Wire structured light bearing and gear three-dimensional measurement method and system

Abstract

The present application relates to the technical field of three-dimensional measurement, in particular to a line structure light bearing and gear three-dimensional measurement method and system, the method comprises the following steps: step one, data acquisition: using the CCD camera provided on the measuring platform to shoot laser images containing laser line patterns from different angles. Step two, image processing: using computer vision algorithms to extract feature points in the laser images. Step three, three-dimensional reconstruction: based on the known CCD camera parameters and the position changes of the feature points on the object to be measured, using spatial coordinate conversion technology to obtain the three-dimensional data of the object to be measured. Step four, error correction: error analysis is performed on the obtained three-dimensional data. Step five, result output: based on the corrected results, the final three-dimensional measurement data of the object to be measured is output. The present application can further shorten the measurement period and improve the production efficiency through automatic and intelligent measurement means.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of three-dimensional measurement, in particular to a line structured light bearing and gear three-dimensional measurement method and system. BACKGROUND

[0002] Line structured light technology is a non-contact three-dimensional measurement technology. Its basic principle is to project one or more laser lines onto the surface of an object and use a camera to capture the deformed patterns formed by the laser lines on the object surface. By analyzing the changes in these patterns, the three-dimensional information of the object surface can be calculated.

[0003] In the prior art, point and line measurements are usually used, which cannot fully reflect the overall appearance of the object to be measured, and the measurement efficiency is low. For objects with complex geometric shapes, some parts may not be directly measured due to occlusion. For high reflectivity or smooth surfaces, laser lines may be totally reflected, making it difficult to identify feature points. Strong ambient light may interfere with the visibility of laser lines, affecting measurement accuracy.

[0004] In summary, how to solve the problem of the prior art that point and line measurements are usually used, which cannot fully reflect the overall appearance of the object to be measured, and the measurement efficiency is low, has become a difficult problem that needs to be solved in the field at present. Therefore, it is necessary to propose a line structured light bearing and gear three-dimensional measurement method and system that can fully reflect the overall appearance of the object and has higher efficiency. SUMMARY

[0005] To solve the above problems, the present application provides a line structured light bearing and gear three-dimensional measurement method and system, which automatically marks the positions of alternating points through image processing technology, thereby obtaining the view point of the special-shaped object to be measured. More complete information is generated using multi-directional data acquisition. Through automated and intelligent measurement means, the measurement cycle can be further shortened and production efficiency can be improved.

[0006] In order to achieve the above purpose, the technical scheme of the present application is as follows: a line structured light bearing and gear three-dimensional measurement method, comprising the following steps:

[0007] Step 1: Data acquisition: place the object to be measured on the measurement platform, project structured laser lines on the object to be measured through the laser provided on the measurement platform; and use the CCD camera provided on the measurement platform to capture laser images containing laser line patterns from different angles.

[0008] Step 2: Image processing: use computer vision algorithms to extract feature points in the laser images and track the position changes of these feature points in different laser images.

[0009] Step three, three-dimensional reconstruction: based on the known CCD camera parameters and the position changes of the feature points on the object to be measured, a three-dimensional reconstruction algorithm is used to calculate the spatial coordinates of each point on the surface of the object to be measured, and a spatial coordinate conversion technique is used to obtain the three-dimensional data of the object to be measured.

[0010] Step four, error correction: error analysis is performed on the obtained three-dimensional data; if the measuring points are blocked, the measuring point positions are adjusted according to the analysis results; if the measuring points are disturbed by external light, the light compensation lamp provided on the measuring platform is used for correction; if the object to be measured has special features, the measuring points of the object to be measured are captured by judging the shadow alternation points of the light.

[0011] Step five, result output: based on the corrected results, the final three-dimensional measurement data of the object to be measured is output.

[0012] Further, in step two, the computer vision algorithm extracts the feature points in the laser image, including the following steps:

[0013] S101, image preprocessing: convert the color laser image to a grayscale image; use a filtering algorithm to remove noise in the grayscale image; and enhance the contrast of the grayscale image by histogram equalization method.

[0014] S102, edge detection: use an edge detection algorithm to determine the edge position in the grayscale image by calculating the gradient direction and size.

[0015] S103, feature point detection: use a corner detection method to detect feature points in the grayscale image; calculate the local feature descriptors around the detected feature points.

[0016] S104, feature matching: use feature descriptors to match feature points between grayscale images taken at different angles.

[0017] S105, feature point optimization: use an algorithm to remove false matching points and optimize the feature point set.

[0018] Further, in step three, the spatial coordinate conversion technique includes the following steps:

[0019] S201, camera pose estimation: use the matched feature points to estimate the initial position and attitude of the camera when each laser image is taken by an algorithm; and use an optimization algorithm to jointly optimize the pose of the camera.

[0020] S202, three-dimensional coordinate calculation: use multi-view geometry to perform triangulation based on the internal and external parameters of the camera and the matched feature points, calculate the coordinates of the feature points in three-dimensional space; and convert the calculated three-dimensional point coordinates to a unified global coordinate system.

[0021] S203, stereo matching: by comparing the position difference of corresponding feature points in two laser images, combined with the baseline length and focal length information of the camera, the depth information of each point on the object surface is calculated, and the three-dimensional coordinates are obtained.

[0022] S204, structured light encoding: in the structured light system, by projecting a specific encoding pattern, the spatial information of each point on the object surface is decoded from the laser image.

[0023] Further, in step four, the processing steps of the measuring point with occlusion are as follows:

[0024] S301, occlusion detection: using error calculation method to analyze the error of three-dimensional data, on the basis of error analysis, identify the inaccurate or missing data of measuring points caused by occlusion.

[0025] S302, occlusion area filling: using multi-view fusion technology, the data under different angles are merged to fill the occlusion area.

[0026] S303, interpolation and filling: for the occluded measuring points under a single angle, use interpolation method to estimate the missing data; use surface reconstruction algorithm to fill the vacancy area.

[0027] S304, iterative optimization: using iterative algorithm to adjust the three-dimensional model, through multiple iterations, the difference between the reconstructed model and the actual object is reduced.

[0028] Further, in step four, the processing steps of the measuring point disturbed by external light are as follows:

[0029] S401, environment evaluation: by comparing the position change of measuring points and the matching rate of feature points under different light conditions, evaluate the interference degree of external light on the accuracy of measuring point data, and judge the back light and front light under different light.

[0030] S402, automatic adjustment: automatically adjust the position, angle and brightness of the fill light according to the position of the measuring point and the direction of the light disturbance.

[0031] S403, reacquisition: after using the fill light to illuminate the measuring point, reacquire the laser image of the object to be measured; and perform feature extraction and matching on the reacquired laser image.

[0032] Further, in step four, the processing steps of the measured object with special features are as follows:

[0033] S501, observe the shadow: use CCD camera to capture shadow image of the measured object under light source from different angles.

[0034] S502, identify alternating points: find the alternating points of shadow and light in the shadow image; the alternating points correspond to the special features of the measured object.

[0035] S503, recording position: automatically marking the alternating point position by image processing technology; the alternating point position corresponds to the depth information of the object to be measured; the view point of the special object to be measured is obtained by recording the alternating point position; if there is no alternating point, it is judged that there is no shadow, and it is judged as a vertical image.

[0036] S504, multi-directional acquisition: according to the alternating point position change, the laser image of the object to be measured is photographed for many times.

[0037] Further, the measuring platform comprises a laser, a CCD camera, a detection plate, a mechanical arm and a controller; the detection plate is circular, and the bottom of the detection plate is provided with a driving assembly for driving the mechanical arm to rotate.

[0038] The driving assembly comprises a driving motor fixedly connected to the bottom of the detection plate; a rotating ring is fixedly connected to the output shaft of the driving motor and rotationally fitted to the outer side of the detection plate; the mechanical arm is fixedly connected to the top of the rotating ring, and the controller is used for controlling the driving motor and the mechanical arm to operate according to the position of the object to be measured; the CCD camera is fixedly connected to the output shaft of the mechanical arm, and the laser is fixedly connected to the top of the detection plate; the controller is used for controlling the CCD camera and the laser to operate; a telescopic rod is fixedly connected to the side of the mechanical arm close to the CCD camera, and a light supplement lamp is fixedly connected to the output shaft of the telescopic rod; the controller is used for controlling the telescopic rod and the light supplement lamp to operate according to the position needing light supplement.

[0039] Further, the line structure light bearing and gear three-dimensional measurement system is suitable for the line structure light bearing and gear three-dimensional measurement method described in any one of the above, comprising the following modules:

[0040] The data acquisition module is used for capturing the laser image formed by the laser line projected on the surface of the object to be measured under different angles.

[0041] The image processing module uses an algorithm to extract the contour information of the laser line formed on the surface of the object to be measured from the collected laser image, and identifies the position of the laser line in the laser image and extracts the feature points.

[0042] The three-dimensional reconstruction module is used for calculating the three-dimensional coordinates of each point on the surface of the object to be measured according to the position information of the feature points.

[0043] The error correction module is used for error analysis on the generated three-dimensional coordinates, and correction is made on the laser image of the object to be measured with shielding, light interference and special features according to the analysis results.

[0044] The intelligent control module is used for controlling the operation of the laser, the driving motor, the mechanical arm, the CCD camera, the telescopic rod and the light supplement lamp according to the imaging condition of the object to be measured.

[0045] A result output module is configured to output the final three-dimensional measurement data of the object to be measured.

[0046] Further, the object to be measured includes but is not limited to a bearing bush and a gear.

[0047] Further, the special-shaped features include but are not limited to an oil groove, an oil hole, a conical surface, an end face flange, a reinforcing rib, a tooth shape, a tooth groove, a non-standard aperture, a keyway, a tooth tip, and a tooth root.

[0048] The above scheme has the following beneficial effects:

[0049] 1. The scheme can capture the details of the surface of the object to be measured by projecting a structured laser line and multi-angle shooting of a CCD camera, so that the result of three-dimensional reconstruction is more accurate. In the case of a bearing bush and a gear being blocked, the feature points missing or inaccurate at certain angles are identified through feature point matching and multi-view image comparison. The corresponding relationship between different angles is determined through a multi-view geometry method using image data at different angles, the data at different angles is fused, and the blocked area is filled. For the blocked measurement points at a single angle, an interpolation method is used to estimate the missing data; based on the data of the surrounding known points, the points in the blocked area are predicted through a mathematical function. An optimization algorithm is used to adjust the three-dimensional model through multiple iterations so that it is closer to the shape of the actual object; thereby improving the overall measurement accuracy and the quality of the model.

[0050] 2. The scheme reduces the interference of external light on the measurement points and improves the measurement accuracy by evaluating and adjusting the light conditions; evaluates the backlight and frontlight conditions under different light conditions, analyzes the influence of light direction on feature point detection, automatically adjusts the position of the fill light according to the position of the measurement points and the direction of light interference, and ensures that the light can uniformly illuminate the measurement point area. The movement of the fill light is controlled by the mechanical arm so that it can adapt to measurement points at different positions, the angle of the fill light is automatically adjusted to avoid strong reflection or shadow, and the visibility of the feature points is improved. The laser image of the object to be measured is reacquired using the adjusted fill light to ensure that the light conditions are suitable for feature point detection. In different measurement environments, the system can adapt to more complex working conditions by dynamically adjusting the fill light, thereby expanding the application range.

[0051] 3. The scheme finds the alternating points of shadows and highlights in the collected shadow images; these alternating points usually correspond to the positions of the mutations on the surface of the object. The positions of the alternating points are automatically marked through image processing technology, thereby obtaining the view points of the special-shaped object to be measured, these positions correspond to the depth information of the surface of the object to be measured, thereby improving the accuracy of three-dimensional reconstruction. The information deficiency under a single angle is made up using multi-directional collected data, a more complete three-dimensional model information is generated, and for objects such as bearing bushes and gears with complex geometric shapes, this method can better capture the special-shaped features and improve the applicability of measurement.

[0052] 4、The scheme finds corresponding feature points in multiple laser images, ensures that the feature points have consistency under different viewing angles, estimates the initial position and attitude of the camera when shooting each laser image by using an algorithm, and jointly optimizes the pose of the camera to improve the accuracy of pose estimation.

[0053] 5、The scheme can remove noise and enhance contrast through image preprocessing, so that feature points are more easily detected. Based on the matched feature points, the coordinates of the feature points in three-dimensional space can be accurately calculated by triangulation using multi-view geometry. Feature matching can ensure that the feature points in different images have consistency, and feature point optimization can remove false matching points to improve the accuracy of the three-dimensional model. Through feature matching of multi-view images, measurement errors caused by occlusion and changes in illumination are reduced. Through automated and intelligent measurement means, the measurement cycle can be further shortened and production efficiency can be improved.

[0054] Additional aspects and advantages of the present application will be partially given in the following description, partially will become obvious from the following description, or will be understood by practicing the present application. BRIEF DESCRIPTION OF DRAWINGS

[0055] Figure 1 It is a flowchart of the three-dimensional measurement method of the linear structured light shaft bearing and gear in the embodiment of the present application.

[0056] Figure 2 It is a flowchart of the computer vision algorithm in the embodiment of the present application.

[0057] Figure 3 It is a flowchart of the space coordinate conversion technology in the embodiment of the present application.

[0058] Figure 4 It is a flowchart of the processing steps of the measurement point with occlusion in the embodiment of the present application.

[0059] Figure 5 It is a flowchart of the processing steps of the measurement point interfered by external light in the embodiment of the present application.

[0060] Figure 6 It is a flowchart of the processing steps of the measured object with special features in the embodiment of the present application.

[0061] Figure 7The structural block diagram of the line structured light bearing and gear three-dimensional measurement system in the embodiment of the present application.

[0062] Figure 8 The axial view of the measurement platform in the embodiment of the present application.

[0063] The reference signs in the drawings of the specification include: 1, laser; 2, CCD camera; 3, detection plate; 4, mechanical arm; 5, driving motor; 6, rotating ring; 7, telescopic rod; 8, light supplement lamp. DETAILED DESCRIPTION

[0064] The following is further described in detail through specific embodiments:

[0065] Embodiment 1

[0066] As shown in the accompanying Figures 1-6 : line structured light bearing and gear three-dimensional measurement method, including the following steps:

[0067] Step one, data acquisition: place the bearing or gear on the measurement platform, project the structured laser line on the bearing or gear through the laser 1 provided on the measurement platform; and use the CCD camera 2 provided on the measurement platform to shoot the laser image containing the laser line pattern from different angles.

[0068] Step two, image processing: use computer vision algorithm to extract feature points in the laser image, and track the position changes of these feature points in different laser images.

[0069] Among them, as Figure 2 shown, the computer vision algorithm extracts the feature points in the image, including the following steps:

[0070] S101, image preprocessing: convert the color laser image to a gray image; use the median filter algorithm to remove noise in the gray image; and enhance the contrast of the gray image through the histogram equalization method.

[0071] S102, edge detection: use the Canny edge detection algorithm to determine the edge position in the gray image by calculating the gradient direction and size.

[0072] S103, feature point detection: use the Harris corner point detection method to detect the feature points in the gray image; calculate the local feature descriptors around the detected feature points, which can capture the detailed information around the feature points and improve the stability of the feature points.

[0073] S104, feature matching: use the feature descriptors to match the feature points between the gray images shot from different angles, to improve the reliability of the matching.

[0074] S105, Feature point optimization: using RANSAC algorithm to remove false matching points, optimizing the feature point set, and improving the positioning accuracy of the feature points.

[0075] Step three, three-dimensional reconstruction: based on the known CCD camera 2 parameters and the position change of the feature points on the object to be measured, using three-dimensional reconstruction algorithm to calculate the spatial coordinates of each point on the surface of the object to be measured, and using spatial coordinate conversion technology to obtain the three-dimensional data of the object to be measured.

[0076] As shown in Figure 3 , the spatial coordinate conversion technology includes the following steps:

[0077] S201, Camera pose estimation: using the matched feature points to estimate the initial position and attitude of the camera at the time of shooting each laser image through the SIFT algorithm; and using the Levenberg-Marquardt optimization algorithm to jointly optimize the pose of the camera.

[0078] S202, Three-dimensional coordinate calculation: according to the internal and external parameters of the camera focal length, principal point position and distortion coefficient, and the matched feature points, using multi-view geometric principle to perform triangulation, calculating the coordinates of the feature points in three-dimensional space; and converting the calculated three-dimensional point coordinates into a unified global coordinate system.

[0079] S203, Stereo matching: by comparing the position difference of corresponding feature points in two laser images, combining the camera baseline length and focal length information, calculating the depth information of each point on the object surface, and obtaining the three-dimensional coordinates.

[0080] S204, Structured light coding: in the structured light system, by projecting a specific coding pattern, the spatial information of each point on the object surface is decoded from the laser image, which adapts to the measurement needs of objects with different shapes, materials and surface characteristics.

[0081] Step four, error correction: error analysis is performed on the obtained three-dimensional data; if the measuring point is blocked, the measuring point position is adjusted according to the analysis result; if the measuring point is disturbed by external light, the light supplement lamp 8 provided on the measuring platform is used for correction; if the object to be measured has special features, the measuring point of the object to be measured is captured by judging the shadow alternation point of the light.

[0082] As shown in Figure 4 , the processing steps of the measuring point with occlusion are as follows:

[0083] S301, Occlusion detection: using error calculation method to analyze the error of three-dimensional data, and on the basis of error analysis, identifying the situation that the measuring point data is inaccurate or missing due to occlusion. For example, finding consistent and inconsistent places in the depth map taken from different angles.

[0084] S302, Occluded area filling: Use multi-view fusion technology to combine data from different angles to fill in the occluded area. For example, if a part is occluded in one image but not in another, the latter data can be used to fill in the missing part.

[0085] S303, Interpolation and filling: For occluded points in a single view, use polynomial interpolation method to estimate the missing data; use Delaunay triangulation surface reconstruction algorithm to fill in the gaps.

[0086] S304, Iterative optimization: Adjust the three-dimensional model using the Iterative Closest Point (ICP) algorithm, through multiple iterations, to narrow the gap between the reconstructed model and the actual object, making it closer to the actual object shape.

[0087] As shown in Figure 5 , the processing steps of the measuring points disturbed by external light are as follows:

[0088] S401, Environment evaluation: By comparing the position changes of measuring points under different light conditions and the matching rate of feature points, evaluate the degree of disturbance of external light on the accuracy of measuring point data, judge the back light and front light under different light, including natural light, indoor light, strong light and weak light in various environments.

[0089] S402, Automatic adjustment: According to the position of the measuring point and the direction of the light disturbance, automatically adjust the position, angle and brightness of the fill light 8, to ensure that the light can uniformly illuminate the measuring point area, avoid direct illumination of the light source to the camera lens, and prevent glare or reflection.

[0090] S403, Re-acquisition: After using the fill light 8 to illuminate the measuring points, re-acquire the laser image of the object to be measured; and use the FLANN algorithm to extract and match the features of the re-acquired laser image to ensure that the feature points are clearly visible.

[0091] As shown in Figure 6 , the processing steps of the measured object with special features are as follows:

[0092] For example, the object to be measured is a gear with standard module, which has a circular hole and uniformly distributed teeth.

[0093] S501, Observe the shadow: Use the CCD camera 2 to capture the shadow image of the gear under the illumination of the light source from different angles, ensure that the position and intensity of the light source are moderate, and the shadow changes of the gear teeth and the circular hole edge can be clearly displayed.

[0094] S502, Identify alternating points: Find the alternating points of the gear tip, tooth root and circular hole shadow and light in the shadow image; the alternating points correspond to the special features of the object to be measured, and this method is very effective for capturing the details of edges, protrusions and depressions.

[0095] S503, record position: automatically mark the alternating point position by image processing technology; the alternating point position corresponds to the depth information of the gear surface; the view point of the profile gear is obtained by recording the alternating point position; if there is no alternating point, it means no shadow, and it is judged as a vertical image; by recording the position of the alternating point, the depth information of the gear surface can be obtained, thereby improving the accuracy of three-dimensional reconstruction.

[0096] S504, multi-directional collection: according to the alternating point position change, the laser image of the gear is photographed for several times to ensure that all tooth shapes and circular hole edge details of the gear surface are captured from different angles.

[0097] Step five, result output: based on the corrected result, the three-dimensional measurement data of the final measured object is output. The whole measurement process is operated intelligently and automatically, reducing the error caused by human operation. At the same time, the accuracy of computer vision algorithm and three-dimensional reconstruction algorithm also further reduces the error of measurement result.

[0098] Example 2:

[0099] As shown in the accompanying Figure 8 , the difference from the above embodiment is that the measurement platform comprises a laser 1, a CCD camera 2, a detection plate 3, a mechanical arm 4 and a controller; the detection plate 3 is circular, and the bottom of the detection plate 3 is provided with a driving assembly for driving the mechanical arm 4 to rotate.

[0100] The driving assembly comprises a driving motor 5, the driving motor 5 is bolted and connected to the bottom of the detection plate 3; the output shaft of the driving motor 5 is bolted and connected with a rotating ring 6, the rotating ring 6 is rotationally fitted on the outside of the detection plate 3; the mechanical arm 4 is bolted and connected to the top of the rotating ring 6, and the controller is used to control the driving motor 5 and the mechanical arm 4 to operate according to the position of the measured object; the CCD camera 2 is bolted and connected to the output shaft of the mechanical arm 4, the laser 1 is bolted and connected to the top of the detection plate 3, and the controller is used to control the operation of the CCD camera 2 and the laser 1; the telescopic rod 7 is bolted and connected to the side of the mechanical arm 4 close to the CCD camera 2, and the output shaft of the telescopic rod 7 is bolted and connected with the light supplement lamp 8; the controller is used to control the operation of the telescopic rod 7 and the light supplement lamp 8 according to the position of the light supplement.

[0101] The specific implementation process is as follows: when three-dimensional measurement is required, first place the bearing or gear on the detection plate 3, turn on the laser 1 through the controller, and project the laser line onto the surface of the object. Since the driving motor 5 is bolted to the bottom of the detection plate 3, the output shaft of the driving motor 5 is bolted to the rotating ring 6, the rotating ring 6 is rotatably connected to the outside of the detection plate 3, and the mechanical arm 4 is bolted to the top of the rotating ring 6, so the driving motor 5 can drive the rotating ring 6 to rotate around the detection plate 3, and the rotating ring 6 drives the mechanical arm 4 to rotate. In this process, the controller is used to accurately control the position of the CCD camera 2 according to the position of the object to be detected, so as to fully capture the image information of the object. When there is insufficient light or the light is blocked, since the mechanical arm 4 is bolted to the side near the CCD camera 2, the output shaft of the telescopic rod 7 is bolted to the light supplement lamp 8, so the controller can adjust the position and brightness of the light supplement lamp 8 to reduce the situation that the object imaging is insufficient when the light is blocked or disturbed.

[0102] The controller controls the driving motor 5 and the mechanical arm 4 to operate according to the position information of the object to be measured, adjusts the position and angle of the CCD camera 2, and ensures that the laser line can be accurately projected onto the surface of the object to be measured. Subsequently, an image processing algorithm is used to extract feature points in the image and calculate the coordinates of the feature points in three-dimensional space. The use of the light supplement lamp 8 can improve the image quality, and the combination of the mechanical arm 4 and the driving assembly can adapt to different sizes and shapes of the object to be measured, enhancing the adaptability of the system. Through the automatic control mechanism, the influence of human factors can be reduced, and the consistency and reliability of the measurement can be improved.

[0103] Example 3:

[0104] As shown in the accompanying Figure 7 The difference between the above embodiment and the line structured light bearing and gear three-dimensional measurement system is that it includes the following modules:

[0105] The data acquisition module is used to capture the laser image formed by the laser line projected on the surface of the bearing and gear at different angles, to ensure that the information on the surface of the object can be captured from different angles, and to reduce the influence of the blockage.

[0106] The image processing module uses the Canny edge detection algorithm to extract the contour information of the laser line formed on the surface of the object from the collected laser image. The position of the laser line in the laser image is identified, and the feature points are extracted to improve the accuracy of subsequent three-dimensional reconstruction.

[0107] The three-dimensional reconstruction module is used to calculate the three-dimensional coordinates of each point on the surface according to the position information of the feature points, and to ensure the accuracy of the measurement results.

[0108] The error correction module is used for error analysis on the generated three-dimensional coordinates, and corrects the laser images of the object to be measured with shielding, light interference and special features according to the analysis results, so as to reduce the measurement error and improve the measurement accuracy.

[0109] The special features include, but are not limited to, oil groove, oil hole, conical surface, end flange, reinforcing rib, tooth shape, tooth groove, non-standard aperture, keyway, tooth crest and tooth root, and the system can collect different special features.

[0110] The intelligent control module is used for controlling the operation of the laser 1, the driving motor 5, the mechanical arm 4, the CCD camera 2, the telescopic rod 7 and the light supplement lamp 8 according to the imaging condition of the object to be measured, so as to realize the automation of the measurement process, improve the measurement efficiency, reduce the error caused by human factors, and ensure the stability and reliability of the measurement results.

[0111] The result output module is used for outputting the final three-dimensional measurement data of the object to be measured, and the output data can be directly applied to subsequent design, production and quality control and the like.

[0112] Obviously, the above embodiments are only examples for clearly illustrating, and are not intended to limit the embodiments. For those skilled in the art, other different forms of changes or variations can be made on the basis of the above description. Here, all the embodiments are not required to be exhausted, and the obvious changes or variations derived therefrom are still within the protection scope of the present application.

Claims

1. A method for three-dimensional measurement of a linear structured light bearing and a gear, characterized in that, The method comprises the following steps: Step one, data collection: place the object to be measured on the measurement platform, project a structured laser line on the object to be measured by the laser (1) provided on the measurement platform; and use the CCD camera (2) provided on the measurement platform to take laser images containing laser line patterns from different angles; Step two, image processing: use computer vision algorithms to extract feature points in the laser images and track the position changes of these feature points in different laser images; Step three, three-dimensional reconstruction: based on the known parameters of the CCD camera (2) and the position changes of the feature points on the object to be measured, use three-dimensional reconstruction algorithms to calculate the spatial coordinates of each point on the surface of the object to be measured, and use spatial coordinate conversion techniques to obtain the three-dimensional data of the object to be measured; Step four, error correction: analyze the errors of the obtained three-dimensional data; If the measurement points are blocked, adjust the measurement point positions according to the analysis results; if the measurement points are disturbed by external light, use the supplementary light (8) provided on the measurement platform to correct them; if the object to be measured has special features, capture the measurement points of the object to be measured by judging the shadow alternation points of the light; The processing steps for the object to be measured with special features are as follows: S501, observe the shadow: use the CCD camera (2) to capture shadow images of the object to be measured under light source illumination from different angles; S502, identify the alternation points: find the alternation points of light and shadow in the shadow images; the alternation points correspond to the special features of the object to be measured; S503, record the position: automatically mark the positions of the alternation points through image processing technology; the alternation point positions correspond to the depth information of the object to be measured; record the positions of the alternation points to obtain the viewpoint of the special object to be measured; if there are no alternation points, it means there is no shadow, and it is judged as a vertical image; S504, multi-directional collection: take multiple laser images of the object to be measured according to the position changes of the alternation points; Step five, result output: output the final three-dimensional measurement data of the object to be measured based on the corrected results.

2. The line structured light axle bearing and gear three-dimensional measurement method according to claim 1, wherein, In step two, the computer vision algorithm for extracting feature points in the laser image comprises the following steps: S101, image preprocessing: convert the color laser image to a grayscale image; use a filtering algorithm to remove noise in the grayscale image; and enhance the contrast of the grayscale image through histogram equalization; S102, edge detection: use an edge detection algorithm to determine the edge positions in the grayscale image by calculating the gradient direction and size; S103, feature point detection: use a corner detection method to detect feature points in the grayscale image; calculate the local feature descriptors around the detected feature points; S104, feature matching: use feature descriptors to match feature points between grayscale images taken from different angles; S105, feature point optimization: use an algorithm to remove false matching points and optimize the feature point set.

3. The line structured light bearing and gear three-dimensional measurement method of claim 2, wherein, In step three, the spatial coordinate conversion technique comprises the following steps: S201, camera pose estimation: use the matched feature points to estimate the initial position and attitude of the camera when taking each laser image through an algorithm; and use an optimization algorithm to jointly optimize the pose of the camera; S202, three-dimensional coordinate calculation: according to the internal and external parameters of the camera and the matched feature points, the three-dimensional coordinates of the feature points in the three-dimensional space are calculated by using the multi-view geometry principle; and the calculated three-dimensional point coordinates are converted to a unified global coordinate system; S203, stereo matching: by comparing the position difference of the corresponding feature points in the two laser images, combined with the camera baseline length and focal length information, the depth information of each point on the surface of the object is calculated, and the three-dimensional coordinates are obtained; S204, structured light coding: in the structured light system, the spatial information of each point on the surface of the object is decoded from the laser image by projecting a specific coding pattern.

4. The line structured light bearing and gear three-dimensional measurement method of claim 3, wherein, In step four, the processing steps of the measuring point with occlusion are as follows: S301, occlusion detection: using error calculation method to analyze the error of three-dimensional data, on the basis of error analysis, identifying the inaccurate or missing data of measuring points caused by occlusion; S302, occlusion area filling: using multi-view fusion technology, the data under different angles are combined to fill the occlusion area; S303, interpolation and filling: for the occluded measuring points under a single view, the missing data is estimated by using interpolation method; the surface reconstruction algorithm is used to fill the vacancy area; S304, iterative optimization: using iterative algorithm to adjust the three-dimensional model, through multiple iterations, the difference between the reconstructed model and the actual object is reduced.

5. The line structured light bearing and gear three-dimensional measurement method of claim 4, wherein, In step four, the processing steps of the measuring point disturbed by external light are as follows: S401, environment evaluation: by comparing the position change of the measuring point and the matching rate of the feature point under different light conditions, the interference degree of the external light on the accuracy of the measuring point data is evaluated, and the back light and front light under different light conditions are judged; S402, automatic adjustment: automatically adjusting the position, angle and brightness of the light supplement lamp (8) according to the position of the measuring point and the direction of the light disturbance; S403, reacquisition: after using the light supplement lamp (8) to illuminate the measuring point, the laser image of the object to be measured is reacquired; And the laser image reacquired is extracted and matched.

6. The line structured light bearing and gear three-dimensional measurement method of claim 5, wherein, The measurement platform comprises a laser (1), a CCD camera (2), a detection plate (3), a mechanical arm (4) and a controller; the detection plate (3) is circular, and the bottom of the detection plate (3) is provided with a driving assembly for driving the mechanical arm (4) to rotate; The driving assembly comprises a driving motor (5) fixedly connected to the bottom of the detection plate (3); the output shaft of the driving motor (5) is fixedly connected with a rotating ring (6) which is rotatably matched with the outer side of the detection plate (3); the mechanical arm (4) is fixedly connected to the top of the rotating ring (6), and the controller is used for controlling the driving motor (5) and the mechanical arm (4) to operate according to the position of the object to be measured; the CCD camera (2) is fixedly connected to the output shaft of the mechanical arm (4), and the laser (1) is fixedly connected to the top of the detection plate (3); the controller is used for controlling the operation of the CCD camera (2) and the laser (1); a telescopic rod (7) is fixedly connected to one side of the mechanical arm (4) close to the CCD camera (2), and the output shaft of the telescopic rod (7) is fixedly connected with a light supplement lamp (8); the controller is used for controlling the operation of the telescopic rod (7) and the light supplement lamp (8) according to the position needing light supplement.

7. A line structured light bearing and gear three-dimensional measurement system, which is suitable for the line structured light bearing and gear three-dimensional measurement method of any one of claims 1-6, characterized in that, It comprises the following modules: A data acquisition module is configured to capture laser images formed by laser lines projected on the surface of the object to be measured at different angles. An image processing module is configured to extract contour information of the laser lines formed on the surface of the object to be measured from the captured laser images using an algorithm. The image processing module is further configured to identify the positions of the laser lines in the laser images and extract feature points. A three-dimensional reconstruction module is configured to calculate three-dimensional coordinates of each point on the surface of the object to be measured according to the position information of the feature points using an algorithm. An error correction module is configured to perform error analysis on the generated three-dimensional coordinates and correct the laser images of the object to be measured with occlusion, light interference and special features according to the analysis results. An intelligent control module is configured to control the operation of the laser (1), the driving motor (5), the mechanical arm (4), the CCD camera (2), the telescopic rod (7) and the light supplement lamp (8) according to the imaging condition of the object to be measured. A result output module is configured to output the final three-dimensional measurement data of the object to be measured.

8. The line structured light axle bearing and gear three-dimensional measurement system of claim 7, wherein, The object to be measured includes a bearing bush and a gear.

9. The line structured light bearing and gear three-dimensional measurement system of claim 8, wherein, The special features include an oil groove, an oil hole, a conical surface, an end face flange, a reinforcing rib, a tooth shape, a tooth groove, a non-standard aperture, a keyway, a tooth tip and a tooth root.

Images