CN115876788A - 3D visual online detection jitter interference and elimination method for marine steel plate - Google Patents
3D visual online detection jitter interference and elimination method for marine steel plate Download PDFInfo
- Publication number
- CN115876788A CN115876788A CN202211445700.2A CN202211445700A CN115876788A CN 115876788 A CN115876788 A CN 115876788A CN 202211445700 A CN202211445700 A CN 202211445700A CN 115876788 A CN115876788 A CN 115876788A
- Authority
- CN
- China
- Prior art keywords
- steel plate
- point cloud
- point
- coordinates
- detected
- 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
Landscapes
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses a 3D visual online detection and jitter interference elimination method for a marine steel plate. The method solves the problem that small-scale surface discontinuity defects cannot be found from the collected 3D point cloud data due to steel plate shaking interference when the steel plate surface discontinuity on-line detection is carried out by using a 3D machine vision technology, and improves the accuracy of steel plate surface defect detection.
Description
Technical Field
The invention belongs to the technical field of ship construction, and particularly relates to a 3D visual online detection jitter interference and elimination method for a marine steel plate.
Background
The shipyard is that the steel sheet makes big household, and the specification of steel sheet is many, in large quantity, and the steel sheet of different varieties, different specifications generally takes open stack to deposit, and the storage time is long, very easily causes the steel sheet quality problem, shows that degree, quantity and the incidence of pit defect in steel sheet surface obviously increase, need polish or repair welding after just can use, consequently need carry out comprehensive detection to marine steel sheet surface, avoids surface defect to flow into follow-up link and causes great loss.
At present, the surface detection of the marine steel plate mainly depends on manual means, the defects of large workload, low efficiency and low accuracy rate exist in manual visual inspection, and the detection quality of the steel plate cannot be ensured. In addition, the main direction of the discontinuous online automatic detection of the surface of the steel plate at present is to use 2D image acquisition equipment, highlight the discontinuous characteristics of the surface of the steel plate through special illumination, manually perform defect classification and marking on the acquired images, perform model training by adopting a deep learning technology, and finally perform reasoning calculation on the images acquired on an actual line by a system to realize the identification and classification of defects. The main problems of such 2D machine vision in inspection detection are:
1. the color difference between the surface defect of the steel plate and the surface color of a normal steel plate is small, and the problem of collection leakage exists;
2.2D machine vision on-line detection technology is greatly influenced by ambient light, a complex shading device is needed for natural light shielding, the system complexity is increased, and the normal working flow of a production line is influenced;
3. the deep learning algorithm model based on the 2D image needs a lot of training to achieve available accuracy, the workload is large, and the skilled time is uncontrollable;
4. the result of the 2D machine vision detection only has defect type information, and information such as depth, area, position and the like of the defect cannot be provided, so that the grade of the steel plate cannot be accurately evaluated actually, and the repair work of the subsequent process is difficult to guide.
On the other hand, although the existing defect detection technology utilizes a 3D machine vision technology to solve the detection problem of the 2D machine vision technology, when the marine steel plate moves on the roller bed, unpredictable and uneven shaking phenomena can be generated due to interaction between the marine steel plate and the roller wheel, and the shaking range of the steel plates with different thicknesses and sizes in the normal direction of the conveying surface of the roller bed is between 0.2mm and 0.4 mm; the requirement of the detection precision on the depth of the discontinuous defect on the surface of the marine steel plate is 0.1mm at the present stage, namely the detection system is required to be capable of distinguishing the depth difference of 0.1mm, and the requirement of high-precision detection of the steel plate cannot be met by using a 3D machine vision technology to perform online measurement under the condition that the steel plate shakes more than 0.2 mm.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provides a 3D visual online detection jitter interference and elimination method for a ship steel plate, and the method solves the problem that small-scale surface discontinuity defects cannot be found in acquired 3D point cloud data due to steel plate jitter interference when the surface discontinuity online detection of the steel plate is carried out by using a 3D machine visual technology, and improves the accuracy of steel plate surface defect detection.
In order to achieve the purpose of the invention, the invention provides the following technical scheme:
A3D visual online detection jitter interference and elimination method for a marine steel plate specifically comprises the following steps:
the method comprises the following steps that firstly, 3D image acquisition sensors are respectively arranged above and below a roller bed, and a steel plate to be detected is placed on the roller bed and is driven to move by the roller bed;
step two, arranging a synchronous encoder above a roller bed to measure the running speed of a steel plate to be detected on the roller bed, and simultaneously sending synchronous image instructions to 3D image acquisition sensors above and below the roller bed by the synchronous encoder;
thirdly, after the 3D image acquisition sensors above and below the roller bed receive the synchronous image instruction, the 3D image acquisition sensors above and below the roller bed simultaneously acquire 3D point cloud images of the upper surface and the lower surface of the steel plate to be detected;
fourthly, acquiring point cloud data sets of the upper surface and the lower surface of the steel plate to be detected from the 3D point cloud image in the third step, and performing low-pass filtering on the acquired point cloud data sets to acquire a standard point cloud sequence;
fifthly, selecting a reference point from the standard point cloud sequence obtained in the fourth step, and calculating an amplitude value through the standard point cloud sequence and the reference point;
sixthly, after the amplitude value is calculated, calculating the amplitude difference value of the corresponding standard point cloud sequence through the amplitude value;
seventhly, after the amplitude difference value is calculated, calculating a static point cloud sequence U (y) of the steel plate corresponding to the point cloud sequence in the fourth step under the static state of the roller bed n ，u n )，D(y n ，d n )；
And eighthly, carrying out subsequent steel plate surface quality detection after the calculation of the static point cloud sequence is finished.
The 3D image acquisition sensors respectively arranged above and below the roller bed are positioned in the same vertical plane, and the synchronous encoder enables the 3D image acquisition sensors respectively arranged above and below the roller bed to synchronously acquire the 3D point cloud images of the upper surface and the lower surface of the steel plate of the same steel plate section.
The point cloud data set in the fourth step is specifically P { P 0 ，P 1 ，...P N }，P′{P′ n ，P′ 1 ，...P′ N P is the coordinate of the upper surface point of the same steel plate section, P' is the coordinate of the lower surface point of the same steel plate section, P n (y n ，z n )，P′ n (y n ，z′ n ) Coordinates of an upper surface point in the same steel plate section and coordinates of a lower surface point in the same steel plate section corresponding to the coordinates of the upper surface point.
The specific steps of the point cloud data set for low-pass filtering are as follows: and comparing the difference value of the Z coordinates of the corresponding upper surface point coordinates and the lower surface point coordinates in the same steel plate section in the acquired point cloud data set with the thickness of the steel plate to be detected, and when the difference value of the Z coordinates of the corresponding upper surface point coordinates and the corresponding lower surface point coordinates in the same steel plate section and the thickness of the steel plate to be detected is in the range of 0.2-0.4 mm, taking the corresponding upper surface point coordinates and the corresponding lower surface point coordinates in the same steel plate section as a standard point cloud sequence.
The reference point in the fourth step is specifically selected by respectively selecting a reference point from the coordinates of the upper surface point of the same section of the steel plate to be detected and the coordinates of the lower surface point of the same section, and the Y coordinates of the two reference points are the same; the reference points of the same section of the steel plate to be detected are fixed.
The fifth step of calculating the amplitude value through the standard point cloud sequence and the reference point comprises the following specific steps: and subtracting the Z coordinates of all point coordinates in the standard point cloud sequence from the Z coordinates of the reference point coordinates to obtain the amplitude values of all point coordinates in the standard point cloud sequence.
The specific steps of calculating the amplitude difference value of the corresponding standard point cloud sequence through the amplitude value in the sixth step are as follows: and subtracting the amplitude value of the reference point and the lower surface coordinate which is the same as the Y coordinate of the upper surface point coordinate in the standard point cloud sequence from the amplitude value of the reference point and the upper surface point coordinate in the standard point cloud sequence to obtain the amplitude difference value corresponding to the upper surface and the lower surface in the same section of the steel plate.
The static point cloud sequence U (y) in the seventh step n ，u n )，D(y n ，d n ) The specific formula of the calculation is as follows:
the method comprises the following steps of A, acquiring a coordinate of the upper surface of the same section of a steel plate to be detected in a static state, and acquiring a coordinate of the lower surface of the same section of the steel plate to be detected in a static state; delta is a set threshold valueThe value is the detection precision value Z of the depth of the discontinuous defect on the surface of the steel plate to be detected n -Z 0 Is the amplitude value, Z, of the upper surface of the same section of the steel plate to be detected 0 Is reference point, Z' n -Z′ 0 Is the lower surface amplitude value, Z 'of the same section of the steel plate to be detected' 0 Is a reference point; (Z) n -Z 0 )-(Z′ n -Z′ 0 ) Is the amplitude difference.
Based on the technical scheme, the 3D visual online detection jitter interference and elimination method for the marine steel plate obtains the following technical advantages through practical application:
1. the method for 3D visual online detection and elimination of the jitter interference of the marine steel plate solves the problem that small-scale surface discontinuity defects cannot be found from acquired 3D point cloud data due to the steel plate jitter interference when the 3D machine vision technology is used for carrying out discontinuous online detection on the surface of the steel plate, and improves the accuracy of detection of the steel plate surface defects.
2. According to the 3D visual online detection jitter interference and elimination method for the marine steel plate, disclosed by the invention, by utilizing a 3D machine visual technology, the influence of ambient light is small, the marine steel plate can normally work in an indoor workshop environment, and the applicability is improved.
Drawings
FIG. 1 is a diagram of a point cloud data set acquisition structure in a 3D visual online detection jitter interference and elimination method for a marine steel plate according to the present invention.
Detailed Description
In order that the objects, aspects and advantages of the present invention will become more apparent, the invention will be described by way of specific examples shown in the accompanying drawings. It is to be understood that such description is merely illustrative and not intended to limit the scope of the present invention. Moreover, in the following description, descriptions of well-known structures and techniques are omitted so as to not unnecessarily obscure the concepts of the present invention.
As shown in fig. 1, the invention belongs to a 3D visual online detection jitter interference and elimination method for a marine steel plate, which specifically comprises the following steps:
firstly, respectively arranging 3D image acquisition sensors 1 above and below a roller bed, and placing a steel plate 3 to be detected on the roller bed and driving the steel plate to move through the roller bed;
step two, arranging a synchronous encoder 2 above a roller bed to measure the running speed of a steel plate 3 to be detected on the roller bed, and simultaneously sending synchronous image instructions to 3D image acquisition sensors 1 above and below the roller bed by the synchronous encoder 2;
thirdly, after the 3D image acquisition sensors 1 above and below the roller bed receive the synchronous image instruction, the 3D image acquisition sensors 1 above and below the roller bed simultaneously acquire 3D point cloud images of the upper surface and the lower surface of the steel plate to be detected;
fourthly, acquiring point cloud data sets of the upper surface and the lower surface of the steel plate 3 to be detected from the 3D point cloud image in the third step, and performing low-pass filtering on the acquired point cloud data sets to acquire a standard point cloud sequence;
fifthly, selecting a reference point from the standard point cloud sequence obtained in the fourth step, and calculating an amplitude value through the standard point cloud sequence and the reference point;
sixthly, after the amplitude value is calculated, calculating the amplitude difference value of the corresponding standard point cloud sequence through the amplitude value;
seventhly, after the amplitude difference value is calculated, calculating a static point cloud sequence U (y) of the steel plate corresponding to the point cloud sequence in the fourth step under the static state of the roller bed n ，u n )，D(y n ，d n )；
And eighthly, carrying out subsequent steel plate surface quality detection after the static point cloud sequence calculation is finished.
The 3D image acquisition sensors 1 respectively arranged above and below the roller bed are positioned in the same vertical plane, and the synchronous encoder 2 enables the 3D image acquisition sensors 1 respectively arranged above and below the roller bed to synchronously acquire 3D point cloud images of the upper surface and the lower surface of the steel plate of the same steel plate section.
The point cloud data set in the fourth step is specifically P { P 0 ，P 1 ，...P N }，P{P′ 0 ,P′ 1 ，...,P′ N Wherein P isThe coordinates of the upper surface points of the same steel plate section, P' is the coordinates of the lower surface points of the same steel plate section, P n (y n ，z n )，P′ n (y n ，z′ n ) Coordinates of an upper surface point in the same steel plate section and coordinates of a lower surface point in the same steel plate section corresponding to the coordinates of the upper surface point.
The specific steps of the point cloud data set for low-pass filtering are as follows: and comparing the difference value of the Z coordinates of the corresponding upper surface point coordinates and the lower surface point coordinates in the same steel plate section in the acquired point cloud data set with the thickness of the steel plate to be detected, and when the difference value of the Z coordinates of the corresponding upper surface point coordinates and the corresponding lower surface point coordinates in the same steel plate section and the thickness of the steel plate to be detected is in the range of 0.2-0.4 mm, taking the corresponding upper surface point coordinates and the corresponding lower surface point coordinates in the same steel plate section as a standard point cloud sequence.
The reference point in the fourth step is specifically selected by respectively selecting a reference point from the coordinates of the upper surface point of the same section and the coordinates of the lower surface point of the same section of the steel plate 3 to be detected, and the Y coordinates of the two reference points are the same; the reference point of the same section of the steel plate 3 to be detected is fixed.
The fifth step of calculating the amplitude value through the standard point cloud sequence and the reference point comprises the following specific steps: and (3) subtracting the Z coordinates of all the point coordinates in the standard point cloud sequence from the Z coordinates of the reference point coordinates to obtain the amplitude values of all the point coordinates in the standard point cloud sequence.
The specific steps of calculating the amplitude difference value of the corresponding standard point cloud sequence through the amplitude value in the sixth step are as follows: and subtracting the amplitude value of the reference point and the lower surface coordinate which is the same as the Y coordinate of the upper surface point coordinate in the standard point cloud sequence from the amplitude value of the reference point and the upper surface point coordinate in the standard point cloud sequence to obtain the amplitude difference value corresponding to the upper surface and the lower surface in the same section of the steel plate.
The static point cloud sequence U (y) in the seventh step n ，u n )，Dn(y n ，d n ) The specific formula of the calculation is as follows:
wherein, U is the coordinate of the upper surface of the same section under the static state of the steel plate 3 to be detected, and D is the coordinate of the lower surface of the same section under the static state of the steel plate 3 to be detected; delta is a set threshold value, the threshold value is a detection precision value of the surface discontinuous defect depth of the steel plate to be detected, Z n -Z 0 The amplitude value Z of the upper surface of the same section of the steel plate 3 to be detected 0 Is reference point, Z' n -Z′ 0 Is the lower surface amplitude value Z 'of the same section of the steel plate 3 to be detected' 0 Is a reference point; (Z) n -Z 0 )-(Z′ n -Z′ 0 ) Is the amplitude difference; the problem that small-scale surface discontinuity defects cannot be found from collected 3D point cloud data due to steel plate shaking interference when the steel plate surface discontinuity on-line detection is carried out by using a 3D machine vision technology is solved, and the detection accuracy of the steel plate surface defects is improved.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention and not to limit it; although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that; modifications to the specific embodiments of the invention or equivalent substitutions for parts of the technical features may be made; without departing from the spirit of the present invention, it is intended to cover all aspects of the invention as defined by the appended claims.
Claims (8)
1. A3D visual online detection jitter interference and elimination method for a marine steel plate is characterized by comprising the following steps:
the method comprises the following steps that firstly, 3D image acquisition sensors (1) are respectively arranged above and below a roller bed, and a steel plate (3) to be detected is placed on the roller bed and is driven to move through the roller bed;
step two, arranging a synchronous encoder (2) above a roller bed to measure the running speed of a steel plate (3) to be detected on the roller bed, and simultaneously sending synchronous image instructions to 3D image acquisition sensors (1) above and below the roller bed by the synchronous encoder (2);
thirdly, after the 3D image acquisition sensors (1) above and below the roller bed receive the synchronous image instruction, the 3D image acquisition sensors (1) above and below the roller bed simultaneously acquire 3D point cloud images of the upper surface and the lower surface of the steel plate to be detected;
fourthly, acquiring point cloud data sets of the upper surface and the lower surface of the steel plate (3) to be detected from the 3D point cloud image in the third step, and performing low-pass filtering on the acquired point cloud data sets to acquire a standard point cloud sequence;
fifthly, selecting a reference point from the standard point cloud sequence obtained in the fourth step, and calculating an amplitude value through the standard point cloud sequence and the reference point;
sixthly, after the amplitude value is calculated, calculating the amplitude difference value of the corresponding standard point cloud sequence through the amplitude value;
seventhly, after the amplitude difference value is calculated, calculating a static point cloud sequence U (y) of the steel plate corresponding to the point cloud sequence in the fourth step in a static state of the roller bed n ，u n )，D(y n ，d n )；
And eighthly, carrying out subsequent steel plate surface quality detection after the static point cloud sequence calculation is finished.
2. The method for 3D visual online detection and elimination of jitter interference of marine steel plate according to claim 1, characterized in that the 3D image capturing sensors (1) respectively arranged above and below the roller bed are in the same vertical plane, and the synchronous encoder (2) enables the 3D image capturing sensors (1) respectively arranged above and below the roller bed to synchronously acquire the 3D point cloud images of the upper surface and the lower surface of the steel plate with the same steel plate section.
3. The method for 3D visual online detection of jitter interference and elimination of marine steel plate according to claim 1, wherein the point cloud data set in the fourth stepSpecific formula is P { P 0 ，P 1 ，...P N }，P′{P′ 0 ，P′ 1 ，...P′ N P is the coordinate of the upper surface point of the same steel plate section, P' is the coordinate of the lower surface point of the same steel plate section, P n (y n ，z n )，P′ n (y n ，z′ n ) Coordinates of an upper surface point in the same steel plate section and coordinates of a lower surface point in the same steel plate section corresponding to the coordinates of the upper surface point.
4. The method for 3D visual online detection and elimination of jitter interference of marine steel plate according to claim 3, wherein the specific steps of performing low-pass filtering on the point cloud data set are as follows: and comparing the difference value of the Z coordinates of the corresponding upper surface point coordinates and the lower surface point coordinates in the same steel plate section in the acquired point cloud data set with the thickness of the steel plate to be detected, and when the difference value of the Z coordinates of the corresponding upper surface point coordinates and the corresponding lower surface point coordinates in the same steel plate section and the thickness of the steel plate to be detected is in the range of 0.2-0.4 mm, taking the corresponding upper surface point coordinates and the corresponding lower surface point coordinates in the same steel plate section as a standard point cloud sequence.
5. The 3D visual online detection jitter interference and elimination method for the marine steel plate according to claim 1, wherein the selection of the reference points in the fourth step is specifically to select a reference point from the coordinates of the upper surface point of the same section of the steel plate (3) to be detected and the coordinates of the lower surface point of the same section, and the Y coordinates of the two reference points are the same; the reference point of the same section of the steel plate (3) to be detected is fixed.
6. The method for 3D visual online detection and elimination of jitter interference of marine steel plate according to claim 1, wherein the step of calculating the amplitude value through the standard point cloud sequence and the reference point comprises the following specific steps: and subtracting the Z coordinates of all point coordinates in the standard point cloud sequence from the Z coordinates of the reference point coordinates to obtain the amplitude values of all point coordinates in the standard point cloud sequence.
7. The method for 3D visual online detection and elimination of jitter interference of marine steel plate according to claim 1, wherein the specific steps of calculating the amplitude difference of the corresponding standard point cloud sequence through the amplitude value in the sixth step are as follows: and subtracting the amplitude value of the reference point and the lower surface coordinate which is the same as the Y coordinate of the upper surface point coordinate in the standard point cloud sequence from the amplitude value of the reference point and the upper surface point coordinate in the standard point cloud sequence to obtain the amplitude difference value corresponding to the upper surface and the lower surface in the same section of the steel plate.
8. The method for 3D visual online detection of jitter interference and elimination of marine steel plate according to claim 1, wherein the stationary point cloud sequence U (y) in the seventh step n ，u n )，D(y n ，d n ) The specific formula of the calculation is as follows:
wherein U is the coordinate of the upper surface of the same section under the static state of the steel plate (3) to be detected, and D is the coordinate of the lower surface of the same section under the static state of the steel plate (3) to be detected; delta is a set threshold value, the threshold value is a detection precision value Z of the depth of the discontinuous defects on the surface of the steel plate to be detected n -Z 0 Is the upper surface amplitude value Z of the same section of the steel plate (3) to be detected 0 Is reference point, Z' n -Z′ 0 Is the lower surface amplitude value Z 'of the same section of the steel plate (3) to be detected' 0 Is a reference point; (Z) n -Z 0 )-(Z′ n -Z′ 0 ) Is the amplitude difference.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211445700.2A CN115876788A (en) | 2022-11-18 | 2022-11-18 | 3D visual online detection jitter interference and elimination method for marine steel plate |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211445700.2A CN115876788A (en) | 2022-11-18 | 2022-11-18 | 3D visual online detection jitter interference and elimination method for marine steel plate |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115876788A true CN115876788A (en) | 2023-03-31 |
Family
ID=85760225
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211445700.2A Pending CN115876788A (en) | 2022-11-18 | 2022-11-18 | 3D visual online detection jitter interference and elimination method for marine steel plate |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115876788A (en) |
-
2022
- 2022-11-18 CN CN202211445700.2A patent/CN115876788A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102636490B (en) | Method for detecting surface defects of dustproof cover of bearing based on machine vision | |
| CN109357630B (en) | Multi-type workpiece batch vision measurement system and method | |
| CN105784716B (en) | Friction plate Quality Inspection System | |
| CN103499585B (en) | Based on noncontinuity lithium battery film defect inspection method and the device thereof of machine vision | |
| CN109900711A (en) | Workpiece, defect detection method based on machine vision | |
| CN110503638B (en) | Spiral adhesive quality online detection method | |
| CN109724984A (en) | A kind of defects detection identification device and method based on deep learning algorithm | |
| CN109919934A (en) | A kind of liquid crystal display panel defect inspection method based on the study of multi-source domain depth migration | |
| US20080055591A1 (en) | Apparatus and methods for two-dimensional and three-dimensional inspection of a workpiece | |
| CN104063873A (en) | Shaft sleeve part surface defect on-line detection method based on compressed sensing | |
| CN102455171B (en) | Method for detecting geometric shape of back of tailor-welding weld and implementing device thereof | |
| CN101859378B (en) | Visual detection method for quality of liquid medicine on high-speed pharmaceutical production line | |
| CN106600593B (en) | A kind of middle aluminium porcelain ball surface defect inspection method | |
| CN112037203A (en) | Side surface defect detection method and system based on complex workpiece outer contour registration | |
| CN110443791B (en) | Workpiece detection method and device based on deep learning network | |
| CN106546263A (en) | A kind of laser leveler shoot laser line detecting method based on machine vision | |
| CN108177660B (en) | Steel rail abrasion detection method with laser image processing step | |
| CN107726985A (en) | A kind of three-dimensional gluing detection method and device in real time | |
| CN110991360B (en) | Robot inspection point position intelligent configuration method based on visual algorithm | |
| CN113325001A (en) | Automatic distinguishing and detecting equipment and method for surface appearance flaws of copper foil | |
| CN113538503B (en) | Solar panel defect detection method based on infrared image | |
| CN111340834B (en) | Lining plate assembly system and method based on laser radar and binocular camera data fusion | |
| CN107803606A (en) | A kind of detection method for quality of welding line and device based on overall process mark | |
| CN113393426A (en) | Method for detecting surface defects of rolled steel plate | |
| CN115063579B (en) | Train positioning pin looseness detection method based on two-dimensional image and three-dimensional point cloud projection |
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 |