A quartz pendulum multi-parameter measurement method and system

Abstract

The present application belongs to the field of optical precision measurement, and relates to a quartz pendulum plate multi-parameter measurement method and system based on white light interference principle, which comprises the following steps: obtaining an interference image sequence through synchronous triggering scanning; automatically identifying sample hollowing, metal plating and transparent region based on cross-correlation analysis; positioning the interference peaks by using the centroid method or the Otsu segmentation combined with the centroid method for different regions, respectively, reconstructing the three-dimensional morphology and calculating the physical thickness of the transparent region medium; analyzing the plating uniformity based on the three-dimensional morphology data, calculating the lower swing angle by region segmentation and plane fitting, and obtaining the two-dimensional size by using the random sampling consensus algorithm for circle fitting. The present application can synchronously and non-destructively obtain the multi-dimensional parameters of the quartz pendulum plate in a single measurement, and has high detection efficiency and comprehensiveness.

Description

Technical Field

[0001] This invention belongs to the field of optical precision measurement and relates to a multi-parameter measurement method and system for a quartz pendulum based on the principle of white light interference. Background Technology

[0002] Quartz flexible accelerometers are classic components in inertial navigation systems, widely used in aerospace, earthquake monitoring, and industrial automation due to their high precision, high sensitivity, high reliability, and compact structure. The quartz pendulum, as the core sensing element of the quartz flexible accelerometer, directly determines the overall measurement accuracy and stability of the accelerometer. For example, the mass of the flexible beam affects the lifespan of the pendulum, while the uniformity of the electrode coating relates to the zero-point deviation and measurement accuracy of the quartz pendulum.

[0003] Currently, parameter measurement methods for quartz pendulums are mainly divided into two categories: contact and non-contact. Traditional contact methods, such as micrometers and vernier calipers, require direct contact with the sample, and improper operation can easily cause irreversible damage to the pendulum. Non-contact measurement methods developed in recent years do not require contact with the sample and can achieve rapid measurement, offering significant advantages. Representative methods include the spectral confocal method and the visual measurement method. The spectral confocal method, based on the principle of spectral confocalization, can automatically measure parameters such as pendulum thickness, step height, and pendulum descent, significantly improving measurement accuracy. However, its single-point scanning mode results in a relatively slow measurement speed, and its efficiency needs further improvement. The visual measurement method identifies the pendulum edges based on differences in image grayscale gradients, establishes search and measurement models for different size features, and achieves rapid measurement of multiple two-dimensional geometric parameters, improving detection efficiency. However, this method also has certain limitations: on the one hand, it is difficult to obtain three-dimensional morphological parameters; on the other hand, it inherits the inherent shortcomings of visual inspection technology, such as sensitivity to ambient light, and the measurement results are easily affected by changes in external light.

[0004] Therefore, there is an urgent need in this field for a novel measurement scheme that can simultaneously, quickly, and non-destructively acquire comprehensive parameters, including two-dimensional dimensions, three-dimensional morphology, film thickness, and even micro-angles, in a single measurement. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a multi-parameter measurement method and system for quartz pendulums. A single measurement can cover the complete three-dimensional morphology of the quartz pendulum and effectively distinguish different structural and material regions. Simultaneously, based on the acquired three-dimensional data, a dedicated data processing workflow is developed to achieve efficient and accurate extraction of multiple dimensional parameters of the pendulum.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] The first aspect of the present invention is to provide a method for measuring multiple parameters of a quartz pendulum, comprising the following steps:

[0008] A complete sequence of interference images is acquired by vertically scanning a quartz pendulum using a white light interferometry system. During the scanning process, the camera simultaneously acquires one frame of interference image for each preset step the high-precision piezoelectric ceramic scanning stage moves, thus forming the interference image sequence.

[0009] Based on the interference image sequence, perform the following operations:

[0010] Structural recognition and material differentiation: Based on the characteristics of the effective white light vertical scanning signal and the number of interference peaks at each pixel, the hollow areas, metal coating areas and transparent medium areas on the quartz pendulum are automatically identified.

[0011] Three-dimensional morphology reconstruction: For different identified regions, corresponding algorithms are used to accurately locate the interference peaks in order to reconstruct the three-dimensional surface morphology of the quartz pendulum; wherein, for the metal coating region, its single interference peak is located; for the transparent medium region, the interference peaks on its upper and lower surfaces are separated and located.

[0012] Multi-parameter calculation: Based on the reconstructed three-dimensional topography data, calculate at least one of the following parameters of the quartz pendulum: coating thickness, coating uniformity, physical thickness of the flexible beam, pendulum angle, and two-dimensional geometric dimensions.

[0013] Furthermore, the aforementioned structure identification and material differentiation specifically include:

[0014] Construct a theoretical model of white light interference signals;

[0015] The theoretical model is cross-correlated with the actual light intensity sampling sequence of each pixel.

[0016] The determination is based on the peak characteristics of the cross-correlation results: areas without significant peaks are identified as hollow areas; areas with a single main peak are identified as metal coatings; and areas exhibiting bimodal characteristics are identified as transparent media areas.

[0017] Furthermore, the formula for the cross-correlation operation is:

[0018]

[0019] in, This is the actual sampling sequence of a certain pixel in the experiment. The theoretical sampling signal model is constructed, where n represents the number of sampling points and k represents the time shift.

[0020] Furthermore, for the metal coating region, the centroid method is used to accurately locate its single interference peak, thereby obtaining the surface morphology. The calculation formula for the centroid method is as follows:

[0021]

[0022] in, Representing pixels The intensity value of the i-th frame in the current sampling frame.

[0023] Furthermore, for the transparent medium region, the interference peak signals of its upper and lower surfaces are first separated using the Otsu threshold segmentation method, and then the centroid method is used to locate the peak positions of the separated interference peaks. The optical thickness is obtained by calculating the scanning axial distance between the interference peaks on the upper and lower surfaces, and combined with the refractive index of the quartz material at the system's center wavelength, the optical thickness is converted into the physical thickness, thereby realizing three-dimensional morphology reconstruction and thickness measurement. The formula for calculating the physical thickness d is:

[0024]

[0025] in, The optical thickness of the flexible beam is represented by the distance between the two coherent peaks. This represents the refractive index of the quartz material at the center wavelength of the current measurement system.

[0026] Furthermore, the calculation of the hem angle specifically includes:

[0027] Based on three-dimensional topographic data, the base area and tongue-shaped pendulum area of ​​the quartz pendulum are automatically identified and segmented;

[0028] Plane fitting is performed on the base region and the tongue-shaped pendulum region respectively to obtain the base fitting plane and the tongue-shaped pendulum fitting plane;

[0029] Calculate the angle between the normal vectors of the two fitted planes; this angle is the swing angle θ. The calculation formula is:

[0030]

[0031] in, and These are the normal vectors of the base fitting plane and the tongue-shaped pendulum fitting plane, respectively.

[0032] Furthermore, the calculation of the two-dimensional geometric dimensions specifically includes:

[0033] Based on the identified hollow and non-hollow areas, the outer contour point set of the quartz pendulum is extracted;

[0034] The external contour point set is filtered to remove outliers;

[0035] A random sampling consensus algorithm is used to perform circle fitting on the filtered contour points to obtain the outer diameter.

[0036] The second aspect of the present invention is to provide a multi-parameter measurement system for a quartz pendulum based on the above method, comprising: a telecentric light source, a sample adjustment stage, an active vibration isolation stage, a camera, a telecentric imaging lens, a plane mirror, a mirror adjustment frame, a motion controller, a high-precision piezoelectric ceramic scanning stage, and a beam splitter.

[0037] The telecentric light source is used to provide uniform and stable large-aperture parallel light illumination for the system. After the beam is split by the beam splitter, it forms a measurement beam and a reference beam, which are projected onto the sample and the plane mirror, respectively. The two beams interfere after reflection and are finally imaged onto the camera through the telecentric imaging lens to complete image acquisition.

[0038] The sample adjustment stage is used to adjust the sample pose, its Z-axis displacement stage is used to adjust the focusing position of the sample, and the pitch and yaw adjustment stage is used to optimize the sample attitude in order to obtain the best interference fringes.

[0039] The active vibration isolation table is used to isolate external vibrations and ensure the stability of the measurement process.

[0040] The mirror adjustment bracket is used to adjust the angle and focus of the plane mirror in the initial state to ensure that the reference beam is incident perpendicularly and reflected.

[0041] The high-precision piezoelectric ceramic scanning stage drives the reference mirror to perform high-precision scanning along a one-dimensional direction, and can move at a preset step size and speed. The motion controller receives instructions from the computer, controls the continuous movement of the piezoelectric ceramic scanning stage, and sends a synchronization trigger signal to the camera according to the set step size to achieve synchronous scanning and acquisition.

[0042] Advantages and beneficial effects of the present invention:

[0043] 1. Employing a non-contact optical sensing principle, the measurement process does not affect the sample in any way. Furthermore, a single scan can acquire the complete three-dimensional morphology of the quartz pendulum, significantly improving measurement efficiency.

[0044] 2. Based on interference signal feature recognition, the system can automatically distinguish different structures and regions of a quartz pendulum, achieving intelligent structure and material identification;

[0045] 3. The entire measurement process is highly automated, and multi-dimensional parameters can be accurately extracted in a single scanning process. Attached Figure Description

[0046] Figure 1This is a system diagram of the quartz pendulum multi-parameter measurement system based on the white light interference principle of this invention;

[0047] Figure 2 This is a flowchart of the multi-parameter evaluation method of the present invention;

[0048] Figure 3 This is a data processing flowchart of the multi-parameter evaluation method during the measurement process of this invention;

[0049] Figure 4 This is a schematic diagram illustrating the measurement of the swing angle during the measurement process of this invention;

[0050] Figure 5 This is a schematic diagram illustrating the measurement of coating height during the measurement process of this invention;

[0051] Figure 1 1: Telecentric light source; 2: Sample adjustment stage; 3: Active vibration isolation stage; 4: Camera; 5: Telecentric lens; 6: Plane mirror; 7: Angle adjustment frame; 8: High-precision piezoelectric ceramic scanning stage; 9: Beam splitter. Detailed Implementation

[0052] The following detailed description of the quartz pendulum multi-parameter measurement system and method based on the principle of white light interference, with reference to examples and accompanying drawings, provides a more comprehensive understanding of the present invention.

[0053] like Figure 1 As shown, the quartz pendulum multi-parameter measurement system based on the principle of white light interference of the present invention includes an adjustment stage 2 for placing the quartz pendulum sample. The adjustment stage 2 consists of a Z-axis displacement stage and an angle adjustment stage. The sample to be measured is mounted on the adjustment stage via a stage. First, the Z-axis displacement stage is used to adjust it to a position where the optical path of the reference optical path and the optical path of the measurement optical path are approximately equal, at which point interference fringes appear in the field of view. Then, the angle adjustment stage is used to further adjust the sample orientation until only 1-2 interference fringes remain in the field of view, which is the optimal fringe position. The parallel beam emitted by the telecentric light source 1 is split into two paths by the beam splitter prism 9: the measurement beam illuminates the sample surface and returns, and the reference beam illuminates the surface of the plane mirror 6 and returns. The two beams converge and interfere at the cemented surface of the beam splitter prism 9, and then are imaged onto the camera 4 by the telecentric lens 5. During the measurement process, the controller receives instructions from the computer and drives the high-precision piezoelectric ceramic scanning stage 8 to move the plane mirror 6 to perform a one-dimensional scan within a set range. Each time the scanning stage 8 moves by a preset step, the controller sends a synchronization pulse signal to the camera 4, which is in external trigger mode, to synchronize scanning and image acquisition. Finally, the system stores the acquired interference image sequence for subsequent data processing and analysis.

[0054] like Figure 2 As shown, the multi-parameter measurement method for a quartz pendulum based on the principle of white light interference of the present invention includes the following steps:

[0055] 1) Start the measurement system, fix the quartz pendulum on the adjustment stage 2, and adjust it to the optimal measurement position through the real-time image of the camera 4.

[0056] (1.1) First, start all hardware devices of the system in sequence, including the light source, camera, piezoelectric scanning stage, and motion controller, and ensure that the computer communicates normally with each device. Then, use the control software on the computer to drive the displacement stage back to the mechanical zero point to complete the system's position initialization. Finally, set the image acquisition parameters in the acquisition software: set the camera's exposure time according to the light intensity, define the region of interest (ROI) to be analyzed, and select an appropriate number of image bits to ensure that the interference signal has sufficient dynamic range.

[0057] (1.2) Based on the imaging sharpness of the quartz pendulum in camera 4, the Z-axis of the adjustment stage 2 is used for coarse focusing to initially image the sample. Once near the focusing position, fine focusing is performed until interference fringes begin to appear in the field of view, indicating that the optical paths of the reference and measurement lights are nearly equal. Subsequently, the pitch / yaw device on the adjustment stage 2 is finely adjusted to change the spatial orientation of the sample relative to the reference light path, thereby reducing the number of interference fringes in the field of view to 1-2 fringes to obtain an interference field with optimal contrast.

[0058] (1.3) The high-precision piezoelectric ceramic scanning stage is initially moved using computer control, while the real-time image from the camera is observed to accurately locate the complete Z-axis position range from the initial appearance of the interference fringes to their peak contrast and eventual disappearance. Subsequently, the determined start and end points are used as the scanning range and formally set in the measurement software. At the same time, based on the requirements of measurement accuracy and efficiency, reasonable scanning step size and scanning speed parameters are set within this range to prepare for subsequent synchronous acquisition.

[0059] 2) The system employs a synchronous trigger mode for scanning and acquisition. In this mode, the motion controller drives the high-precision piezoelectric ceramic scanning stage 8 to move the plane mirror 6 to begin one-dimensional scanning. For each precise step the scanning stage moves, the controller sends a high-precision synchronous trigger pulse to the camera 4. The camera synchronously acquires a frame of interference fringe image based on this pulse, ensuring that each scanning position corresponds to a clear interference pattern, ultimately obtaining a set of strictly corresponding interference image sequences.

[0060] 3) After scanning, the system performs automated microstructure identification and material analysis on the acquired complete interference image sequence. Subsequently, the centroid method is used to accurately locate the peak positions of the effective interference signals, and the three-dimensional morphology data of the sample surface is reconstructed pixel by pixel. Finally, based on this three-dimensional morphology data, a series of key parameters are further calculated and extracted, including but not limited to coating height, coating uniformity, and the flatness and roughness of the quartz substrate.

[0061] (3.1) Read all the interferometric image sequences acquired during the scanning process from the specified storage path.

[0062] (3.2) Based on the actual light source spectral distribution and core parameters such as the numerical aperture of the objective lens in the measurement system, a theoretical white light interference signal model is established as the benchmark for feature matching. During the identification process, this theoretical model is cross-correlated with the actual light intensity signal collected by each pixel during vertical scanning. By analyzing the number and distribution of peaks in the cross-correlation results, regions without significant peaks are identified as hollow structures without reflective surfaces; regions with a single main peak are identified as metal coatings with only one effective reflective interface; and regions exhibiting bimodal characteristics are identified as transparent medium regions with upper and lower reflective interfaces.

[0063]

[0064] in, This is the actual sampling sequence of a certain pixel in the experiment. The theoretical sampling signal model is constructed, where n represents the number of sampling points and k represents the time shift.

[0065] (3.3) For the different material regions identified, the system adopts corresponding morphology reconstruction strategies. For hollow regions, since there is no effective interference signal, the morphology data of this region is directly left blank and does not participate in the reconstruction. For metal coating regions, the centroid method is used to accurately locate the surface morphology by utilizing the single coherent peak in its interference signal.

[0066]

[0067] in, Representing pixels The intensity value of the i-th frame in the current sampling frame.

[0068] For the transparent medium region, the Otsu threshold segmentation method is first used to separate the interference peak signals on its upper and lower surfaces, and then the centroid method is applied to locate the peak positions. Based on this, the distance between the interference peaks on the upper and lower surfaces of each pixel in the transparent region is calculated, and this distance corresponds to the optical thickness of the flexible beam at that point. Combined with the known refractive index of the quartz material at the center wavelength of the system, the optical thickness can be converted into the physical thickness, ultimately achieving accurate reconstruction of the three-dimensional morphology and thickness measurement of the thin film structure.

[0069]

[0070] in, The optical thickness of the flexible beam is represented by the distance between the two coherent peaks. This represents the refractive index of the quartz material at the center wavelength of the current measurement system.

[0071] (3.4) Based on the acquired high-precision three-dimensional topographic data, the system can quantitatively analyze the height and uniformity of the metal coating, providing a reliable basis for process quality assessment. Furthermore, by utilizing the connectivity characteristics of regions in three-dimensional space, automatic identification and effective segmentation of different structural regions of the quartz pendulum (such as the base region and the tongue-shaped pendulum region) can be achieved. Based on this, least-squares plane fitting is performed on the base region and the tongue-shaped pendulum region respectively to obtain the base fitting plane and the tongue-shaped pendulum fitting plane. By calculating the angle between the normal vectors of the two planes in three-dimensional space, non-destructive and accurate measurement of the pendulum's downward swing angle can be achieved.

[0072]

[0073] in, and These are the normal vectors of the base fitting plane and the tongue-shaped pendulum fitting plane, respectively.

[0074] (3.5) Based on the hollow and non-hollowed areas identified in step (3.2), their clear boundary contours can be further extracted, providing a basis for the accurate measurement of two-dimensional geometric dimensions. Taking the measurement of the outer diameter of a quartz pendulum as an example: First, the contour extraction operator is used to extract the complete set of external contour points; then, the minimum connected component analysis technique is used to effectively filter out abnormal discrete points caused by noise or local defects, thereby improving the purity of the contour data; on this basis, the random sampling consensus algorithm is applied to perform robust circle fitting. This algorithm can effectively resist the interference of local incompleteness or outliers in the contour, accurately extract key geometric parameters such as the diameter of the outer circle and the coordinates of the center, thereby realizing high-precision and automated measurement of key two-dimensional dimensions such as the outer diameter of the quartz pendulum.

[0075] To verify the repeatability and accuracy of the method of the present invention, measurements were performed on laboratory quartz pendulum samples.

[0076] Measurement process: according to Figure 1 The system shown and Figure 2 Follow the procedure shown. Set the piezoelectric ceramic scanning range to 330 μm and the step size to 50 nm, acquiring a total of 6600 interference images. Process the acquired interference images according to... Figure 3 The data processing flow shown can obtain the coating thickness, coating uniformity, physical thickness of the flexible beam, swing angle, and two-dimensional geometric dimensions of the quartz pendulum.

[0077] Table 1 shows the results of five repeatable experiments on the thickness of the transparent medium. The measurement results are taken as the average of the height difference between the upper and lower surfaces of all pixels in the measurement area, which is used as the single optical thickness measurement value. The physical thickness is obtained by dividing the optical thickness by the refractive index of the quartz material at the center wavelength of the measurement system (n=1.4587). The mean value of the physical thickness measurement is 9.5521 μm, and the standard deviation of the measurement is 0.0025 μm.

[0078] Table 1

[0079]

[0080] Table 2 presents the evaluation results of the surface microstructure of the quartz pendulum substrate region. These evaluation results are based on the analysis of the surface morphology characteristics of the test samples using the three-dimensional surface texture parameters provided in ISO 25178-2, "Specifications for Product Surface Geometry".

[0081] Table 2

[0082]

[0083] Table 3 shows the measurement results of five repeatable measurements of the four circular characteristic dimensions of the quartz pendulum. After verification by a metrology institution, the verified values ​​for the outer diameter and inner diameter of the pendulum are 15.5416 μm and 13.2546 μm, respectively, with a measurement uncertainty of U = 1.5 μm (k = 2). The measurement results demonstrate excellent accuracy of the measurement system and method in measuring two-dimensional geometric parameters, with relative deviations from the verified values ​​of 0.76% and 0.37%, respectively. The measurement results show that the standard deviations of each parameter are all on the order of micrometers, further verifying the high repeatability and stability of the measurement system.

[0084] Table 3

[0085]

[0086] Figure 4 This is a schematic diagram illustrating the measurement of the pendulum angle according to the present invention. The quartz pendulum base plane and the tongue-shaped pendulum plane are labeled and classified based on regional connectivity. The normal vectors of the fitting plane are respectively... =[1.770×10 -4 2.699×10 -3 [,-1] and -5.885×10 -6 -1.845×10 -5 The measured angle of the pendulum's downward swing is 0.156°. .

[0087] Figure 5 A schematic diagram illustrating the measurement of plating height and uniformity of the present invention is shown. The measurement results were obtained by characterizing the gold plating layer using multiple measurement contour lines. Comparative analysis of the gold plating layer thickness and contour shape in different regions revealed significant differences in plating uniformity across different areas.

[0088] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several modifications and improvements can be made without departing from the inventive concept, and these all fall within the protection scope of the present invention.

Claims

1. A method for measuring multiple parameters of a quartz pendulum, characterized in that, Includes the following steps: A complete sequence of interference images is acquired by vertically scanning a quartz pendulum using a white light interferometry system. During the scanning process, the camera simultaneously acquires one frame of interference image for each preset step the high-precision piezoelectric ceramic scanning stage moves, thus forming the interference image sequence. Based on the interference image sequence, perform the following operations: Structural recognition and material differentiation: Based on the characteristics of the effective white light vertical scanning signal and the number of interference peaks at each pixel, the hollow areas, metal coating areas and transparent medium areas on the quartz pendulum are automatically identified. Three-dimensional morphology reconstruction: For different identified regions, corresponding algorithms are used to accurately locate the interference peaks in order to reconstruct the three-dimensional surface morphology of the quartz pendulum; wherein, for the metal coating region, its single interference peak is located; for the transparent medium region, the interference peaks on its upper and lower surfaces are separated and located. Multi-parameter calculation: Based on the reconstructed three-dimensional topography data, calculate at least one of the following parameters of the quartz pendulum: coating thickness, coating uniformity, physical thickness of the flexible beam, pendulum angle, and two-dimensional geometric dimensions.

2. The method for multi-parameter measurement of a quartz pendulum according to claim 1, characterized in that, The aforementioned structure identification and material differentiation specifically include: Construct a theoretical model of white light interference signals; The theoretical model is cross-correlated with the actual light intensity sampling sequence of each pixel. The determination is based on the peak characteristics of the cross-correlation results: areas without significant peaks are identified as hollow areas; areas with a single main peak are identified as metal coatings; and areas exhibiting bimodal characteristics are identified as transparent media areas.

3. The method for multi-parameter measurement of a quartz pendulum according to claim 2, characterized in that, The formula for the cross-correlation operation is: ； in, This is the actual sampling sequence of a certain pixel in the experiment. The theoretical sampling signal model is constructed, where n represents the number of sampling points and k represents the time shift.

4. The method for multi-parameter measurement of a quartz pendulum according to claim 1, characterized in that, For the metal coating region, the centroid method is used to accurately locate the peak position of its single interference peak, thereby obtaining the surface morphology. The calculation formula for the centroid method is as follows: ； in, Representing pixels The intensity value of the i-th frame in the current sampling frame.

5. The method for multi-parameter measurement of a quartz pendulum according to claim 1, characterized in that, For the transparent medium region, the Otsu threshold segmentation method is first used to separate the interference peak signals of its upper and lower surfaces. Then, the centroid method is used to locate the peak positions of the separated interference peaks. The optical thickness is obtained by calculating the scanning axial distance between the interference peaks of the upper and lower surfaces. Combined with the refractive index of the quartz material at the system center wavelength, the optical thickness is converted into the physical thickness, thereby realizing three-dimensional morphology reconstruction and thickness measurement. The formula for calculating the physical thickness d is: ； in, The optical thickness of the flexible beam is represented by the distance between the two coherent peaks. This represents the refractive index of the quartz material at the center wavelength of the current measurement system.

6. The method for multi-parameter measurement of a quartz pendulum according to claim 1, characterized in that, The calculation of the hem angle specifically includes: Based on three-dimensional topographic data, the base region and tongue-shaped pendulum region of the quartz pendulum are automatically identified and segmented based on regional connectivity. Plane fitting is performed on the base region and the tongue-shaped pendulum region respectively to obtain the base fitting plane and the tongue-shaped pendulum fitting plane; Calculate the angle between the normal vectors of the two fitted planes; this angle is the swing angle θ. The calculation formula is: ； in, and These are the normal vectors of the base fitting plane and the tongue-shaped pendulum fitting plane, respectively.

7. The method for multi-parameter measurement of a quartz pendulum according to claim 1, characterized in that, The calculation of the two-dimensional geometric dimensions specifically includes: Based on the identified hollow and non-hollow areas, the outer contour point set of the quartz pendulum is extracted; The external contour point set is filtered to remove outliers; A random sampling consensus algorithm is used to perform circle fitting on the filtered contour points to obtain the outer diameter.

8. The quartz pendulum multi-parameter measurement system according to any one of claims 1 to 7, characterized in that, include: Telecentric light source (1), sample adjustment stage (2), active vibration isolation stage (3), camera (4), telecentric imaging lens (5), plane mirror (6), mirror adjustment frame (7), motion controller, high-precision piezoelectric ceramic scanning stage (8), and beam splitter (9); The telecentric light source (1) is used to provide parallel light, which is split by the beam splitter (9) to form a measurement beam and a reference beam, which are projected onto the sample and the plane mirror (6) respectively. After reflection, they merge and interfere, and are imaged onto the camera (4) by the telecentric imaging lens (5). The sample adjustment stage (2) is used to adjust the spatial position and orientation of the sample; The high-precision piezoelectric ceramic scanning stage (8) is used to drive the plane mirror (6) to perform one-dimensional scanning; The motion controller is used to control the movement of the high-precision piezoelectric ceramic scanning stage (8) and send a synchronous trigger signal to the camera (4) according to the preset step size to realize the synchronous scanning and acquisition.

Images