammetry and application in manufacture |
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 |