Non-contact measurement system and method for fiber optic ring skeleton based on line structured light

Abstract

The present invention discloses a non-contact measurement system and method for a fiber optic ring skeleton based on line structured light. The method comprises: continuously capturing line structured light strip images of the rotating fiber optic ring skeleton surface through a camera and extracting the centerline of the light strip; converting the two-dimensional coordinates of the light strip centerline into three-dimensional coordinates in the camera coordinate system based on camera parameters and light plane parameters; registering the three-dimensional coordinate point cloud data of the same light strip cross section at different viewing angles to obtain a cross-sectional point cloud set; splicing the cross-sectional point cloud set based on the linear equation of the fiber optic ring skeleton's rotation axis to obtain skeleton point cloud data; and reconstructing a skeleton point cloud data model based on the skeleton point cloud data to calculate parameters such as the fiber optic ring skeleton's groove width, perpendicularity, cylindricity, and concentricity. The present invention uses a single camera non-contact method based on three-dimensional machine vision to measure the fiber optic ring skeleton, improving measurement accuracy and speed.

Description

Technical Field

[0001] The present invention relates to the field of optical detection technology, and more particularly to a non-contact measurement system and method for an optical fiber ring skeleton based on line structured light. Background Art

[0002] The fiber optic ring coil is the core component of the fiber optic gyroscope (FOG). Currently, most FOGs utilize a debonded fiber optic ring, in which the coil bobbin is removed after winding. Compared to conventional debonded fiber optic rings, this design not only prevents the fiber from being sensitive to the compression and contraction stresses generated between the coil bobbin and the optical fiber when the temperature of the measurement environment changes, but also significantly reduces the temperature error of the fiber optic ring. This significantly improves the temperature stability of the fiber optic ring, thereby enhancing the measurement accuracy of the FOG.

[0003] However, during the production of de-bonded fiber optic rings, multiple skeleton components must be constantly assembled and disassembled. This repeated assembly and disassembly can cause installation errors or deformation, affecting the accuracy of the fiber optic ring. Therefore, it is necessary to measure the assembly parameters of the fiber optic ring skeleton. Currently, this measurement is often done manually, requiring re-measurement after each disassembly and reassembly. This is not only inaccurate but also very slow. More importantly, contact measurement can damage the spray coating on the fiber optic ring skeleton, rendering the skeleton unusable. All of these factors can seriously affect the production quality of fiber optic gyros.

[0004] With the development of 3D machine vision technology, non-contact measurement methods based on this technology are increasingly being used in shaft inspection, such as shaft diameter measurement using single-camera imaging and disc cam measurement using line structured light. However, due to the recent application of frameless fiber optic technology, research on fiber optic ring frame dimensional measurement based on machine vision is limited, and the methods are still immature. In particular, compared to measuring the simple diameter of shaft parts, the parameters of fiber optic ring frames are more complex and more difficult to measure.

[0005] Therefore, how to provide a non-contact measurement system and method for an optical fiber ring skeleton that is easy to implement and can measure accurately is an urgent problem that needs to be solved by those skilled in the art. Summary of the Invention

[0006] In view of this, the present invention provides a non-contact measurement system and method for optical fiber ring skeletons based on line structured light, which adopts a single camera non-contact method based on three-dimensional machine vision to measure the optical fiber ring skeleton. It can not only improve the shortcomings of traditional contact measurement such as time-consuming and labor-intensive, low precision, and fragile skeleton, but also make up for the technical deficiencies of existing machine vision non-contact measurement methods in measuring optical fiber ring skeletons, improve measurement accuracy and measurement speed, and meet a higher cost-effectiveness.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions:

[0008] The non-contact measurement method of optical fiber ring skeleton based on line structured light includes:

[0009] The camera parameters, the light plane parameters of the line structured light plane equation and the straight line equation of the fiber ring skeleton rotation axis are obtained through calibration;

[0010] A laser is used to project a line structured light from a determined position, which is then irradiated on the surface of the rotating fiber ring skeleton. A calibrated camera is used to capture the light strip image on the surface of the fiber ring skeleton and extract the center line of the light strip.

[0011] Based on the camera parameters and light plane parameters, the 2D coordinates of the light strip centerline are converted into 3D coordinates in the camera coordinate system. The 3D point cloud data collected from different camera angles for the same light strip cross section are registered to obtain a cross-sectional point cloud set.

[0012] Based on the linear equation of the optical fiber ring skeleton rotation axis, the light strip cross-section point cloud set is spliced to obtain the skeleton point cloud data;

[0013] The skeleton point cloud data model is reconstructed based on the skeleton point cloud data to obtain the fiber ring skeleton parameters, including slot width, verticality, cylindricity, concentricity and other parameters.

[0014] Preferably, when calibrating the camera parameters, it is assumed that the calibration plate is located on the plane Zw=0 of the world coordinate system Ow-XwYwZw, and the calibration plate is photographed from multiple angles using a camera. The coordinates of the feature points in the image pixel coordinate system and the world coordinate system are extracted, and the homography matrix of each image is calculated. It is linearly decomposed to obtain the intrinsic and extrinsic parameters, and then the distortion coefficient is obtained according to the distortion equation. The calibration of the camera's intrinsic and extrinsic parameters is completed through optimization using the Levenberg-Marquardt algorithm.

[0015] Preferably, the calibration method of the linear equation of the optical fiber ring skeleton rotation axis is:

[0016] The calibrated camera is fixed in position, and the planar checkerboard target is fixed on the rotation axis of the fiber optic ring skeleton, so that the axis of the fiber optic ring skeleton rotation axis passes through the plane where the planar checkerboard is located. Multiple images of the rotating planar checkerboard target are taken, and the equations of the multiple planes where the planar checkerboard is located are calculated. The intersection line of each plane is obtained, which is the straight line equation of the rotation axis of the fiber optic ring skeleton.

[0017] Preferably, the calibration method of the line structured light plane equation includes:

[0018] Let the line structure light plane equation be AX+BY+CZ+D=0, then the light plane parameters of the line structure light plane equation include the normal vector (A, B, C) and the plane parameter D;

[0019] Extract the center line equation of the light strip image projected by the line structured light on the planar checkerboard target and calculate the vanishing point;

[0020] Continuously change the target position on the plane chessboard and repeat the above steps to obtain multiple vanishing points;

[0021] The vanishing line is fitted through multiple vanishing points to obtain the normal vector (A, B, C) of the light plane;

[0022] The plane parameter D is obtained through geometric constraints.

[0023] Preferably, the specific process of extracting the center line equation of the light stripe image projected by the line structured light on the planar checkerboard target is as follows:

[0024] The grayscale image with light stripe information is obtained through the calibrated camera, and the grayscale image is filtered to remove noise points;

[0025] Segment the filtered image to obtain the light stripe region of interest;

[0026] In the light stripe region of interest, the Steger algorithm based on the Hessian matrix is used to extract the light stripe center, and the sub-pixel coordinates of the light stripe center are obtained. The least squares method is used for straight line fitting to obtain the centerline equation of the light stripe image projected by the line structured light on the planar checkerboard target.

[0027] The fiber ring skeleton non-contact measurement system based on line structured light includes: a camera, a laser, a rotating platform, a calibration module, an extraction module, a point cloud data processing module and a model reconstruction module. The fiber ring skeleton is fixed on the rotating platform.

[0028] Rotating platform: used to carry the optical fiber ring skeleton to perform circular motion around the rotating axis of the rotating platform;

[0029] Laser: used to emit line structured light onto the surface of the rotating optical fiber ring skeleton;

[0030] Calibration module: obtains camera parameters, light plane parameters of the line structured light plane equation, and the linear equation of the fiber ring skeleton rotation axis through calibration;

[0031] Camera: used to collect light strip images on the surface of the optical fiber ring skeleton;

[0032] Extraction module: used to extract the light stripe center line of the light stripe image;

[0033] Point cloud data processing module: Based on the camera parameters and light plane parameters, the 2D coordinates of the light strip centerline are converted into 3D coordinates in the camera coordinate system. The 3D point cloud data collected from different camera angles for the same light strip cross section are registered to obtain a cross-sectional point cloud set.

[0034] The cross-section point cloud set is spliced based on the straight line equation of the optical fiber ring skeleton rotation axis to obtain the skeleton point cloud data;

[0035] Model reconstruction module: Reconstruct the skeleton point cloud data model based on the skeleton point cloud data to obtain the fiber ring skeleton parameters, including slot width, verticality, cylindricity, concentricity and other parameters.

[0036] It can be seen from the above technical solution that compared with the existing technology, the present invention discloses a non-contact measurement system and method for an optical fiber ring skeleton based on line structured light, which uses a single camera three-dimensional vision measurement technology based on line structured light to measure the feature point position and size of the optical fiber ring skeleton, thereby improving the measurement efficiency and accuracy of the optical fiber gyroscope skeleton. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0038] Figure 1 The accompanying drawing is a flow chart of the non-contact measurement method of the optical fiber ring skeleton based on line structured light provided by the present invention.

[0039] Figure 2 The accompanying drawing is a structural diagram of the optical fiber ring skeleton non-contact measurement system based on line structured light provided by the present invention.

[0040] Figure 3 The accompanying drawing is a schematic diagram of the calibration of the plane equation of line structured light provided by the present invention.

[0041] Among them, 1. Rotating platform, 2. Fiber optic ring skeleton, 3. Camera, 4. Laser. DETAILED DESCRIPTION

[0042] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0043] The embodiment of the present invention discloses a non-contact measurement method of an optical fiber ring skeleton based on line structured light, such as Figure 1 Shown, including:

[0044] The camera parameters, the light plane parameters of the line structured light plane equation and the straight line equation of the fiber ring skeleton rotation axis are obtained through calibration;

[0045] The line structured light is irradiated onto the surface of the optical fiber ring skeleton, and the light strip image on the surface of the optical fiber ring skeleton is captured by a camera, and the center line of the light strip is extracted;

[0046] Based on the camera parameters and light plane parameters, the 2D coordinates of the light strip centerline are converted into 3D coordinates in the camera coordinate system. The 3D point cloud data of the same light strip cross section at different viewing angles are registered to obtain a cross-section point cloud set.

[0047] The cross-section point cloud set is spliced based on the straight line equation of the optical fiber ring skeleton rotation axis to obtain the skeleton point cloud data;

[0048] The skeleton point cloud data model is reconstructed based on the skeleton point cloud data to obtain the fiber ring skeleton parameters.

[0049] Furthermore, in terms of online structured light plane calibration technology, current plane calibration is mostly determined by the intersection of the light plane and a planar checkerboard target. This approach, based on the principle that two non-parallel lines define a plane, is used to fit the plane equation by solving multiple intersection equations. This method is simple, but requires multiple coordinate transformations using the camera's internal and external parameters, resulting in large errors and being time-consuming. To improve light plane calibration accuracy, the present invention employs a light plane calibration method based on the principle of blanking.

[0050] The image point of the infinite point on the line L is called the vanishing point of the line L. Parallel lines intersect the plane at infinity at the same point at infinity, so parallel lines have the same vanishing point. The vanishing point is only related to the direction of the line and has nothing to do with the position of the line. Let v represent the vanishing point, the direction of the line be d, the intrinsic parameter of the camera be K, and v = Kd. The projection line of the infinite line L on the plane π on the image plane is called the vanishing line of the plane. Parallel planes intersect at the same line on the plane at infinity, so parallel planes have the same vanishing line. The vanishing line is only related to the normal vector of the plane (or the direction of the plane) and has nothing to do with the position of the plane. Let the vanishing line be l, the direction of the plane be n, and l = K -T n, where K -T is the inverse transpose matrix of K.

[0051] According to the principle of vanishing, parallel planes in three-dimensional space have the same vanishing line, which is related only to the plane's normal vector. Similarly, parallel straight lines have the same vanishing point, which is related only to the direction vector of the line.

[0052] like Figure 3As shown in the figure, S1 and S2 are images of the intersection of the spatial parallel plane and the planar checkerboard target, that is, they represent the light strip images of the line structured light projected on the planar checkerboard target. The two parallel light planes are formed by fixing the laser on a linear guide and moving it; L1 and L2 are inherent parallel straight lines of the planar checkerboard target; U1 and U3 are the perpendicular lines between L1 and L2 and the perpendicular line between S1 and S2, respectively; U2 is the line connecting the intersection of S1 and L1 and the intersection of S2 and L2.

[0053] Let the plane equation of the line structured light be AX+BY+CZ+D1=0. First, by extracting the equation of the center line of the light strip projected by the line structured light on the plane checkerboard target, the intersection of the two light strips, that is, the vanishing point, is calculated; then, the position of the plane checkerboard target is changed, and the equation of the center of the light strip projected by the line structured light on the plane checkerboard target is extracted again to obtain another vanishing point. The position of the plane checkerboard target is continuously changed to obtain multiple vanishing points. The vanishing line is fitted by the least squares method, and then l=K is used. -T n, we get the normal vector (A, B, C) of the light plane. Therefore, the spatial plane equation can be expressed as:

[0054]

[0055] For parameters D1 and D2, the distance between the two planes is a fixed number and is known as d, which is the distance the laser moves on the linear guide. The distance between the straight lines of the planar checkerboard target is also d. Angle α is the angle between L1 and S1, which can be determined by the direction vectors of the two straight lines. is the angle between U1 and U3. According to the geometric relationship, the angle α is equal to the angle Then the length l of line segment U2 can be expressed as

[0056]

[0057] And l can be expressed as the distance between the two intersection points. Assume that the camera coordinates of the two intersection points are V1 and V2 respectively, then according to the equation group

[0058]

[0059] D1 and D2 can be solved.

[0060] This method only considers camera intrinsic parameters when solving for the light plane normal, eliminating the need for extrinsic parameters, resulting in more accurate results. Compared to existing methods that require extensive data coordinate transformations, this method only requires coordinate transformation when finding the intersection point, effectively reducing transformation errors.

[0061] In this embodiment, if Figure 2As shown, an embodiment of the present invention provides a non-contact measurement system of an optical fiber ring skeleton based on line structured light, comprising: a camera 3, a laser 4, a rotating platform 1, a calibration module, an extraction module, a point cloud data processing module, and a model reconstruction module, wherein the optical fiber ring skeleton is fixed on the rotating platform;

[0062] Rotating platform 1: used to carry the optical fiber ring skeleton 2 to perform circular motion around the rotating axis of the rotating platform;

[0063] Laser 4: used to emit line structured light onto the surface of the optical fiber ring skeleton;

[0064] Calibration module: obtains camera parameters, light plane parameters of the line structured light plane equation, and the linear equation of the fiber ring skeleton rotation axis through calibration;

[0065] Camera 3: used to collect the light stripe image on the surface of the optical fiber ring skeleton 2;

[0066] Extraction module: used to extract the light stripe center line of the light stripe image;

[0067] Point cloud data processing module: Based on the camera parameters and light plane parameters, the 2D coordinates of the light strip centerline are converted into 3D coordinates in the camera coordinate system. The 3D point cloud data of the same light strip cross section from different perspectives are registered to obtain a cross-sectional point cloud set.

[0068] The cross-section point cloud set is spliced based on the straight line equation of the optical fiber ring skeleton rotation axis to obtain the skeleton point cloud data;

[0069] Model reconstruction module: Reconstruct the skeleton point cloud data model based on the skeleton point cloud data to obtain parameters such as fiber ring skeleton groove width, verticality, cylindricity and concentricity.

[0070] The specific working process of the system of the present invention is:

[0071] Before the system works, the calibration module is used to calibrate the camera 3, laser 4 and the rotation axis, and the corresponding camera parameters, the light plane parameters of the line structure light plane equation and the linear equation of the fiber ring skeleton rotation axis are obtained.

[0072] When starting the measurement, the fiber ring skeleton 2 is fixed on the rotating platform 1, and the motor of the rotating platform 1 drags the skeleton to make a circular motion around the platform's rotation axis;

[0073] Place the calibrated camera and laser as acquisition modules at position 1, and collect data on the skeleton. Figure 2The light strip image at the cross section shown is processed to obtain the point cloud data at position 1, which is one side of the skeleton; after the acquisition and processing at position 1 is completed, the acquisition device is moved to position 2, and the point cloud data at position 2, which is the other side of the skeleton, is acquired and processed to obtain the same cross section as shown in the figure; at positions 1 and 2, it is necessary to ensure that the camera can collect point cloud data on both sides of the same cross section, and pay special attention to the complete acquisition of the point cloud data of the two side walls.

[0074] The point clouds collected twice on the same light strip cross section are spliced to form a complete skeleton cross section point cloud data. If the point cloud data of each cross section is rotated around the rotation axis by a fixed angle, a skeleton point cloud model is formed. (That is, when collecting the point cloud data on one side, the positions of the camera and laser remain unchanged, only the skeleton rotates. Therefore, the absolute position of the laser on the skeleton remains unchanged. In the camera coordinate system, these point cloud data are all located in the same position, but because these cross sections are obtained around the rotation axis, they belong to different cross sections. By recording the rotation angle of each cross section relative to the initial position, it can be reversed to its original position.) The skeleton point cloud model is reconstructed and the measurement results of parameters such as the fiber ring skeleton groove width, verticality, cylindricity, and concentricity are calculated.

[0075] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0076] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A non-contact measurement method of optical fiber ring skeleton based on line structured light, characterized in that: include: The camera parameters, the light plane parameters of the line structured light plane equation and the straight line equation of the fiber ring skeleton rotation axis are obtained through calibration; A laser is used to project a line structured light from a determined position, which is then irradiated on the surface of the rotating fiber ring skeleton. A calibrated camera is used to capture the light strip image on the surface of the fiber ring skeleton and extract the center line of the light strip. Based on the camera parameters and light plane parameters, the 2D coordinates of the light strip centerline are converted into 3D coordinates in the camera coordinate system. The 3D point cloud data collected from different camera angles for the same light strip cross section are registered to obtain a cross-sectional point cloud set. Based on the linear equation of the optical fiber ring skeleton's rotation axis, the light strip cross-section point cloud set is spliced to obtain skeleton point cloud data; The skeleton point cloud data model is reconstructed based on the skeleton point cloud data to obtain the fiber ring skeleton parameters.

2. The non-contact measurement method of optical fiber ring skeleton based on line structured light according to claim 1, characterized in that: When calibrating camera parameters, assume that the calibration plate is located on the plane Zw=0 of the world coordinate system Ow-XwYwZw. Use the camera to shoot the calibration plate from multiple angles, extract the coordinates of the feature points in the image pixel coordinate system and the world coordinate system, calculate the homography matrix of each image, linearly decompose it to obtain the intrinsic and extrinsic parameters, and then obtain the distortion coefficient according to the distortion equation. The calibration of the camera's intrinsic and extrinsic parameters is completed through optimization using the Levenberg-Marquardt algorithm.

3. The non-contact measurement method of optical fiber ring skeleton based on line structured light according to claim 2, characterized in that: The calibration method of the linear equation of the optical fiber ring skeleton rotation axis is: The calibrated camera is fixed in position, and the planar checkerboard target is fixed on the rotation axis of the fiber optic ring skeleton. The axis of the fiber optic ring skeleton rotation axis passes through the plane where the planar checkerboard target is located. Multiple images of the rotating planar checkerboard target are taken, and the equations of the multiple planes where the planar checkerboard target is located are calculated. The intersection lines of each plane are obtained, which is the straight line equation of the rotation axis of the fiber optic ring skeleton.

4. The non-contact measurement method of optical fiber ring skeleton based on line structured light according to claim 2, characterized in that: The calibration method of the line structured light plane equation includes: Let the line structure light plane equation be AX+BY+CZ+D=0, then the light plane parameters of the line structure light plane equation include the normal vector (A, B, C) and the plane parameter D; Extract the center line equation of the light strip image projected by the line structured light on the planar checkerboard target and calculate the vanishing point; Continuously change the target position on the plane chessboard and repeat the above steps to obtain multiple vanishing points; The vanishing line is fitted through multiple vanishing points to obtain the normal vector (A, B, C) of the light plane; The plane parameter D is obtained through geometric constraints.

5. The non-contact measurement method of optical fiber ring skeleton based on line structured light according to claim 4, characterized in that: The center line equation of the light stripe image projected by the line structured light on the planar checkerboard target is extracted. The specific process is as follows: The grayscale image with light stripe information is obtained through the calibrated camera, and the grayscale image is filtered to remove noise points; Segment the filtered image to obtain the light stripe region of interest; In the light stripe region of interest, the Steger algorithm based on the Hessian matrix is used to extract the light stripe center, and the sub-pixel coordinates of the light stripe center are obtained. The least squares method is used for straight line fitting to obtain the centerline equation of the light stripe image projected by the line structured light on the planar checkerboard target.

6. The non-contact measurement method of optical fiber ring skeleton based on line structured light according to claim 1, characterized in that: Fiber optic ring skeleton parameters include slot width, perpendicularity, cylindricity and concentricity.

7. The fiber ring skeleton non-contact measurement system based on line structured light is characterized by: include: Camera, laser, rotating platform, calibration module, extraction module, point cloud data processing module and model reconstruction module, and the optical fiber ring skeleton is fixed on the rotating platform; Rotating platform: used to carry the optical fiber ring skeleton to perform circular motion around the rotating axis of the rotating platform; Laser: used to emit line structured light onto the surface of the rotating optical fiber ring skeleton; Calibration module: obtains camera parameters, light plane parameters of the line structured light plane equation, and the linear equation of the fiber ring skeleton rotation axis through calibration; Camera: used to collect light strip images on the surface of the optical fiber ring skeleton; Extraction module: used to extract the light stripe center line of the light stripe image; Point cloud data processing module: Based on the camera parameters and light plane parameters, the 2D coordinates of the light strip centerline are converted into 3D coordinates in the camera coordinate system. The 3D point cloud data collected from different camera angles for the same light strip cross section are registered to obtain a cross-sectional point cloud set. The cross-section point cloud set is spliced based on the straight line equation of the optical fiber ring skeleton rotation axis to obtain the skeleton point cloud data; Model reconstruction module: reconstructs the skeleton point cloud data model based on the skeleton point cloud data to obtain the fiber ring skeleton parameters.

Images