CN111207693A - Three-dimensional measurement method of turbine blade ceramic core based on binocular structured light - Google Patents
Three-dimensional measurement method of turbine blade ceramic core based on binocular structured light Download PDFInfo
- Publication number
- CN111207693A CN111207693A CN202010025374.4A CN202010025374A CN111207693A CN 111207693 A CN111207693 A CN 111207693A CN 202010025374 A CN202010025374 A CN 202010025374A CN 111207693 A CN111207693 A CN 111207693A
- Authority
- CN
- China
- Prior art keywords
- ceramic core
- phase
- camera
- point
- structured light
- 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
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
- G01B11/25—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures by projecting a pattern, e.g. one or more lines, moiré fringes on the object
- G01B11/254—Projection of a pattern, viewing through a pattern, e.g. moiré
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
A three-dimensional measurement method of a turbine blade ceramic core based on binocular structured light comprises the steps of firstly calibrating a left camera and a right camera, projecting structured light stripes onto the ceramic core, simultaneously collecting deformation stripe patterns modulated by the surface of the ceramic core by the left camera and the right camera, carrying out image preprocessing on the deformation stripe patterns by a computer, demodulating phases to obtain wrapping phases, only extracting wrapping phase distribution of the ceramic core through image segmentation, and unwrapping to obtain absolute phase distribution; carrying out stereo matching on absolute phase distribution based on polar line constraint and phase constraint, and calculating a single point cloud of the ceramic core according to a matching result and camera calibration parameters; rotating the high-precision rotary table to obtain single point clouds of the ceramic core at different angles, splicing to obtain a complete point cloud model of the ceramic core, transmitting the point cloud model to an industrial personal computer for clamping pose measurement and to-be-repaired path extraction, and guiding a laser to finish automatic repair of the ceramic core; the invention realizes on-line three-dimensional measurement and meets the requirements of high precision and high speed of automatic shaping of the ceramic core.
Description
Technical Field
The invention belongs to the technical field of binocular structured light-based three-dimensional measurement, and particularly relates to a binocular structured light-based three-dimensional measurement method for a turbine blade ceramic core.
Background
The ceramic core is a core component for forming a complex inner cavity structure of the hollow turbine blade of the aero-engine. At present, ceramic cores are mainly formed through hot-pressing injection, the problems that the formed ceramic cores are easy to cause flash, burrs, hole blockage and the like at parting surfaces are solved through shape modification, the size precision of the ceramic cores can be guaranteed only after the problems are solved, the ceramic cores are limited by the immaturity of an online measurement technology, automatic shape modification is not achieved at present, and manual shape modification is mostly adopted. Most of the ceramic cores are thin-wall fragile and free complex curved surface structures, and the problems of low yield (less than 20%), poor size precision (0.5mm), irregular micro structure and the like exist in manual shaping. In order to solve the difficult problem of the ceramic core trimming of the turbine blade, high-precision and automatic ceramic core trimming equipment must be developed. Because the ceramic core has a free complex curved surface structure, the shaping path of the ceramic core is a complex space curve, the clamping pose and the shaping path are required to be obtained based on the point cloud model, and then the shaping path after pose correction is fed back to a numerical control machine tool processing system to guide the laser to complete automatic shaping along the path to be shaped. Therefore, the online measurement of the high-precision point cloud model is a prerequisite condition for realizing the measurement of the clamping pose and the extraction of the path to be repaired, and is key input for realizing the automation of the ceramic core repairing.
The three-dimensional measurement system is used for carrying out online measurement on the measured object to generate a high-precision point cloud model, and compared with an image, the point cloud model can well solve the problem that core clamping pose and depth information are difficult to acquire. At present, there are various three-dimensional measurement methods, wherein the structured light three-dimensional measurement method is non-contact measurement and can be used for measuring a component with a free complex curved surface, such as a ceramic core; the method has high measurement speed and high precision, is suitable for online measurement, and is convenient for realizing the automation of the ceramic core shaping; the anti-interference performance to the environment is strong, and the laser trimming method can adapt to the complex industrial environment during laser trimming; the ceramic core is a white surface diffuse reflection component, and can overcome the defect that the structured light measurement method is easily influenced by color and smooth plane reflection. In the structured light three-dimensional measurement method, the binocular structured light measurement is wider than the monocular measurement range, so the binocular structured light three-dimensional measurement method is an ideal choice for reconstructing the ceramic core three-dimensional point cloud model. The existing products based on the structured light three-dimensional measurement method mainly have two types: one type is a structured light depth camera which is simple and convenient to use but has poor precision and cannot meet the requirement of accurate measurement of a fine structure of a ceramic core; the other type is a commercial three-dimensional scanner, the device can realize high-precision measurement, but the product positioning mostly realizes high-precision off-line measurement, the time consumption is long, and the high-efficiency on-line measurement is difficult to realize when the device is integrated with a numerical control processing system.
Disclosure of Invention
In order to overcome the existing defects, the invention aims to provide a binocular structured light-based three-dimensional measurement method for a turbine blade ceramic core, which is used for realizing high-precision three-dimensional online measurement of the ceramic core and providing a high-precision point cloud model for clamping pose measurement and core to-be-repaired path extraction.
In order to achieve the purpose, the invention adopts the technical scheme that:
a three-dimensional measurement method of a turbine blade ceramic core based on binocular structured light comprises the following steps:
the method comprises the following steps: simultaneously shooting at least 5 calibration plate images at different angles by using a left camera and a right camera to calibrate the cameras, and respectively obtaining internal parameters and distortion parameters of the left camera and the right camera and a rotation matrix and a translation vector representing the position relationship of the left camera and the right camera;
step two: generating structural light stripes with specific patterns through computer software, projecting the structural light stripes onto the ceramic core through a structural light projection device, enabling the stripes to be deformed due to the modulation of the surface height of the ceramic core, and simultaneously acquiring deformation stripe patterns containing the surface depth characteristics of the ceramic core by a left camera and a right camera and transmitting the deformation stripe patterns to a computer;
step three: the computer carries out image preprocessing on the deformed fringe patterns acquired by the left camera and the right camera;
step four: performing phase demodulation on the left and right camera deformed fringe patterns after image preprocessing, converting light intensity information directly acquired by the left and right cameras into phase information, and obtaining wrapped phase value distribution of each point in the left and right images;
step five: calculating the modulation degree of the wrapping phase diagrams of the left camera and the right camera, carrying out image segmentation, separating the ceramic core from the background, and only extracting the wrapping phase distribution of the ceramic core;
step six: performing phase unwrapping on each point in the ceramic core wrapped phase distribution left and right images to obtain left and right images of the absolute phase distribution of the ceramic core;
step seven: polar line correction is carried out on the left and right images of the absolute phase distribution of the ceramic core obtained in the sixth step by utilizing the camera calibration result obtained in the first step, the internal parameters and distortion parameters of the left and right cameras and the rotation matrix and translation vector representing the position relationship of the left and right cameras, and a new corrected absolute phase left and right image is obtained by adopting a bilinear interpolation method;
step eight: and C, stereo matching is carried out on the new absolute phase left and right images after polar line correction in the step seven based on polar line constraint and phase constraint: setting a phase region template, obtaining an initial matching point based on a template matching method, and performing sub-pixel stereo matching by using a linear interpolation method;
step nine: using a stereo matching result, combining the camera calibration result of the step one, the internal parameters and distortion parameters of the left camera and the right camera, and the rotation matrix and translation vector representing the position relationship of the left camera and the right camera, and calculating three-dimensional coordinates corresponding to all stereo matching points on the ceramic core by utilizing a triangulation principle, namely obtaining a single point cloud of the ceramic core;
step ten: respectively rotating a high-precision turntable loaded with the special fixture for the ceramic core by 90 degrees, 180 degrees and 270 degrees, repeating the second step to the ninth step to obtain single point clouds of the ceramic cores at four different angles, calculating a pose transformation matrix of the point clouds of the ceramic cores at different angles based on corner information, and realizing splicing of the point clouds at multiple viewing angles to obtain a complete point cloud model of the ceramic core;
step eleven: transmitting the complete point cloud model of the ceramic core to an industrial personal computer of a numerical control machine tool for clamping pose measurement and to-be-repaired path extraction, and guiding a laser to finish automatic repair of the ceramic core;
the device adopted by the binocular structured light-based three-dimensional measurement method for the ceramic core of the turbine blade comprises a structured light projection device 1, two CCD cameras, a special ceramic core clamp 5, a high-precision rotary table 7 and a computer which is communicated with the structured light projection device 1 and the two CCD cameras, wherein the structured light projection device 1 and the two CCD cameras are fixed on a stand column 6 of a numerical control machine, and the two CCD cameras are a left camera 2 and a right camera 3; the ceramic core 4 is fixed on a high-precision turntable 7 by a ceramic core special fixture 5, and the left camera 2 and the right camera 3 are aligned with the ceramic core 4.
In the step one, a Zhangyingyou plane calibration method is utilized to calibrate the left camera and the right camera to respectively obtain the internal parameter matrixes A of the left camera and the right cameral,ArDistortion coefficient k1l,k2l,k1r,k2rAnd a rotation matrix R and a translational vector T characterizing the positional relationship of the left camera.
And in the second step, the structured light stripes of the specific pattern are projected as blue light sine stripes, and the continuous change of the gray scale of the stripes is required to be ensured when the stripes are collected.
And in the third step, the computer preprocesses the deformed fringe images collected by the left camera and the right camera by a Gaussian frequency domain filtering method, so that random noise interference is reduced.
The phase demodulation in the fourth step is to solve the phase information by a four-step phase shift method, and specifically, the phase demodulation is performed on the left image and the right image by the following formula:
in the formula:
the phase value of the envelope is distributed in (-pi, pi)]；I1、I2、I3、I4For the light intensity of a deformed fringe pattern acquired by a four-step phase-shift CCD cameraAnd (4) distribution.
In the step five, before phase unwrapping, removing backgrounds except the ceramic core in the left and right deformation stripe images; calculating a modulation parameter, wherein the modulation parameter is defined as:
in the formula: n is the total number of phase shift steps, I represents the number of phase shift steps, IiRepresenting the gray scale of the fringe image corresponding to the ith step phase shift; the modulation degree calculated by the stripes on the ceramic core is higher than the background, so that the threshold value is set to separate the modulation degree from the background, and an iterative method is adopted to solve the optimal threshold value.
And sixthly, performing phase unwrapping on wrapped phase values of all points in the left image and the right image by adopting a multi-frequency heterodyne method, wherein the frequency combination of unwrapping by adopting the multi-frequency heterodyne method is selected from 74, 68 and 63, and obtaining the left image and the right image of the absolute phase distribution of the ceramic core.
And seventhly, performing polar line correction on the left and right images of the absolute phase distribution of the ceramic core, wherein the corrected system structure is a head-up binocular standard geometric structure, so that a certain point in the left image is distributed on the same horizontal line at a matching point in the right image, and calculating new absolute phases for the corrected left and right camera phase diagrams by adopting a bilinear interpolation method.
And step eight, setting a phase region template by utilizing epipolar constraint and phase constraint, and obtaining an initial matching point by a method for calculating region similarity based on template matching.
The formula according to when calculating the three-dimensional coordinates in the ninth step is as follows:
in the formula: b is the base length, f is the focal length, cc1Is the abscissa of the optical center of the left camera, c1The horizontal coordinate of the current pixel point is v, the vertical coordinate of the optical center is subtracted from the vertical coordinate of the current pixel point, d is the parallax of the space point in the left image and the right image, and d is expressed as:
d＝cc1-c1+c2-cc2
in the formula: c. Cc2Is the abscissa of the optical center of the right camera, c2And the abscissa of the matching point in the right image of the current pixel point.
The invention has the beneficial effects that:
1. the method realizes the point cloud splicing of different angles based on the acquisition of the corner information of the high-precision rotary table. Compared with the method for continuously iterating by the common ICP algorithm, the method for realizing splicing by using data matrix transformation has the advantage of higher speed; meanwhile, the high precision of the rotary table is benefited, and the precision of the splicing process is higher. The advantages of rapidness, high precision and the like provide convenient conditions for the automation of the shaping of the ceramic core.
2. Because the method of the invention carries out image segmentation on the deformed stripe images collected by the left camera and the right camera by calculating the modulation degree parameter, the removal of the background can greatly improve the stereo matching speed; meanwhile, the point cloud model of the ceramic core is generated more accurately, background points do not need to be manually removed after the point cloud is generated, and convenience is provided for automation of ceramic core shaping.
3. According to the invention, the initial matching point is obtained by using the template matching method during stereo matching, and the accuracy of matching is improved compared with point matching by using area matching; the linear interpolation is adopted to carry out sub-pixel level matching, and the matching point calculation is more accurate, so that the method has the advantage of more accurate point cloud model generation.
Drawings
FIG. 1 is a schematic view of a measuring apparatus according to an embodiment of the present invention.
FIG. 2 is a flowchart of a measurement method according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of a binocular stereoscopic imaging model according to an embodiment of the present invention.
Detailed Description
The invention is further illustrated with reference to the following figures and examples.
As shown in fig. 1, a three-dimensional measuring device for a turbine blade ceramic core based on binocular structured light comprises a structured light projection device 1, two CCD cameras, a ceramic core special fixture 5, a high-precision rotary table 7 and a computer which is communicated with the structured light projection device 1 and the two CCD cameras, wherein the structured light projection device 1 and the two CCD cameras are fixed on a column 6 of a numerical control machine, and the two CCD cameras are a left camera 2 and a right camera 3; the ceramic core 4 is fixed on a high-precision turntable 7 by a ceramic core special fixture 5, and the left camera 2 and the right camera 3 are aligned with the ceramic core 4.
As shown in fig. 2, a three-dimensional measurement method of a ceramic core of a turbine blade based on binocular structured light comprises the following steps:
the method comprises the following steps: using a left camera 2 and a right camera 3 to shoot 15 calibration plate images with different angles simultaneously for camera calibration to obtain internal parameter matrixes A of the left camera and the right camera respectivelyl,ArAnd distortion parameter k1l,k2l,k1r,k2rThe right camera coordinate system is converted into a rotation matrix R and translation vector T of 3 x 3 of the left camera coordinate system;
step two: blue light sine stripes are generated through computer software and projected onto the ceramic core 4 through the structured light projection device 1, the blue light has stronger anti-interference performance compared with common white light, the stripes are modulated by the height of the surface of the ceramic core 4 to generate deformation, the left camera and the right camera simultaneously acquire deformation stripe images containing the surface depth characteristics of the ceramic core and transmit the deformation stripe images to the computer, the deformation stripe images are out of focus when the images are acquired, higher harmonics are filtered, the continuous change of the gray scale of the stripes is ensured, and the stripe images are not over exposed;
step three: the computer carries out image preprocessing on the deformed fringe patterns acquired by the left camera and the right camera by adopting a Gaussian frequency domain filtering method so as to reduce random noise interference;
step four: performing phase demodulation on the left and right camera deformed fringe patterns after image preprocessing, converting light intensity information directly acquired by the left and right cameras into phase information, and obtaining wrapped phase value distribution of each point in the left and right images; phase demodulation is completed based on a four-step phase shift method, and phase information is solved by the following formula:
in the formula:
the phase value of the envelope is distributed in (-pi, pi)]；I1、I2、I3、I4The light intensity distribution of the deformed fringe pattern acquired by the four-step phase-shift CCD camera is obtained;
step five: calculating the modulation degree of the wrapping phase diagrams of the left camera and the right camera, carrying out image segmentation, separating the ceramic core from the background, and only extracting the wrapping phase distribution of the ceramic core;
the image segmentation is to remove the background except the ceramic core in the left and right deformation stripe images before the phase unwrapping, compared with the prior art, the added background removal step can ensure that the generated point cloud model has no background interference, reduce the complexity of subsequent point cloud processing and greatly improve the matching speed in the subsequent steps; the modulation parameter is defined as:
in the formula: n is the total number of phase shift steps, I represents the number of phase shift steps, IiRepresenting the gray scale of the fringe image corresponding to the ith step phase shift;
the modulation degree obtained by calculating the stripes on the ceramic core is higher than the background, so that a threshold value is set for the modulation degree, a binary template is generated after binarization, points with the value of 1 on the template are effective points, and points with the value of 0 on the template are ineffective points; only unwrapping points with the value of 1 when unwrapping the phase, and obtaining continuous phases for invalid points by an interpolation method; the selection of the regulation degree threshold is very important, the threshold is too large, the number of invalid points is large, and difficulty is brought to later-stage interpolation; the threshold is too small, the reliability of the point considered as the reliable point is reduced, and an iterative method is adopted to solve the optimal threshold;
step six: performing phase unwrapping on each point in the ceramic core wrapped phase distribution left and right images after image segmentation to obtain left and right images of the absolute phase distribution of the ceramic core; by adopting a multi-frequency heterodyne method, the spatial period of the frequency combination is selected to be stripes as 74, 68 and 63,
formula of multi-frequency heterodyne method:
in the formula: psi1The phase of the fringes, phi, representing the frequency one2The phase of the fringes representing the frequency two,
is a wrapped phase value for the frequency one,
for the wrapped phase value of frequency two, INT denotes the rounding operation, f1,f2,f12Respectively representing a frequency one, a frequency two and a synthesized frequency of the frequency one and the frequency two; therefore, when the grating frequency is increased, the digitization error of the sinusoidal grating is increased; when the grating frequency is reduced, the resolution of the digital grating on the surface of the object is reduced, which is not beneficial to the precise measurement of the surface of the ceramic core; the heterodyne frequency combinations with fringe space periods of 74, 68, 63 are a set of frequencies with the smallest average error as determined by the plate flatness experiment;
step seven: utilizing the camera calibration result of the step one, and obtaining the internal parameter matrix A of the left camera and the right cameral,ArAnd distortion parameter k1l,k2l,k1r,k2rConverting the right camera coordinate system into a rotation matrix R and a translation vector T of 3 x 3 of the left camera coordinate system, calculating a basic matrix F of a binocular system, and performing epipolar line correction on the left and right images of the absolute phase distribution of the ceramic core obtained in the step six; after correction, according to the epipolar constraint criterion, a certain point in the left image and a matching point in the right image are distributed on the same horizontal line, so that the matching efficiency is greatly improved; obtaining a corrected new absolute phase left and right image by a bilinear interpolation method;
step eight: performing sub-pixel level stereo matching on the left and right images subjected to polar line correction in the step seven by using a linear interpolation method based on polar line constraint and phase constraint; setting a phase area template, and obtaining an initial matching point based on a template matching method, but because a camera is an area array CCD, a certain point in a left image does not necessarily fall on the whole pixel point of the image in a right image, and sub-pixel level matching is carried out by adopting linear interpolation;
the area template settings are as follows:
let a point in the left image after polar line correction be P1With P1Centering the point, creating a 3 x 3 phase template, which is then aligned with P in the corrected right image1Searching a region with the highest similarity with the left image phase template after epipolar line correction on a straight line with the same vertical coordinate as a matching region, calculating the sum of Absolute values of differences of corresponding phase values of each pixel by using an SAD (sum of Absolute differences) algorithm as a measure of the similarity, and determining the central point of the matching region as an initial matching point P20；
Because a certain point in the left image does not necessarily fall on the whole pixel point of the image at the matching point in the right image, linear interpolation is adopted to carry out sub-pixel level matching;
let P1The dot phase value is represented as IL(xl,yl),P20The dot phase value is represented as IR(xR,yR)，P20The adjacent points of the points on the polar line are P21(xR-1,yR),P22(xR+1,yR) (ii) a Calculating P20Point and P1The phase difference of the points is positive, if the difference is positive, the sub-pixel matching point is at P21(xR-1,yR) And P20Between points; if the difference is negative, the sub-pixel matching point is at P20Point sum P22(xR+1,yR) To (c) to (d); dividing the whole pixel into 1000 equal parts, and taking P1The point with the minimum point of the point phase difference value is the final matching point P2；
Step nine: using the stereo matching result and combining the camera calibration result of the step one with the internal parameter matrix A of the left camera and the right cameral,ArAnd distortion parameter k1l,k2l,k1r,k2rThe right camera coordinate system is converted into a rotation matrix R and a translation vector T of 3 x 3 of the left camera coordinate system, three-dimensional coordinates corresponding to all the stereo matching points on the ceramic core 4 are calculated by utilizing a triangulation principle, namely, a single point cloud of the ceramic core 4 is obtained, and according to a similar triangle in the graph 3, a formula according to which the three-dimensional coordinates are calculated is as follows:
in the formula: b is the base length, f is the focal length, cc1Is the abscissa of the optical center of the left camera, c1The horizontal coordinate of the current pixel point is v, the vertical coordinate of the optical center is subtracted from the vertical coordinate of the current pixel point, d is the parallax of the space point in the left image and the right image, and d is expressed as:
d＝cc1-c1+c2-cc2
in the formula: c. Cc2Is the abscissa of the optical center of the right camera, c2The abscissa of the matching point in the right image of the current pixel point is taken as the coordinate;
step ten: respectively rotating a high-precision turntable 7 loaded with the ceramic core special fixture 5 by 90 degrees, 180 degrees and 270 degrees, repeating the second step to the eighth step to obtain single point clouds of the ceramic cores 4 at four different angles, calculating a pose transformation matrix of the point clouds at the different angles based on corner information, and realizing splicing of the point clouds at multiple angles to obtain a complete point cloud model of the ceramic cores 4;
step eleven: and transmitting the complete point cloud model of the ceramic core 4 to an industrial personal computer of a numerical control machine tool for clamping pose measurement and to-be-repaired path extraction, and guiding a laser to finish automatic repair of the ceramic core.
Claims (10)
1. A three-dimensional measurement method for a turbine blade ceramic core based on binocular structured light is characterized by comprising the following steps:
the method comprises the following steps: simultaneously shooting at least 5 calibration plate images at different angles by using a left camera and a right camera to calibrate the cameras, and respectively obtaining internal parameters and distortion parameters of the left camera and the right camera and a rotation matrix and a translation vector representing the position relationship of the left camera and the right camera;
step two: generating structural light stripes with specific patterns through computer software, projecting the structural light stripes onto the ceramic core through a structural light projection device, enabling the stripes to be deformed due to the modulation of the surface height of the ceramic core, and simultaneously acquiring deformation stripe patterns containing the surface depth characteristics of the ceramic core by a left camera and a right camera and transmitting the deformation stripe patterns to a computer;
step three: the computer carries out image preprocessing on the deformed fringe patterns acquired by the left camera and the right camera;
step four: performing phase demodulation on the left and right camera deformed fringe patterns after image preprocessing, converting light intensity information directly acquired by the left and right cameras into phase information, and obtaining wrapped phase value distribution of each point in the left and right images;
step five: calculating the modulation degree of the wrapping phase diagrams of the left camera and the right camera, carrying out image segmentation, separating the ceramic core from the background, and only extracting the wrapping phase distribution of the ceramic core;
step six: performing phase unwrapping on each point in the ceramic core wrapped phase distribution left and right images to obtain left and right images of the absolute phase distribution of the ceramic core;
step seven: utilizing the camera calibration result of the step one, and obtaining the internal parameter matrix A of the left camera and the right cameral,ArAnd distortion parameter k1l,k2l,k1r,k2rConverting the right camera coordinate system into a rotation matrix R and a translation vector T of 3 x 3 of the left camera coordinate system, performing polar line correction on the left and right images of the ceramic core absolute phase distribution obtained in the step six, and obtaining corrected new absolute phase left and right images by adopting a bilinear interpolation method;
step eight: and C, stereo matching is carried out on the new absolute phase left and right images after polar line correction in the step seven based on polar line constraint and phase constraint: setting a phase region template, obtaining an initial matching point based on a template matching method, and performing sub-pixel stereo matching by using a linear interpolation method;
step nine: using a stereo matching result, combining the camera calibration result of the step one, the internal parameters and distortion parameters of the left camera and the right camera, and the rotation matrix and translation vector representing the position relationship of the left camera and the right camera, and calculating three-dimensional coordinates corresponding to all stereo matching points on the ceramic core by utilizing a triangulation principle, namely obtaining a single point cloud of the ceramic core;
step ten: respectively rotating a high-precision turntable loaded with the special fixture for the ceramic core by 90 degrees, 180 degrees and 270 degrees, repeating the second step to the ninth step to obtain single point clouds of the ceramic cores at four different angles, calculating a pose transformation matrix of the point clouds of the cores at the different angles based on corner information, and realizing splicing of the point clouds at multiple viewing angles to obtain a complete point cloud model of the ceramic core;
step eleven: transmitting the complete point cloud model of the ceramic core to an industrial personal computer of a numerical control machine tool for clamping pose measurement and to-be-repaired path extraction, and guiding a laser to finish automatic repair of the ceramic core;
the device adopted by the binocular structured light-based three-dimensional measurement method for the ceramic core of the turbine blade comprises a structured light projection device (1), two CCD cameras, a special fixture (5) for the ceramic core, a high-precision rotary table (7) and a computer which is communicated with the structured light projection device (1) and the two CCD cameras, wherein the structured light projection device (1) and the two CCD cameras are fixed on a stand column (6) of a numerical control machine tool, and the two CCD cameras are a left camera (2) and a right camera (3); the ceramic core (4) is fixed on the high-precision rotary table (7) through the special fixture (5) for the ceramic core, and the left camera (2) and the right camera (3) are aligned to the ceramic core (4).
2. The binocular structured light-based three-dimensional measurement method for the turbine blade ceramic core, as claimed in claim 1, wherein in the first step, the left and right cameras are calibrated by using a Zhang Yong plane calibration method, 15 images of different positions of a calibration plate are taken, and internal parameter matrixes A of the left and right cameras are respectively obtainedl,ArDistortion coefficient k1l,k2l,k1r,k2rAnd a rotation matrix R and a translational vector T characterizing the positional relationship of the left camera.
3. The binocular structured light-based three-dimensional measurement method for the turbine blade ceramic core, as claimed in claim 1, wherein in the second step, the structured light stripes with specific patterns are projected as blue light sine stripes, and the stripes are collected while ensuring continuous change of the gray scale of the stripes.
4. The binocular structured light-based three-dimensional measurement method for the turbine blade ceramic core is characterized in that the method for preprocessing the deformed fringe images acquired by the left camera and the right camera in the third step is Gaussian filtering, and random noise interference is reduced.
5. The binocular structured light-based three-dimensional measurement method for the turbine blade ceramic core, as claimed in claim 1, wherein the phase demodulation in the four steps is performed by a four-step phase shift method to solve the phase information, specifically, the phase demodulation is performed on the left and right images by the following formula:
in the formula:
the phase value of the envelope is distributed in (-pi, pi)]；I1、I2、I3、I4The light intensity distribution of the deformed fringe pattern acquired by the four-step phase-shift CCD camera is shown.
6. The binocular structured light-based three-dimensional measurement method for the turbine blade ceramic core, as claimed in claim 1, wherein the fifth step is to perform image segmentation on the left and right cameras by calculating modulation parameters, and separate the ceramic core from the background; calculating a modulation parameter, wherein the modulation parameter is defined as:
in the formula: n is the total number of phase shift steps, I represents the number of phase shift steps, IiRepresenting the gray scale of the fringe image corresponding to the ith step phase shift; the modulation degree calculated by the stripes on the ceramic core is higher than the background, and a threshold value is set to separate the modulation degree from the background; an iterative approach is used to solve for the optimal threshold.
7. The binocular structured light-based three-dimensional measurement method for the turbine blade ceramic core, as claimed in claim 1, wherein in the sixth step, the wrapping phase values of each point in the left and right images are subjected to phase unwrapping by using a multi-frequency heterodyne method, wherein the frequency combination of unwrapping by using the multi-frequency heterodyne method is selected from 74, 68 and 63, and left and right images of the absolute phase distribution of the ceramic core are obtained.
8. The binocular structured light-based three-dimensional measurement method for the turbine blade ceramic core, as claimed in claim 1, wherein in the seventh step, epipolar correction is performed on the left and right images of the absolute phase distribution of the ceramic core, the corrected system structure is a head-up binocular standard geometry, so that a certain point in the left image is certainly distributed on the same horizontal line at the matching point in the right image, and a new absolute phase is calculated for the corrected left and right camera phase maps by using a bilinear interpolation method.
9. The binocular structured light-based three-dimensional measurement method for the turbine blade ceramic core, as claimed in claim 1, wherein the eighth step sets a phase region template by using epipolar constraint and phase constraint, obtains an initial matching point based on a template matching method, and performs sub-pixel level matching by using linear interpolation;
the area template settings are as follows:
let a point in the left image after polar line correction be P1With P1Centering the point, creating a 3 x 3 phase template, which is then aligned with P in the corrected right image1Searching a region with the highest similarity with the epipolar line corrected left image phase template on a straight line with the same vertical coordinate as a matching region, and calculating the similarity by adopting an SAD (sum of absolute differences) algorithmThe sum of absolute values of the difference between the phase values corresponding to each pixel is used as a measure, and the central point of the matching area is the initial matching point P20；
Because a certain point in the left image does not necessarily fall on the whole pixel point of the image at the matching point in the right image, linear interpolation is adopted to carry out sub-pixel level matching;
let P1The dot phase value is represented as IL(xl,yl),P20The dot phase value is represented as IR(xR,yR)，P20The adjacent points of the points on the polar line are P21(xR-1,yR),P22(xR+1,yR) (ii) a Calculating P20Point and P1The phase difference of the points is positive, if the difference is positive, the sub-pixel matching point is at P21(xR-1,yR) And P20Between points; if the difference is negative, the sub-pixel matching point is at P20Point sum P22(xR+1,yR) To (c) to (d); dividing the whole pixel into 1000 equal parts, and taking P1The point with the minimum point of the point phase difference value is the final matching point P2。
10. The binocular structured light-based three-dimensional measurement method for the turbine blade ceramic core according to claim 1, wherein the nine steps are based on the following formula when calculating the three-dimensional coordinates:
in the formula: b is the base length, f is the focal length, cc1Is the abscissa of the optical center of the left camera, c1The horizontal coordinate of the current pixel point is v, the vertical coordinate of the optical center is subtracted from the vertical coordinate of the current pixel point, d is the parallax of the space point in the left image and the right image, and d is expressed as:
d＝cc1-c1+c2-cc2
in the formula: c. Cc2Is the abscissa of the optical center of the right camera, c2And the abscissa of the matching point in the right image of the current pixel point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010025374.4A CN111207693A (en) | 2020-01-10 | 2020-01-10 | Three-dimensional measurement method of turbine blade ceramic core based on binocular structured light |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010025374.4A CN111207693A (en) | 2020-01-10 | 2020-01-10 | Three-dimensional measurement method of turbine blade ceramic core based on binocular structured light |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111207693A true CN111207693A (en) | 2020-05-29 |
Family
ID=70788881
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010025374.4A Pending CN111207693A (en) | 2020-01-10 | 2020-01-10 | Three-dimensional measurement method of turbine blade ceramic core based on binocular structured light |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111207693A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111473744A (en) * | 2020-06-03 | 2020-07-31 | 北京航空航天大学 | Three-dimensional shape vision measurement method and system based on speckle embedded phase shift stripe |
| CN111750805A (en) * | 2020-07-06 | 2020-10-09 | 山东大学 | Three-dimensional measuring device and method based on binocular camera imaging and structured light technology |
| CN111862193A (en) * | 2020-07-21 | 2020-10-30 | 太仓光电技术研究所 | Binocular vision positioning method and device for electric welding spots based on shape descriptors |
| CN111899326A (en) * | 2020-06-18 | 2020-11-06 | 苏州小优智能科技有限公司 | Three-dimensional reconstruction method based on GPU parallel acceleration |
| CN111932632A (en) * | 2020-07-13 | 2020-11-13 | 太仓中科信息技术研究院 | Phase correction method in three-dimensional reconstruction of mechanical part |
| CN112102456A (en) * | 2020-08-20 | 2020-12-18 | 佛山科学技术学院 | Ceramic wafer height detection method and device and computer readable storage medium |
| CN112325799A (en) * | 2021-01-07 | 2021-02-05 | 南京理工大学智能计算成像研究院有限公司 | High-precision three-dimensional face measurement method based on near-infrared light projection |
| CN112990373A (en) * | 2021-04-28 | 2021-06-18 | 四川大学 | Convolution twin point network blade profile splicing system based on multi-scale feature fusion |
| CN113916153A (en) * | 2021-10-12 | 2022-01-11 | 深圳市其域创新科技有限公司 | Active and passive combined structured light three-dimensional measurement method |
| CN114067078A (en) * | 2021-11-19 | 2022-02-18 | 天津理工大学 | Point cloud registration method for accurately measuring pose of ceramic sanitary appliance |
| CN114972544A (en) * | 2022-07-28 | 2022-08-30 | 星猿哲科技(深圳)有限公司 | Method, device and equipment for self-calibration of external parameters of depth camera and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8922647B2 (en) * | 2011-08-03 | 2014-12-30 | The Boeing Company | Projection aided feature measurement using uncalibrated camera |
| CN104331897A (en) * | 2014-11-21 | 2015-02-04 | 天津工业大学 | Polar correction based sub-pixel level phase three-dimensional matching method |
| CN105953747A (en) * | 2016-06-07 | 2016-09-21 | 杭州电子科技大学 | Structured light projection full view three-dimensional imaging system and method |
-
2020
- 2020-01-10 CN CN202010025374.4A patent/CN111207693A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8922647B2 (en) * | 2011-08-03 | 2014-12-30 | The Boeing Company | Projection aided feature measurement using uncalibrated camera |
| CN104331897A (en) * | 2014-11-21 | 2015-02-04 | 天津工业大学 | Polar correction based sub-pixel level phase three-dimensional matching method |
| CN105953747A (en) * | 2016-06-07 | 2016-09-21 | 杭州电子科技大学 | Structured light projection full view three-dimensional imaging system and method |
Non-Patent Citations (2)
| Title |
|---|
| 余良伟: "离心泵叶轮的结构光测量技术研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
| 吴琳琳: "基于双目立体视觉的三维测量系统的实现", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111473744A (en) * | 2020-06-03 | 2020-07-31 | 北京航空航天大学 | Three-dimensional shape vision measurement method and system based on speckle embedded phase shift stripe |
| CN111899326A (en) * | 2020-06-18 | 2020-11-06 | 苏州小优智能科技有限公司 | Three-dimensional reconstruction method based on GPU parallel acceleration |
| CN111750805B (en) * | 2020-07-06 | 2021-12-10 | 山东大学 | Three-dimensional measuring device and method based on binocular camera imaging and structured light technology |
| CN111750805A (en) * | 2020-07-06 | 2020-10-09 | 山东大学 | Three-dimensional measuring device and method based on binocular camera imaging and structured light technology |
| CN111932632A (en) * | 2020-07-13 | 2020-11-13 | 太仓中科信息技术研究院 | Phase correction method in three-dimensional reconstruction of mechanical part |
| CN111862193A (en) * | 2020-07-21 | 2020-10-30 | 太仓光电技术研究所 | Binocular vision positioning method and device for electric welding spots based on shape descriptors |
| CN112102456A (en) * | 2020-08-20 | 2020-12-18 | 佛山科学技术学院 | Ceramic wafer height detection method and device and computer readable storage medium |
| CN112102456B (en) * | 2020-08-20 | 2023-12-26 | 佛山科学技术学院 | Ceramic wafer height detection method and device and computer readable storage medium |
| CN112325799A (en) * | 2021-01-07 | 2021-02-05 | 南京理工大学智能计算成像研究院有限公司 | High-precision three-dimensional face measurement method based on near-infrared light projection |
| CN112990373B (en) * | 2021-04-28 | 2021-08-03 | 四川大学 | Convolution twin point network blade profile splicing system based on multi-scale feature fusion |
| CN112990373A (en) * | 2021-04-28 | 2021-06-18 | 四川大学 | Convolution twin point network blade profile splicing system based on multi-scale feature fusion |
| CN113916153A (en) * | 2021-10-12 | 2022-01-11 | 深圳市其域创新科技有限公司 | Active and passive combined structured light three-dimensional measurement method |
| CN113916153B (en) * | 2021-10-12 | 2023-11-28 | 深圳市其域创新科技有限公司 | Active and passive combined structured light three-dimensional measurement method |
| CN114067078A (en) * | 2021-11-19 | 2022-02-18 | 天津理工大学 | Point cloud registration method for accurately measuring pose of ceramic sanitary appliance |
| CN114067078B (en) * | 2021-11-19 | 2024-04-30 | 天津理工大学 | Point cloud registration method for accurately measuring pose of ceramic sanitary ware |
| CN114972544A (en) * | 2022-07-28 | 2022-08-30 | 星猿哲科技(深圳)有限公司 | Method, device and equipment for self-calibration of external parameters of depth camera and storage medium |
| CN114972544B (en) * | 2022-07-28 | 2022-10-25 | 星猿哲科技(深圳)有限公司 | Method, device and equipment for self-calibration of external parameters of depth camera and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111207693A (en) | Three-dimensional measurement method of turbine blade ceramic core based on binocular structured light | |
| CN109242828B (en) | Three-dimensional defect detection method for 3D printed product based on grating projection multistep phase shift method | |
| CN110288642B (en) | Three-dimensional object rapid reconstruction method based on camera array | |
| CN105783775B (en) | A kind of minute surface and class minute surface object surface appearance measuring device and method | |
| CN110514143B (en) | Stripe projection system calibration method based on reflector | |
| CN110296667B (en) | High-reflection surface three-dimensional measurement method based on line structured light multi-angle projection | |
| CN110207614B (en) | High-resolution high-precision measurement system and method based on double telecentric camera matching | |
| CN100561120C (en) | A kind of formation method of three-dimension measuring system | |
| CN109648202B (en) | Additive manufacturing forming precision control method for non-flat surface autonomous recognition robot | |
| CN113205592B (en) | Light field three-dimensional reconstruction method and system based on phase similarity | |
| CN112465912A (en) | Three-dimensional camera calibration method and device | |
| CN111353997B (en) | Real-time three-dimensional surface defect detection method based on fringe projection | |
| CN108269289A (en) | A kind of two step optimization methods of camera parameter calibration | |
| Feng et al. | Repair volume extraction method for damaged parts in remanufacturing repair | |
| CN111256587A (en) | High-reflectivity surface three-dimensional measurement method based on double-line structured light scanning | |
| CN114972538A (en) | Thickness measuring device and method of flattened ultrathin heat pipe based on binocular structured light | |
| Apolinar et al. | Three-dimensional microscope vision system based on micro laser line scanning and adaptive genetic algorithms | |
| CN114170284B (en) | Multi-view point cloud registration method based on active landmark point projection assistance | |
| CN117450955B (en) | Three-dimensional measurement method for thin object based on space annular feature | |
| Li et al. | Fast phase-based stereo matching method for 3D shape measurement | |
| CN115855955A (en) | Mold surface structure defect detection device and method based on multi-beam laser | |
| Wang et al. | Light-field-assisted phase unwrapping of fringe projection profilometry | |
| Hafeez et al. | The effect of patterns on image-based modelling of texture-less objects | |
| CN116428982A (en) | Fiber ring skeleton non-contact measurement system and method based on line structured light | |
| CN113298882B (en) | Camera calibration device and method for microscopic three-dimensional topography measurement 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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200529 |
|
| WD01 | Invention patent application deemed withdrawn after publication |