Three-dimensional measurement method and system for linear structured light bearing bush and gear

Abstract

The invention relates to the technical field of three-dimensional measurement, in particular to a linear structured light bearing bush and gear three-dimensional measurement method and system, and the method comprises the following steps: 1, data collection: employing a CCD camera arranged on a measurement platform to shoot laser images containing laser line patterns from different angles; step 2, image processing: extracting feature points in the laser image by using a computer vision algorithm; and 3, three-dimensional reconstruction: based on the known CCD camera parameters and the position change of the feature points on the to-be-measured object, obtaining three-dimensional data of the to-be-measured object by using a space coordinate conversion technology. And step 4, error correction: carrying out error analysis on the obtained three-dimensional data. And step 5, result output: outputting final three-dimensional measurement data of the to-be-measured object based on the corrected result. Through the automatic and intelligent measurement means, the measurement period can be further shortened, and the production efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of three-dimensional measurement technology, and in particular to a three-dimensional measurement method and system for linear structure optical bearings and gears. Background Art

[0002] Line structured light technology is a non-contact 3D 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 deformation patterns formed by these laser lines on the surface of the object. By analyzing the changes in these patterns, the 3D information of the object surface can be calculated.

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

[0004] To sum up, how to solve the problem that the existing technology mainly relies on a small number of point and line measurements, which is difficult to fully reflect the overall picture of the object to be measured and has low measurement efficiency has become a difficult problem that needs to be urgently solved in this field. Therefore, it is necessary to propose a three-dimensional measurement method and system for linear structure optical bearings and gears that can fully reflect the overall picture of the object and has higher efficiency. Summary of the invention

[0005] To solve the above problems, the present invention provides a linear structure optical bearing and gear three-dimensional measurement method and system, which automatically marks the position of alternating points through image processing technology, thereby obtaining the viewpoint of the special-shaped object to be measured, and uses multi-directional collected data to generate more complete information. Through automated and intelligent measurement methods, the measurement cycle can be further shortened and production efficiency can be improved.

[0006] In order to achieve the above object, the technical solution of the present invention is as follows: a linear structure optical bearing and gear three-dimensional measurement method, comprising the following steps:

[0007] Step 1: Data collection: Place the object to be measured on the measuring platform, project a structured laser line on the object to be measured through a laser provided on the measuring platform; and use a CCD camera provided on the measuring 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 laser images and track the position changes of these feature points in different laser images.

[0009] Step 3: 3D reconstruction: Based on the known CCD camera parameters and the position changes of the feature points on the object to be measured, a 3D reconstruction algorithm is used to calculate the spatial coordinates of each point on the surface of the object to be measured, and the 3D data of the object to be measured is obtained using spatial coordinate conversion technology.

[0010] Step 4, error correction: perform error analysis on the obtained 3D data; if the measuring point is blocked, adjust the measuring point position according to the analysis result; if the measuring point is disturbed by external light, use the fill light on the measuring platform to correct it; if the object to be measured has special-shaped features, capture the measuring point of the object to be measured by judging the alternating points of light and shadow.

[0011] Step 5: Result output: Output the final three-dimensional measurement data of the object to be measured based on the corrected result.

[0012] Further, in step 2, the computer vision algorithm extracts feature points in the laser image including the following steps:

[0013] S101, image preprocessing: converting the color laser image into a grayscale image; using a filtering algorithm to remove noise in the grayscale image; and enhancing the contrast of the grayscale image through a histogram equalization method.

[0014] S102, edge detection: using 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: using a corner point detection method to detect feature points in the grayscale image; calculating local feature descriptors around the detected feature points.

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

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

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

[0019] S201, camera pose estimation: using the matched feature points, the algorithm estimates the initial position and pose of the camera when each laser image is taken; and the camera pose is jointly optimized through the optimization algorithm.

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

[0021] S203, stereo matching: by comparing the position difference of 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 to obtain the three-dimensional coordinates.

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

[0023] Furthermore, in step 4, the processing steps for measuring point occlusion are as follows:

[0024] S301, occlusion detection: using an error calculation method to perform error analysis on the three-dimensional data, and based on the error analysis, identifying situations where the measurement point data is inaccurate or missing due to occlusion.

[0025] S302, filling in occluded areas: using multi-view fusion technology, merging data from different perspectives to fill in occluded areas.

[0026] S303, interpolation and filling: For obscured measuring points under a single viewing angle, use an interpolation method to estimate the missing data; and use a surface reconstruction algorithm to fill in the missing areas.

[0027] S304, iterative optimization: use an iterative algorithm to adjust the three-dimensional model, and through multiple iterations, narrow the gap between the reconstructed model and the actual object.

[0028] Furthermore, in step 4, the processing steps for the measurement point being disturbed by external light are as follows:

[0029] S401, Environmental assessment: By comparing the position changes of the measuring points and the matching rates of the feature points under different lighting conditions, the interference degree of the external light on the accuracy of the measuring point data is evaluated, and the backlighting and frontlighting conditions under different lighting conditions are determined.

[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 light interference.

[0031] S403, re-collection: after illuminating the measuring point with a fill light, re-collect the laser image of the object to be measured; and perform feature extraction and matching on the re-collected laser image.

[0032] Furthermore, in step 4, the processing steps for the presence of irregular features in the object to be tested are as follows:

[0033] S501. Observe the shadow: Use a CCD camera to capture the shadow image of the object to be measured under the illumination of the light source from different angles.

[0034] S502, identifying alternating points: finding alternating points of shadow and light in the shadow image; the alternating points correspond to the special-shaped features of the object to be measured.

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

[0036] S504, multi-directional acquisition: taking laser images of the object to be measured multiple times according to the position changes of the alternating points.

[0037] Furthermore, the measuring platform includes a laser, a CCD camera, a detection board, a mechanical arm and a controller; the detection board is circular, and a driving component for driving the mechanical arm to rotate is provided at the bottom of the detection board.

[0038] The driving assembly includes a driving motor, which is fixedly connected to the bottom of the detection board; the output shaft of the driving motor is fixedly connected to a rotating ring, and the rotating ring rotates and cooperates with the outside of the detection board; the mechanical arm is fixedly connected to the top of the rotating ring, and the controller is used to control the operation of the driving motor and the mechanical arm 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 board, and the controller is used to control the operation of the CCD camera and the laser; a telescopic rod is fixedly connected to the side of the mechanical arm close to the CCD camera, and a fill light is fixedly connected to the output shaft of the telescopic rod; the controller is used to control the operation of the telescopic rod and the fill light according to the position where fill light is required.

[0039] Furthermore, the linear structured light bearing and gear three-dimensional measurement system is suitable for any of the linear structured light bearing and gear three-dimensional measurement methods described above, and includes the following modules:

[0040] The data acquisition module is used to capture the laser images formed by the laser line projected on the surface of the object to be measured at different angles.

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

[0042] The 3D reconstruction module is used to calculate the 3D coordinates of each point on the surface of the object to be measured using an algorithm based on the position information of the feature points.

[0043] The error correction module is used to perform error analysis on the generated three-dimensional coordinates and to correct the laser images of the objects to be measured with occlusion, light interference and special-shaped features according to the analysis results.

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

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

[0046] Furthermore, the objects to be tested include but are not limited to bearings and gears.

[0047] Furthermore, the special-shaped features include, but are not limited to, oil grooves, oil holes, tapered surfaces, end flanges, reinforcing ribs, tooth shapes, tooth grooves, non-standard hole diameters, keyways, tooth tops and tooth roots.

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

[0049] 1. This solution can capture the details of the surface of the object to be measured through the projection of structured laser lines and multi-angle shooting of CCD cameras, making the results of 3D reconstruction more accurate. When the bearings and gears are blocked, feature points that are missing or inaccurate at certain viewing angles are identified through feature point matching and multi-view image comparison. Using image data from different viewing angles, the correspondence between different viewing angles is determined through multi-view geometry methods, and the data from different viewing angles are fused to fill the blocked areas. For the obscured measuring points at a single viewing angle, the interpolation method is used to estimate the missing data; based on the data of the surrounding known points, mathematical functions are used to predict the points in the blocked area. Using the optimization algorithm, the 3D model is adjusted through multiple iterations to make it closer to the shape of the actual object; thereby improving the overall measurement accuracy and the quality of the model.

[0050] 2. This solution improves the measurement accuracy by evaluating and adjusting the light conditions to reduce the interference of external light on the measuring points; evaluates the backlight and front light conditions under different light conditions, and analyzes the impact of light direction on feature point detection. According to the position of the measuring point and the direction of light interference, the position of the fill light is automatically adjusted to ensure that the light can evenly illuminate the measuring point area. The movement of the fill light is controlled by the robotic arm so that it can adapt to the measuring points in different positions, and the angle of the fill light is automatically adjusted to avoid excessive reflections or shadows, thereby improving the visibility of the feature points. The laser image of the object to be measured is recaptured using the adjusted fill light to ensure that the light conditions are suitable for the detection of feature points. In different measurement environments, by dynamically adjusting the fill light, the system can adapt to more complex working conditions and expand the scope of application.

[0051] 3. This solution searches for the alternating points of shadow and light in the acquired shadow image; these alternating points usually correspond to the mutation positions on the surface of the object. The positions of the alternating points are automatically marked by image processing technology to obtain the viewpoints of the irregular 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 3D reconstruction. The data collected from multiple angles is used to make up for the lack of information under a single perspective and generate more complete 3D model information. For objects with complex geometric shapes such as bearings and gears, this method can better capture their irregular features and improve the applicability of measurement.

[0052] 4. This solution finds corresponding feature points in multiple laser images, ensures that these feature points are consistent under different viewing angles, and uses algorithms to estimate the initial position and posture of the camera when each laser image is taken; jointly optimize the camera's posture to improve the accuracy of posture estimation. Triangulation is performed through multi-view geometry principles to reduce the problem of insufficient information under a single view; structured light coding technology can maintain high measurement accuracy under large changes in lighting. Combined with the camera baseline length and focal length information, the depth information of each point on the surface of the object is calculated; the depth information is combined with the two-dimensional image coordinates to obtain three-dimensional coordinates.

[0053] 5. This solution can remove noise and enhance contrast through image preprocessing, making feature points easier to detect. Based on the matched feature points, triangulation is performed using the principle of multi-view geometry to accurately calculate the coordinates of the feature points in three-dimensional space. Feature matching can ensure the consistency of feature points in different images, and feature point optimization can remove incorrect matching points and improve the accuracy of the three-dimensional model. Through feature matching of multi-view images, measurement errors caused by occlusion and illumination changes can be reduced. Through automated and intelligent measurement methods, the measurement cycle can be further shortened and production efficiency can be improved.

[0054] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a schematic flow chart of the linear structure optical bearing and gear three-dimensional measurement method in an embodiment of the present invention.

[0056] Figure 2 Schematic diagram of the flow of the computer vision algorithm in the embodiment of the present invention.

[0057] Figure 3 Schematic diagram of the process of spatial coordinate conversion technology in an embodiment of the present invention.

[0058] Figure 4 The figure is a flowchart of the processing steps for the case where the measuring point is blocked in the embodiment of the present invention.

[0059] Figure 5 The figure is a flow chart of the processing steps for the measurement point being disturbed by external light in an embodiment of the present invention.

[0060] Figure 6 The figure is a flowchart of the processing steps for detecting the presence of irregular features on an object to be detected in an embodiment of the present invention.

[0061] Figure 7It is a structural block diagram of the linear structure optical bearing and gear three-dimensional measurement system in an embodiment of the present invention.

[0062] Figure 8 It is an axonometric diagram of the measuring platform in the embodiment of the present invention.

[0063] The figure marks in the drawings of the specification include: 1. laser; 2. CCD camera; 3. detection board; 4. robotic arm; 5. driving motor; 6. rotating ring; 7. telescopic rod; 8. fill light. DETAILED DESCRIPTION

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

[0065] Embodiment 1:

[0066] As attached Figure 1-Figure 6 As shown: The three-dimensional measurement method of the line structure light bearing and gear includes the following steps:

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

[0068] Step 2: Image processing: Use computer vision algorithms to extract feature points in laser images and track the position changes of these feature points in different laser images.

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

[0070] S101, image preprocessing: converting the color laser image into a grayscale image; using a median filter algorithm to remove noise in the grayscale image; and enhancing the contrast of the grayscale image through a histogram equalization method.

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

[0072] S103, feature point detection: use Harris corner point detection method to detect feature points in the grayscale 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: using feature descriptors to match feature points between grayscale images taken at different angles to improve matching reliability.

[0074] S105, feature point optimization: Use the RANSAC algorithm to remove erroneous matching points, optimize the feature point set, and improve the positioning accuracy of the feature points.

[0075] Step 3, 3D reconstruction: Based on the known parameters of CCD camera 2 and the position changes of feature points on the object to be measured, a 3D reconstruction algorithm is used to calculate the spatial coordinates of each point on the surface of the object to be measured, and the 3D data of the object to be measured is obtained using spatial coordinate conversion technology.

[0076] Among them, Figure 3 As shown, the spatial coordinate transformation technology includes the following steps:

[0077] S201, camera pose estimation: using the matched feature points, the SIFT algorithm is used to estimate the initial position and pose of the camera when each laser image is taken; and the camera pose is jointly optimized using the Levenberg-Marquardt optimization algorithm.

[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, triangulation is performed using the multi-view geometry principle to calculate the coordinates of the feature points in three-dimensional space; and the calculated three-dimensional point coordinates are converted into a unified global coordinate system.

[0079] S203, stereo matching: by comparing the position difference of 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 to obtain 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 to meet the measurement needs of objects with different shapes, materials and surface characteristics.

[0081] Step 4, error correction: perform error analysis on the obtained three-dimensional data; if the measuring point is blocked, adjust the measuring point position according to the analysis result; if the measuring point is interfered by external light, use the fill light 8 provided on the measuring platform to correct it; if the object to be measured has special-shaped features, capture the measuring point of the object to be measured by judging the alternating points of light and shadow.

[0082] Among them, Figure 4 As shown in the figure, the processing steps for measuring point occlusion are as follows:

[0083] S301, occlusion detection: Use error calculation methods to perform error analysis on 3D data, and identify inaccurate or missing measurement point data due to occlusion based on the error analysis. For example, find consistency and inconsistency in depth maps taken from different angles.

[0084] S302, fill in the occluded area: use multi-view fusion technology to merge data from different perspectives to fill in the occluded area. For example, if a part is occluded in one image but not in another image, the data of the latter can be used to fill in the missing part of the former.

[0085] S303, interpolation and filling: For obscured measurement points under a single viewing angle, a polynomial interpolation method is used to estimate the missing data; and a Delaunay triangulated surface reconstruction algorithm is used to fill in the missing areas.

[0086] S304, iterative optimization: use the iterative closest point ICP algorithm to adjust the three-dimensional model, and through multiple iterations, narrow the gap between the reconstructed model and the actual object, making it closer to the shape of the actual object.

[0087] like Figure 5 As shown in the figure, the processing steps for the measurement point to be disturbed by external light are as follows:

[0088] S401, Environmental Assessment: By comparing the position changes of the measuring points and the matching rates of the feature points under different lighting conditions, the interference degree of the external light on the accuracy of the measuring point data is evaluated, and the backlighting and frontlighting conditions under different lighting conditions are judged, including natural light, indoor light, strong light and weak light environments.

[0089] S402, automatic adjustment: automatically adjust the position, angle and brightness of the fill light 8 according to the position of the measuring point and the direction of light interference, to ensure that the light can evenly illuminate the measuring point area, avoid the light source directly irradiating the camera lens, and prevent glare or reflection.

[0090] S403, re-collection: after using the fill light 8 to illuminate the measuring point, re-collect the laser image of the object to be measured; and use the FLANN algorithm to perform feature extraction and matching on the re-collected laser image to ensure that the feature points are clearly visible.

[0091] like Figure 6 As shown, the processing steps for the presence of irregular features in the object to be tested are as follows:

[0092] For example, the object to be measured is a gear with a standard module, which has a round hole and evenly spaced teeth.

[0093] S501. Observe the shadow: Use CCD camera 2 to capture the shadow image of the gear under the illumination of the light source from different angles, and ensure that the position and intensity of the light source are appropriate so as to clearly display the tooth shape of the gear and the shadow changes of the circular hole edge.

[0094] S502, identifying alternating points: finding alternating points between the shadow and light of the tooth tip, tooth root and circular hole in the shadow image; the alternating points correspond to the special-shaped features of the object to be measured. This method is very effective in capturing the details of edges, protrusions and depressions.

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

[0096] S504, multi-directional acquisition: The laser image of the gear is taken multiple times according to the change of the alternating point position to ensure that all the detailed features of the tooth shape and circular hole edge of the gear surface are captured from different angles.

[0097] Step 5, result output: Output the final 3D measurement data of the object to be measured based on the corrected result. The entire measurement process reduces the error caused by human operation through intelligent and automated operation. At the same time, the accuracy of computer vision algorithms and 3D reconstruction algorithms also further reduces the error of measurement results.

[0098] Embodiment 2:

[0099] As attached Figure 8 As shown, the difference from the above embodiment is that the measuring platform includes a laser 1, a CCD camera 2, a detection board 3, a mechanical arm 4 and a controller; the detection board 3 is circular, and a driving component for driving the mechanical arm 4 to rotate is provided at the bottom of the detection board 3.

[0100] The driving assembly includes a driving motor 5, which is bolted and fixedly connected to the bottom of the detection plate 3; the output shaft of the driving motor 5 is bolted and fixedly connected to a rotating ring 6, and the rotating ring 6 rotates and cooperates with the outside of the detection plate 3; the mechanical arm 4 is bolted and fixedly connected to the top of the rotating ring 6, and the controller is used to control the operation of the driving motor 5 and the mechanical arm 4 according to the position of the object to be measured; the CCD camera 2 is bolted and fixedly connected to the output shaft of the mechanical arm 4, and the laser 1 is bolted and fixedly 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 side of the mechanical arm 4 close to the CCD camera 2 is bolted and fixedly connected to a telescopic rod 7, and the output shaft of the telescopic rod 7 is bolted and fixedly connected to a fill light 8; the controller is used to control the operation of the telescopic rod 7 and the fill light 8 according to the position where fill light is required.

[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 a drive motor 5 is bolted and fixedly connected to the bottom of the detection plate 3, the output shaft of the drive motor 5 is bolted and fixedly connected to a rotating ring 6, the rotating ring 6 is rotatably matched with the outside of the detection plate 3, and the mechanical arm 4 is bolted and fixedly connected to the top of the rotating ring 6, 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 encountering insufficient light or being blocked, since the mechanical arm 4 is bolted and fixedly connected to the side of the CCD camera 2, the output shaft of the telescopic rod 7 is bolted and fixedly connected to a fill light 8, the position and brightness of the fill light 8 can be adjusted through the controller to reduce the situation of insufficient imaging of the object when the light is blocked or disturbed.

[0102] The controller controls the operation of the drive motor 5 and the mechanical arm 4 according to the position information of the object to be measured, and adjusts the position and angle of the CCD camera 2 to ensure that the laser line can be accurately projected onto the surface of the object to be measured. Subsequently, the image processing algorithm is used to extract the feature points in the image and calculate the coordinates of the feature points in three-dimensional space. The use of the fill light 8 can improve the image quality and meet the requirements of poor light conditions. The combination of the mechanical arm 4 and the drive assembly can adapt to objects of different sizes and shapes to be measured, and enhance the adaptability of the system. The automatic control mechanism can reduce the influence of human factors and improve the consistency and reliability of the measurement.

[0103] Embodiment 3:

[0104] As attached Figure 7 As shown, the difference from the above embodiment is that the linear structure optical bearing and gear three-dimensional measurement system includes the following modules:

[0105] The data acquisition module is used to capture the laser images formed by the laser lines projected on the bearing and gear surfaces at different angles, ensuring that the information of the object surface can be captured from different angles and reducing the impact of occlusion.

[0106] The image processing module uses the Canny edge detection algorithm to extract the contour information formed by the laser line on the surface of the object from the collected laser image. It also identifies the position of the laser line in the laser image and extracts the feature points to improve the accuracy of subsequent 3D reconstruction.

[0107] The 3D reconstruction module is used to calculate the 3D coordinates of each point on the body surface based on the position information of the feature points using the 3D reconstruction algorithm to ensure the accuracy of the measurement results.

[0108] The error correction module is used to perform error analysis on the generated three-dimensional coordinates, and to correct the laser images of the objects to be measured with occlusion, light interference and special-shaped features according to the analysis results, which can reduce measurement errors and improve measurement accuracy.

[0109] Among them, special-shaped features include but are not limited to oil grooves, oil holes, tapered surfaces, end face flanges, reinforcing ribs, tooth shapes, tooth grooves, non-standard hole diameters, keyways, tooth tops and tooth roots. The system can collect different special-shaped features.

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

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

[0112] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. Line structure optical bearing and gear three-dimensional measurement method, characterized in that: The following steps are involved: Step 1, data collection: placing the object to be measured on the measuring platform, projecting a structured laser line on the object to be measured by a laser (1) provided on the measuring platform; and using a CCD camera (2) provided on the measuring platform to capture laser images containing the laser line pattern from different angles; Step 2: Image processing: Use computer vision algorithms to extract feature points in laser images and track the position changes of these feature points in different laser images; Step 3, 3D 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, a 3D reconstruction algorithm is used to calculate the spatial coordinates of each point on the surface of the object to be measured, and the 3D data of the object to be measured is obtained by using the spatial coordinate conversion technology; Step 4, error correction: error analysis of the obtained three-dimensional data; If the measuring point is blocked, the position of the measuring point is adjusted according to the analysis result; if the measuring point is disturbed by external light, it is corrected by using the fill light (8) provided on the measuring platform; if the object to be measured has special-shaped features, the measuring point of the object to be measured is captured by judging the alternating point of the light and shadow; Step 5: Output the results: Output the final three-dimensional measurement data of the object to be measured based on the corrected results.

2. The linear structured optical bearing and gear three-dimensional measurement method according to claim 1 is characterized in that: In step 2, the computer vision algorithm extracts feature points from the laser image, including the following steps: S101, image preprocessing: converting the color laser image into a grayscale image; using a filtering algorithm to remove noise in the grayscale image; and enhancing the contrast of the grayscale image by a histogram equalization method; S102, edge detection: using an edge detection algorithm to determine the edge position in the grayscale image by calculating the gradient direction and magnitude; S103, feature point detection: using a corner point detection method to detect feature points in the grayscale image; calculating local feature descriptors around the detected feature points; S104, feature matching: matching feature points between grayscale images taken at different angles using feature descriptors; S105, feature point optimization: use an algorithm to remove incorrect matching points and optimize the feature point set.

3. The linear structured optical bearing and gear three-dimensional measurement method according to claim 2 is characterized in that: In step three, the spatial coordinate transformation technology includes the following steps: S201, camera pose estimation: using the matched feature points to estimate the initial position and pose of the camera when each laser image is taken through an algorithm; and jointly optimizing the camera pose through an optimization algorithm; S202, three-dimensional coordinate calculation: according to the internal and external parameters of the camera and the matched feature points, triangulation is performed using the multi-view geometry principle to calculate the coordinates of the feature points in the three-dimensional space; and the calculated three-dimensional point coordinates are converted into a unified global coordinate system; S203, stereo matching: by comparing the position difference of 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 to obtain the three-dimensional coordinates; S204, structured light coding: In a structured light system, 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 linear structured optical bearing and gear three-dimensional measurement method according to claim 3 is characterized in that: In step 4, the processing steps for measuring point occlusion are as follows: S301, occlusion detection: using an error calculation method to perform error analysis on the three-dimensional data, and based on the error analysis, identifying the situation where the measurement point data is inaccurate or missing due to occlusion; S302, filling the occluded area: using multi-view fusion technology to merge data from different perspectives to fill the occluded area; S303, interpolation and filling: for obscured measuring points under a single viewing angle, use an interpolation method to estimate missing data; use a surface reconstruction algorithm to fill in the missing areas; S304, iterative optimization: use an iterative algorithm to adjust the three-dimensional model, and through multiple iterations, narrow the gap between the reconstructed model and the actual object.

5. The linear structured optical bearing and gear three-dimensional measurement method according to claim 4 is characterized in that: In step 4, the processing steps for the measurement point being disturbed by external light are as follows: S401, Environmental Assessment: By comparing the position changes of the measuring points and the matching rates of the feature points under different light conditions, the interference degree of the external light on the accuracy of the measuring point data is evaluated, and the backlighting and frontlighting conditions under different light conditions are determined; S402, automatic adjustment: automatically adjusting the position, angle and brightness of the fill light (8) according to the position of the measuring point and the direction of the light interference; S403, re-collection: after illuminating the measuring point with a fill light (8), re-collect the laser image of the object to be measured; And perform feature extraction and matching on the re-collected laser images.

6. The linear structured optical bearing and gear three-dimensional measurement method according to claim 5 is characterized in that: In step 4, the processing steps for the presence of irregular features on the object to be tested are as follows: S501, observing the shadow: using a CCD camera (2) to capture the shadow image of the object to be measured under the illumination of the light source from different angles; S502, identifying alternating points: finding alternating points of shadow and light in the shadow image; the alternating points correspond to the special-shaped features of the object to be measured; S503, recording positions: automatically marking the positions of alternating points by image processing technology; the positions of alternating points correspond to the depth information of the object to be measured; and the viewpoint of the irregular object to be measured is obtained by recording the positions of alternating points; If there are no alternating dots, indicating no shadow, it is considered a vertical image; S504, multi-directional acquisition: taking laser images of the object to be measured multiple times according to the position changes of the alternating points.

7. The linear structured optical bearing and gear three-dimensional measurement method according to claim 6 is characterized in that: The measuring 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 a driving component for driving the mechanical arm (4) to rotate is provided at the bottom of the detection plate (3); The driving assembly comprises a driving motor (5), wherein the driving motor (5) is fixedly connected to the bottom of the detection plate (3); the output shaft of the driving motor (5) is fixedly connected to a rotating ring (6), and the rotating ring (6) 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 to control the operation of the driving motor (5) and the mechanical arm (4) according to the position of the object to be detected; 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), and the controller is used to control the operation of the CCD camera (2) and the laser (1); a telescopic rod (7) is fixedly connected to the side of the mechanical arm (4) close to the CCD camera (2), and a fill light (8) is fixedly connected to the output shaft of the telescopic rod (7); and the controller is used to control the operation of the telescopic rod (7) and the fill light (8) according to the position where fill light is required.

8. A linear structured light bearing and gear three-dimensional measurement system, suitable for carrying out the linear structured light bearing and gear three-dimensional measurement method according to any one of claims 1 to 7, characterized in that: Includes the following modules: A data acquisition module is used to capture laser images formed by laser lines projected onto the surface of the object to be measured at different angles; The image processing module uses an algorithm to extract the contour information formed by the laser line on the surface of the object to be measured from the collected laser image; And identify the position of the laser line in the laser image and extract the feature points; A 3D reconstruction module is used to calculate the 3D coordinates of each point on the surface of the object to be measured using an algorithm based on the position information of the feature points; The error correction module is used to perform error analysis on the generated three-dimensional coordinates and correct the laser images of the objects to be measured with occlusion, light interference and special-shaped features according to the analysis results; An intelligent control module, used for controlling the operation of the laser (1), the drive motor (5), the mechanical arm (4), the CCD camera (2), the telescopic rod (7) and the fill light (8) according to the imaging condition of the object to be measured; The result output module is used to output the final three-dimensional measurement data of the object to be measured.

9. The linear structured optical bearing and gear three-dimensional measurement system according to claim 8, characterized in that: The objects to be tested include but are not limited to bearings and gears.

10. The linear structured optical bearing and gear three-dimensional measurement system according to claim 9, characterized in that: Special-shaped features include, but are not limited to, oil grooves, oil holes, tapered surfaces, end flanges, ribs, tooth shapes, tooth grooves, non-standard hole diameters, keyways, tooth tops, and tooth roots.

Images