CN111932632A - Phase correction method in three-dimensional reconstruction of mechanical part - Google Patents
Phase correction method in three-dimensional reconstruction of mechanical part Download PDFInfo
- Publication number
- CN111932632A CN111932632A CN202010668243.8A CN202010668243A CN111932632A CN 111932632 A CN111932632 A CN 111932632A CN 202010668243 A CN202010668243 A CN 202010668243A CN 111932632 A CN111932632 A CN 111932632A
- Authority
- CN
- China
- Prior art keywords
- phase
- image
- point cloud
- grating image
- calibration
- 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
- 238000000034 method Methods 0.000 title claims abstract description 80
- 238000012937 correction Methods 0.000 title claims abstract description 20
- 238000001914 filtration Methods 0.000 claims abstract description 16
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 13
- 238000005070 sampling Methods 0.000 claims abstract description 10
- 238000013507 mapping Methods 0.000 claims abstract description 5
- 230000010363 phase shift Effects 0.000 claims description 11
- 238000007619 statistical method Methods 0.000 claims description 7
- 230000000737 periodic effect Effects 0.000 claims description 6
- 230000006872 improvement Effects 0.000 description 9
- 230000008569 process Effects 0.000 description 9
- 238000010587 phase diagram Methods 0.000 description 8
- 238000005259 measurement Methods 0.000 description 7
- 238000000691 measurement method Methods 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000003321 amplification Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000005315 distribution function Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000005484 gravity Effects 0.000 description 2
- 238000003199 nucleic acid amplification method Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000011089 mechanical engineering Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/20—Finite element generation, e.g. wire-frame surface description, tesselation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses a phase correction method in three-dimensional reconstruction of mechanical parts, which comprises the following steps: moving the calibration plate in space for multiple times, and calibrating the acquired image by using a camera calibration algorithm after acquiring the image of the calibration plate by using a binocular camera to acquire internal parameters and external parameters of the binocular camera; acquiring a part grating image, demodulating the acquired part grating image, acquiring phase information of the part grating image, correcting the phase information of the part grating image, and performing part point cloud reconstruction on the corrected phase information; and carrying out down-sampling and filtering on the part point cloud. The invention can accurately calculate the internal parameters and the external parameters of the camera system. Correcting the phase of the demodulated image to obtain an unfolded phase which is smooth and has no jump error; and (3) carrying out stereo matching on the corrected phase information of the left camera and the right camera, and quickly and accurately reconstructing the three-dimensional point cloud of the mechanical part through phase-height mapping.
Description
Technical Field
The invention relates to the technical field of three-dimensional reconstruction, in particular to a phase correction method in three-dimensional reconstruction of mechanical parts.
Background
Computer vision has been rapidly developed since birth, and computer three-dimensional reconstruction technology is continuously playing an important role in the fields of human production and life. Compared with two-dimensional information, the three-dimensional space information can reflect more characteristics of objects and can be accepted by people more intuitively and comprehensively. Three-dimensional reconstruction has been widely used in various fields, particularly in medical treatment, cultural relic protection, game development, industrial design, and the like.
In the field of manufacturing, the reconstruction of mechanical parts aims at precise measurements. The model reconstruction of the mechanical part has the specialized special requirements, and the digital model provides support for the applications of product processing and manufacturing, reverse mold manufacturing, virtual simulation, product redesign and the like in the field of mechanical engineering, so the requirement on the precision of the reconstruction method is high; most mechanical parts have smooth surfaces, less texture information, reflective materials and the like, and the characteristics of the mechanical parts put higher requirements on a model reconstruction algorithm.
At present, three-dimensional reconstruction methods are mainly classified into contact and non-contact. The traditional method is mostly contact measurement, such as a three-coordinate measuring machine. The three-coordinate measuring machine has the advantages of good repeatability, high precision and the like, but different measuring heads need to be adapted according to different measured objects, and the measuring heads can damage the surfaces of parts. The non-contact measurement method is mainly classified into a passive measurement method and an active measurement method, the passive measurement method is mainly a stereoscopic vision method, and the active measurement method includes a structured light method and the like. Among them, binocular stereo vision and structured light three-dimensional reconstruction are becoming widely used in industry as important research directions. The binocular stereo vision needs to extract and match feature points of two images, and for mechanical parts with smooth surfaces and less texture information, the part surface feature points are less, and the reconstruction effect is poorer. The structured light three-dimensional reconstruction technology mainly comprises the steps of actively projecting a pattern with certain coded information to a measured object through a digital projector, collecting an image through a camera, and recovering the three-dimensional shape of the measured object according to a triangulation principle. The measurement principle is similar to that of binocular stereo vision, but the problem that weak texture or repeated texture areas are difficult to match in the binocular stereo vision can be solved. At present, the structured light equipment is mostly combined by a monocular camera and a projector, and the projector needs to be calibrated in the reconstruction process. Most of digital projectors are calibrated by adopting an inverse camera principle, the calibration is complex, the calibration precision is low, and the reconstruction precision of mechanical parts is influenced. On the basis of research on monocular structured light, the binocular stereo vision principle is fused, the problem that the weak texture or repeated texture area of a mechanical part is difficult to match in binocular stereo vision is solved, calibration of a projector is avoided, and the precision of three-dimensional reconstruction of the part is improved. The three-dimensional measurement technology based on the binocular structured light has the advantages of high precision and high speed, but in actual measurement, due to the influence of random noise such as measurement environment, object surface characteristics, equipment precision and the like, a phase diagram obtained by solving an actually acquired grating image has high noise, a discontinuous point of phase value jumping occurs, and the reconstruction effect is greatly influenced.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a phase correction method in three-dimensional reconstruction of mechanical parts.
In order to achieve the above object, an embodiment of the present invention provides the following technical solutions:
a phase correction method in three-dimensional reconstruction of mechanical parts comprises the following steps:
step S1: moving the calibration plate in space for multiple times, and calibrating the acquired image by using a camera calibration algorithm after acquiring the image of the calibration plate by using a binocular camera to acquire internal parameters and external parameters of the binocular camera;
step S2: acquiring a part grating image, demodulating the acquired part grating image, acquiring phase information of the part grating image, correcting the phase information of the part grating image, and performing part point cloud reconstruction on the corrected phase information;
step S3: and carrying out down-sampling and filtering on the part point cloud.
As a further improvement of the present invention, in step S2, a grating modulated by a periodic function is projected onto the surface of the part by a digital projector, and the part grating image is acquired by a binocular camera.
As a further improvement of the present invention, in step S2, the method for demodulating the obtained part grating image is a method combining a phase shift method and multi-frequency heterodyne, and an absolute phase of the part grating image is obtained.
As a further improvement of the present invention, in the step S2, the following steps are adopted to correct the phase information of the part grating image: firstly, correcting the jump error of +/-2 pi generated by the phase after multi-frequency heterodyne expansion, and then removing the outlier noise phase by adopting the region-by-region median denoising.
As a further improvement of the present invention, in step S2, performing a part point cloud reconstruction on the corrected phase information includes performing a stereo matching on the corrected phase information, and calculating a three-dimensional coordinate of the part through a phase-height mapping.
As a further improvement of the present invention, in step S3, the method for down-sampling the part point cloud is a voxel grid method.
As a further improvement of the present invention, in step S3, the method for filtering the part point cloud is a statistical analysis filtering method.
As a further improvement of the present invention, the statistical analysis filtering method includes calculating the average distance from each point to all neighboring points, and assuming that the obtained distribution is gaussian, obtaining a mean μ and a standard deviation σ, and regarding all points in the neighborhood point set and the points whose neighborhood distances are greater than the interval μ + σ as noise points.
As a further improvement of the present invention, in step S1, the calibration board is a dot calibration board or a checkerboard calibration board.
As a further improvement of the present invention, in step S1, the camera calibration algorithm is a gnomone calibration method.
The invention has the beneficial effects that:
(1) the calibration of the binocular camera system is carried out by using the dot calibration plate through a Zhang-Yongyou calibration method, and the internal parameters and the external parameters of the camera system can be accurately calculated.
(2) Correcting the phase of the demodulated image, correcting phase noise, acquiring smooth image phase information, and obtaining an unfolded phase which is smooth and has no jump error
(3) And (3) carrying out stereo matching on the corrected phase information of the left camera and the right camera, and quickly and accurately reconstructing the three-dimensional point cloud of the mechanical part through phase-height mapping.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flowchart of a phase correction method in three-dimensional reconstruction of a mechanical part according to a preferred embodiment of the present invention;
FIG. 2(a) is a schematic structural diagram of a dot calibration plate according to a preferred embodiment of the present invention;
FIG. 2(b) is a checkerboard calibration plate;
FIG. 3 is a flow chart of a three-dimensional reconstruction of a preferred embodiment of the present invention;
FIG. 4(a) is an unwrapped phase diagram using equation (6);
FIG. 4(b) is an unwrapped phase diagram using equation (7);
FIG. 5(a) is a diagram of the phase after the jump error correction;
FIG. 5(b) is a region diagram;
FIG. 5(c) is a phase diagram after outlier error correction;
FIG. 6 is a point cloud illustration of a reconstructed part according to a preferred embodiment of the present invention;
FIG. 7 is a graph of the results of downsampling in accordance with a preferred embodiment of the present invention;
FIG. 8(a) is a graph of the results after filtering;
fig. 8(b) shows the filtered noise points.
Detailed Description
In order to make those skilled in the art better understand the technical solution of the present invention, the technical solution in the embodiment of the present invention will be clearly and completely described below with reference to the drawings in the embodiment of the present invention, and it is obvious that the described embodiment is only a part of the embodiment of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, a phase correction method in three-dimensional reconstruction of a mechanical part includes the following steps:
step S1: and moving the calibration plate in the space for multiple times, and calibrating the acquired image by using a camera calibration algorithm after acquiring the image of the calibration plate by using the binocular camera to acquire the internal parameters and the external parameters of the binocular camera.
In the calibration process, the position of the calibration plate is transformed for many times, and the calibration plate is collected through a binocular camera system to obtain a plurality of groups of image pairs. In the embodiment, the calibration board is a dot calibration board, as shown in fig. 2(a), the dot calibration board is used as a calibration object, and the dot calibration extracts the center of a dot, so that the optical information of the center of the feature point can be easily obtained. It is to be understood that the calibration plate is not limited to the dot calibration plate, but may be a checkerboard calibration plate, as shown in fig. 2 (b).
In the three-dimensional reconstruction process, camera calibration directly influences the subsequent processing precision such as matching point searching, triangular calculation and the like, so that a camera calibration algorithm is important for the three-dimensional reconstruction precision. In the process of realizing three-dimensional reconstruction, the invention needs to select a proper camera calibration method according to the requirement of a three-dimensional reconstruction algorithm and the reconstruction precision index. In the embodiment, a Zhangyingyou calibration method is adopted for camera calibration, and is a method between a traditional calibration method and a self-calibration method, so that the defects of high equipment requirement, complex operation and the like of the traditional method are overcome, the precision is higher than that of the self-calibration method, the method is a common method for calibrating a computer vision system, and the Zhangyingyou calibration method is used for calculating images to obtain internal parameters and external parameters of a binocular camera system.
Step S2: collecting a part grating image, demodulating the obtained part grating image, acquiring phase information of the part grating image, correcting the phase information of the part grating image, and performing part point cloud reconstruction on the corrected phase information.
The three-dimensional part reconstruction process of the present invention is shown in fig. 3.
Step S2.1: a grating modulated by a periodic function is projected to the surface of a part through a digital projector, a part grating image is collected through a binocular camera, and polar line correction is carried out on the part grating image. Specifically, a fringe grating pattern modulated by a sine periodic function is projected to the surface of a part through a digital projector, a left camera and a right camera of a binocular camera simultaneously acquire a structured light image modulated by the three-dimensional surface profile of the part, and polar line correction is respectively performed on a left fringe grating image acquired by the left camera and a right fringe grating image acquired by the right camera, so that corresponding feature points of the left fringe grating image and the right fringe grating image are on the same scanning line.
Step S2.2: and demodulating the obtained part grating image by combining a phase shift method and multi-frequency heterodyne to obtain the absolute phase of the part grating image.
And performing three-dimensional reconstruction by projecting the modulation grating to obtain a grating projection method. The principle of the grating projection method is that a grating modulated by a periodic function is projected on the surface of a measured part, the phase of the grating stripe of each point is shifted due to the change of the surface height of the part, and the height information of an object point can be solved by phase information through the relation between the phase shift amount and the surface height in a measuring system. The phase shift method is widely used because of high precision and strong anti-interference performance. The phase shift method is to project a group of stripe grating images with certain phase difference values to the surface of the part to be measured, the stripe grating images are deformed after the surface modulation of the part, and the phase principal value of each point on the image can be obtained after the stripe grating images are decoded. Assuming that the image intensity follows a standard sinusoidal distribution, the intensity distribution function is:
wherein, Ii(x, y) is the gray value at point (x, y) on the ith image, I' (x, y) is the average gray of the image, I "(x, y) is the gray modulation intensity, and
for the phase main value (relative phase value) to be obtained, N is the phase shift step number, i.e. the number of fringe patterns, and θ represents the phase of the grating image.
The invention adopts a four-step phase shift method to solve the phase principal value.
Using the four-step phase shift method requires projecting four grating images spaced at pi/2 phase shifts, i.e., θ is 0, pi/2, pi, and 3 pi/2, respectively, and the light intensity distribution function under different phase shift values is:
from equation (3), the phase principal value of the grating image can be calculated as:
according to the characteristics of the arctan function, the solved phase principal value is truncated between (-pi, pi), but because a plurality of periodic stripes exist in the whole measuring space range, the phase principal value changes periodically. In order to obtain a global complete continuous phase, the phase principal value of a space point needs to be subjected to phase unwrapping so as to obtain a unique absolute phase value in a global range.
The phase unwrapping algorithm based on the multi-frequency heterodyne principle has the advantages that the phase unwrapping process mainly depends on phase principal values of grating images with different frequencies, the phase unwrapping algorithm is insensitive to the surface color of a part to be measured, the calculation process is more stable, and the phase unwrapping algorithm can be used for carrying out phase unwrapping on the grating images with each frequency. The invention adopts multi-frequency heterodyne to perform phase unwrapping.
The multi-frequency heterodyne method is to wrap two different frequencies into a phase
And
by adding, a phase function of smaller frequency can be generated
Definition of
Respectively at a frequency of f1、f2、fbIf f is1、f2As is known, f can be calculated by the formula (4)b。
Is provided with
And
respectively of the phase principal value functions of1(x, y) and ω2(x，y)，
And
the function of the phase principal value after the heterodyne is omegab(x, y), formula (5) can be seen:
according to the obtained omegabFrom equation (6), theObtain the phase omega of the principal value1(x，y)。
As can be seen from the observation of the formulas (5) and (6), ω is expressed bybMultiplying by a factor
Will make the utility model
Is amplified by many times, resulting in the desired omega1A large error is generated.
And because of omega1Can also be expressed as formula (7), and the error amplification phenomenon can be avoided by using the formula to carry out solution, wherein N1Is an integer-level stripe, and can be obtained by equation (8).
Step S2.3: correcting phase information of the part grating image, including correcting jump errors of +/-2 pi generated by the phase after multi-frequency heterodyne expansion, and then removing the outlier noise phase by adopting region-by-region median denoising.
The wrapping phase obtained by the equations (2) and (3) is caused by the influence of the equipment precision and the environmental noise
Errors inevitably occur. The phase unwrapped using equation (6) is denoted as ω1Unwrapping the wrapped phase using equation (6) results in
Is amplified to the original
Multiple, resulting in the desired ω1A large error is generated. The principal value phase ω is obtained from equation (6)1(x, y) the phase diagram of a certain line is shown in FIG. 4(a), in which the unit of abscissa x is pixel (pixel) and ordinate ω is ordinate ω1The unit of (2) is rad (radian), and the noise points in the phase information obtained by the formula (6) are too much, which causes great interference to the phase matching of the left and right camera images and leads to the clutter of reconstructed part point clouds.
The phases developed by equations (7) and (8) are denoted as ω'1And the wrapping phase is unfolded by adopting a formula (7) and a formula (8), so that the error amplification phenomenon of the formula (6) is avoided. FIG. 4(b) shows a phase diagram of one line of principal value phases obtained from equations (7) and (8), in which the unit of abscissa x is pixel and ordinate ω'1The unit of (d) is rad, and as can be seen from the figure, the unwrapped phase is not smooth, and has some jump errors, and analysis formula (8) shows that, when the phase is unwrapped, the number of fringe series is rounded, and the error of rounding causes the phase to have jump errors of + -2 pi.
Albeit omega1The error is large, but can not reach the multiple error of 2 pi, and is close to the true value as a whole, so that the error can be adopted to omega'1And (6) carrying out correction. The correction process is shown in equation (9), and the corrected phase diagram is shown in fig. 5 (a).
As can be seen from fig. 5(a), the corrected phase is smooth, but outlier noise is still present at an abscissa of 500. The reason for this is shown in FIG. 4(a) and FIG. 4(b), and ω is1And ω'1All the coordinates have outlier error and cannot pass through omega1To ω'1And (6) carrying out correction.
The method removes outlier noise phase by adopting region-by-region median denoising, and obtains a smooth absolute phase. In this embodiment, the width and the height of a photo taken by the binocular camera are 2440 pixels and 2048 pixels, the photo width direction is an x coordinate axis direction, 1/20 of an abscissa x is a width of one region, the width and the height of each region are 2440 × 1/20 to 122 and 1, as shown in fig. 5(b), each cell represents one region, each row has 20 regions and is 2048 rows, a median value of each region is obtained from each row region by region, a noise threshold is set, each point of the region is compared with the median value of the region, if an absolute value of a difference is greater than the noise threshold, the point is a noise point to be removed, and the next row repeats the operation of the previous row. The median value of each region is preferably obtained by taking the median value after ascending sorting in the bubble sorting method. Fig. 5(c) shows a phase diagram of a line subjected to region-by-region median denoising. As shown in fig. 5(c), the denoised phase values are smooth and free of outlier noise.
Step S2.4: and (3) carrying out stereo matching on the corrected phase information of the left camera and the right camera, calculating the three-dimensional coordinates of the part through phase-height mapping, and recovering the three-dimensional morphology of the part to be detected, wherein the point cloud reconstruction result of the part is shown in fig. 6.
Step S3: and carrying out down-sampling and filtering on the part point cloud.
Step S3.1: and carrying out down-sampling on the part point cloud by a voxel grid method.
The point cloud data obtained by three-dimensional reconstruction through structured light is too large, the density distribution of the point cloud is not uniform, and down sampling of the point cloud is required. The voxel grid method has the advantage that the shape characteristics of the point cloud can be saved during down-sampling. The method creates a three-dimensional voxel grid for the input point cloud data, and the center of gravity of all points in each voxel is used for approximately displaying other points in the voxel, so that all points in the voxel are finally represented by a center of gravity point. The data volume of the point cloud can be compressed by a voxel grid method, and the problem of point cloud unevenness can be solved. The point cloud down-sampling results are shown in fig. 7.
Step S3.2: and carrying out point cloud filtering by a statistical analysis filtering method.
In the process of three-dimensional reconstruction by using structured light, due to the influence of the precision of structured light equipment, environmental noise, background light and the like, reconstructed part point cloud data has noise points, the noise points can complicate estimation of local point cloud characteristics (such as surface normal or curvature change), and wrong values can be generated, so that subsequent point cloud registration, point cloud gridding and the like can be influenced. The invention adopts a statistical analysis filtering method to carry out point cloud filtering and remove noise points. The method uses statistical analysis techniques to centrally remove measurement noise points from a point cloud data. Specifically, K neighborhood points of each point in the point cloud are obtained by using a K-Means clustering algorithm to form a neighborhood point set, the average distance between each point and all adjacent points is calculated for each point, a mean value mu and a standard deviation sigma can be obtained by assuming that the obtained distribution is Gaussian distribution, and then all points in the neighborhood point set and points outside the neighborhood distance of the points is larger than the interval mu + sigma can be regarded as noise points. The point cloud filtering results are shown in fig. 8(a) and 8(b), where fig. 8(a) shows the filtered results and fig. 8(b) shows the filtered noise points.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Furthermore, it should be understood that although the present description refers to embodiments, not every embodiment may contain only a single embodiment, and such description is for clarity only, and those skilled in the art should integrate the description, and the embodiments may be combined as appropriate to form other embodiments understood by those skilled in the art.
Claims (10)
1. A phase correction method in three-dimensional reconstruction of mechanical parts is characterized by comprising the following steps:
step S1: moving the calibration plate in space for multiple times, and calibrating the acquired image by using a camera calibration algorithm after acquiring the image of the calibration plate by using a binocular camera to acquire internal parameters and external parameters of the binocular camera;
step S2: acquiring a part grating image, demodulating the acquired part grating image, acquiring phase information of the part grating image, correcting the phase information of the part grating image, and performing part point cloud reconstruction on the corrected phase information;
step S3: and carrying out down-sampling and filtering on the part point cloud.
2. The method for phase correction in three-dimensional reconstruction of mechanical part according to claim 1, wherein in step S2, a grating modulated by a periodic function is projected onto the surface of the part by a digital projector, and the part grating image is collected by a binocular camera.
3. The method for correcting phase in three-dimensional reconstruction of mechanical part according to claim 1 or 2, wherein in step S2, the method for demodulating the obtained part grating image is a method combining phase shift method and multi-frequency heterodyne method, and the absolute phase of the part grating image is obtained.
4. The method of claim 3, wherein the step S2 of correcting the phase information of the part grating image comprises the following steps: firstly, correcting the jump error of +/-2 pi generated by the phase after multi-frequency heterodyne expansion, and then removing the outlier noise phase by adopting the region-by-region median denoising.
5. The method of claim 4, wherein the step S2, the point cloud reconstructing the corrected phase information comprises stereo matching the corrected phase information, and calculating the three-dimensional coordinates of the part through phase-height mapping.
6. The method for phase correction in three-dimensional reconstruction of mechanical part according to claim 1, wherein the down-sampling method for the part point cloud in step S3 is voxel grid method.
7. The method for phase correction in three-dimensional reconstruction of mechanical part according to claim 1, wherein the filtering method for the part point cloud in step S3 is a statistical analysis filtering method.
8. The method of claim 7, wherein the statistical analysis filtering method comprises calculating an average distance from each point to all neighboring points, and assuming that the obtained distribution is Gaussian, and a mean μ and a standard deviation σ are obtained, all points in the neighborhood point set whose neighborhood distance is greater than the interval μ + σ are considered as noise points.
9. The method for phase correction in three-dimensional reconstruction of mechanical part according to claim 1, wherein in step S1, the calibration plates are circular calibration plates or checkerboard calibration plates.
10. The method for correcting phase in three-dimensional reconstruction of mechanical part according to claim 1, wherein in step S1, the camera calibration algorithm is a tensor calibration method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010668243.8A CN111932632A (en) | 2020-07-13 | 2020-07-13 | Phase correction method in three-dimensional reconstruction of mechanical part |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010668243.8A CN111932632A (en) | 2020-07-13 | 2020-07-13 | Phase correction method in three-dimensional reconstruction of mechanical part |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111932632A true CN111932632A (en) | 2020-11-13 |
Family
ID=73312465
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010668243.8A Pending CN111932632A (en) | 2020-07-13 | 2020-07-13 | Phase correction method in three-dimensional reconstruction of mechanical part |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111932632A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112766256A (en) * | 2021-01-25 | 2021-05-07 | 北京淳中科技股份有限公司 | Grating phase diagram processing method and device, electronic equipment and storage medium |
| CN114219841A (en) * | 2022-02-23 | 2022-03-22 | 武汉欧耐德润滑油有限公司 | Automatic lubricating oil tank parameter identification method based on image processing |
| CN116912429A (en) * | 2023-09-13 | 2023-10-20 | 江苏普旭科技股份有限公司 | Three-dimensional reconstruction method and system for high-definition video IG (inter-group) material |
| WO2024021654A1 (en) * | 2022-07-28 | 2024-02-01 | 江苏集萃智能光电系统研究所有限公司 | Error correction method used for line structured light 3d camera, and apparatus |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104330052A (en) * | 2014-11-21 | 2015-02-04 | 天津工业大学 | Heterodyne three-frequency unequal range phase displacement solution phase method |
| CN107607060A (en) * | 2017-08-24 | 2018-01-19 | 东南大学 | A kind of phase error compensation method in the measurement applied to grating tripleplane |
| CN108335350A (en) * | 2018-02-06 | 2018-07-27 | 聊城大学 | The three-dimensional rebuilding method of binocular stereo vision |
| CN109186492A (en) * | 2018-08-14 | 2019-01-11 | 博众精工科技股份有限公司 | Three-dimensional rebuilding method, apparatus and system based on one camera |
| CN110160468A (en) * | 2019-04-29 | 2019-08-23 | 东南大学 | It is a kind of to defocus optical grating projection method for three-dimensional measurement for Moving Objects |
| CN111207693A (en) * | 2020-01-10 | 2020-05-29 | 西安交通大学 | Three-dimensional measurement method of turbine blade ceramic core based on binocular structured light |
-
2020
- 2020-07-13 CN CN202010668243.8A patent/CN111932632A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104330052A (en) * | 2014-11-21 | 2015-02-04 | 天津工业大学 | Heterodyne three-frequency unequal range phase displacement solution phase method |
| CN107607060A (en) * | 2017-08-24 | 2018-01-19 | 东南大学 | A kind of phase error compensation method in the measurement applied to grating tripleplane |
| CN108335350A (en) * | 2018-02-06 | 2018-07-27 | 聊城大学 | The three-dimensional rebuilding method of binocular stereo vision |
| CN109186492A (en) * | 2018-08-14 | 2019-01-11 | 博众精工科技股份有限公司 | Three-dimensional rebuilding method, apparatus and system based on one camera |
| CN110160468A (en) * | 2019-04-29 | 2019-08-23 | 东南大学 | It is a kind of to defocus optical grating projection method for three-dimensional measurement for Moving Objects |
| CN111207693A (en) * | 2020-01-10 | 2020-05-29 | 西安交通大学 | Three-dimensional measurement method of turbine blade ceramic core based on binocular structured light |
Non-Patent Citations (6)
| Title |
|---|
| FPP AND DL等: "双目结构光的三维重建方法（相移法+多频外差相位解包）（含相位代码）", pages 1, Retrieved from the Internet <URL:https://blog.csdn.net/qq_32638769/article/details/108169187> * |
| 刘飞等: "基于多频外差的全频解相方法", 《激光与光电子学进展》, vol. 56, no. 1, pages 173 - 180 * |
| 张峰: "基于双目结构光的机械零件三维重建方法研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》, no. 1, pages 029 - 447 * |
| 朱登明等: "基于图像的启发式三维火焰重建算法", 《高技术通讯》, no. 2, pages 122 - 130 * |
| 王素琴等: "面向复杂机械零件形貌测量的高精度三维重建方法", 《红外与激光工程》, vol. 51, no. 7, pages 330 - 340 * |
| 赵成星: "光栅四步相移法的三维重建", 《激光杂志》, vol. 41, no. 10, pages 34 - 38 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112766256A (en) * | 2021-01-25 | 2021-05-07 | 北京淳中科技股份有限公司 | Grating phase diagram processing method and device, electronic equipment and storage medium |
| CN112766256B (en) * | 2021-01-25 | 2023-05-30 | 北京淳中科技股份有限公司 | Grating phase diagram processing method and device, electronic equipment and storage medium |
| CN114219841A (en) * | 2022-02-23 | 2022-03-22 | 武汉欧耐德润滑油有限公司 | Automatic lubricating oil tank parameter identification method based on image processing |
| WO2024021654A1 (en) * | 2022-07-28 | 2024-02-01 | 江苏集萃智能光电系统研究所有限公司 | Error correction method used for line structured light 3d camera, and apparatus |
| CN116912429A (en) * | 2023-09-13 | 2023-10-20 | 江苏普旭科技股份有限公司 | Three-dimensional reconstruction method and system for high-definition video IG (inter-group) material |
| CN116912429B (en) * | 2023-09-13 | 2023-12-08 | 江苏普旭科技股份有限公司 | Three-dimensional reconstruction method and system for high-definition video IG (inter-group) material |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111932632A (en) | Phase correction method in three-dimensional reconstruction of mechanical part | |
| CN104061879B (en) | A kind of structural light three-dimensional face shape vertical survey method continuously scanned | |
| CN107798698B (en) | Structured light stripe center extraction method based on gray correction and adaptive threshold | |
| WO2018040017A1 (en) | Method and system for correcting distortion of projector lens based on adaptive fringes | |
| CN109945802B (en) | Structured light three-dimensional measurement method | |
| CN112465912B (en) | Stereo camera calibration method and device | |
| CN109141291A (en) | A kind of fast phase unwrapping algorithm | |
| CN111174730B (en) | Rapid phase unwrapping method based on phase encoding | |
| CN110174079B (en) | Three-dimensional reconstruction method based on four-step phase-shift coding type surface structured light | |
| CN111563952B (en) | Method and system for realizing stereo matching based on phase information and spatial texture characteristics | |
| CN108362226B (en) | Double four-step phase shift method for improving phase measurement precision of image overexposure area | |
| CN111947599B (en) | Three-dimensional measurement method based on learning fringe phase retrieval and speckle correlation | |
| CN107014313B (en) | Method and system for weighted least square phase unwrapping based on S-transform ridge value | |
| JP5761750B2 (en) | Image processing method and apparatus | |
| CN110223384A (en) | A kind of white light interference three-dimensional appearance method for reconstructing, device, system and storage medium | |
| CN110375675B (en) | Binocular grating projection measurement method based on space phase expansion | |
| Yang et al. | A dual-platform laser scanner for 3D reconstruction of dental pieces | |
| CN112802084B (en) | Three-dimensional morphology measurement method, system and storage medium based on deep learning | |
| He et al. | A composite-structured-light 3D measurement method based on fringe parameter calibration | |
| CN111583323B (en) | Single-frame structure light field three-dimensional imaging method and system | |
| CN111947600B (en) | Robust three-dimensional phase unfolding method based on phase level cost filtering | |
| KR101001894B1 (en) | Apparatus and method for 3-D profilometry using color projection moire technique | |
| CN116934981A (en) | Stripe projection three-dimensional reconstruction method and system based on dual-stage hybrid network | |
| CN115290004B (en) | Underwater parallel single-pixel imaging method based on compressed sensing and HSI | |
| Gdeisat et al. | Simple and accurate empirical absolute volume calibration of a multi-sensor fringe projection system |
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 |