A high-precision angular displacement measurement method and system based on a phase target
By setting multiple radial positioning strips on the rotating object as phase targets, and using complex Gabor filtering and robust fitting techniques combined with a Fourier eccentric trajectory model, the problem of unstable target center positioning and non-ideal trajectory distortion in visual angular displacement measurement is solved, thus achieving high-precision and stable angular displacement measurement.
Patent Information
- Application Number
- CN202611378692.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-07
- Publication Date
- 2026-11-17
AI Technical Summary
In existing visual angular displacement measurements, the target center positioning is easily affected by imaging quality. Abnormal positioning during continuous measurement can easily cause angle jumps. Eccentric installation and non-ideal factors can lead to trajectory distortion, reducing the accuracy of angular displacement inversion.
A high-precision angular displacement measurement method based on phase targets is adopted. By setting up phase targets with multiple radial positioning strips, complex Gabor filtering and robust fitting techniques are used to determine the sub-pixel center. Combined with multi-directional geometric constraints and Fourier eccentric trajectory models, high-precision angular displacement measurement is achieved.
It improves the center positioning accuracy and continuous measurement stability under complex imaging conditions, reduces the impact of installation errors and abnormal observations on angular displacement results, and enhances the applicability and engineering deployment flexibility of the measurement system.
Smart Images

Figure CN123111142A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of machine vision and precision measurement technology, specifically relating to a high-precision angular displacement measurement method and system based on a phase target. Background Technology
[0002] Angular displacement is an important physical quantity characterizing the geometric state, pointing state, and dynamic performance of a rotating mechanism. In applications such as aerospace, precision machine tools, inertial platforms, optical payloads, micro-vibration testing, and state monitoring of rotating actuators, minute angular changes can affect the pointing accuracy, assembly consistency, and operational reliability of the system through mechanical structures, optical paths, or control closed loops. Therefore, it is necessary to continuously acquire angular displacement time histories while minimizing interference with the motion state of the object under test.
[0003] Existing methods for angular displacement measurement include photoelectric encoders, autocollimators, grating sensors, position-sensitive detectors, laser interferometers, or inertial measurement units. While these methods can achieve high resolution, they typically require establishing strict mechanical or optical relationships with the rotation axis or measurement reference, and impose certain requirements on installation space, reflective surface quality, optical path collimation, and environmental stability. When the object being measured is inaccessible, installation space is limited, the measurement distance is long, or multiple rotating targets are present, the placement of dedicated angular displacement sensors and the establishment of a reference become increasingly difficult.
[0004] Machine vision measurement uses a camera to acquire target images and calculates motion by analyzing changes in the positions of features within the image. It offers advantages such as non-contact operation, flexible deployment, and traceability of the measured images. Existing monocular vision angle measurement schemes utilize checkerboard, circular, or other cooperative targets, calculating rotation angles through camera intrinsic and extrinsic parameters, perspective projection models, target pose, or feature point reprojection. Other schemes utilize coded regions to identify multiple visual targets. However, in high-precision continuous angular displacement measurement, the accuracy of angle calculation is primarily limited by the stability of the target feature positions in each frame. Centroid, template matching, or single edge fitting can all introduce center shifts under conditions of exposure variations, edge blurring, local occlusion, background texture, and small imaging size, and this center error will continue to propagate to the final angle result.
[0005] In recent years, extracting local phase from images using complex Gabor filters and achieving sub-pixel localization through phase zero-crossing has improved the ability to detect minute displacements; published research has even used the intersection of phase zero-crossing points of two intersecting lines to determine the target center. The challenge in further improving the performance of existing schemes for angular displacement measurement lies not only in obtaining the sub-pixel center of a single frame, but also in ensuring that the high-precision center of a single frame can be reliably incorporated into continuous angle inversion. On the one hand, multi-directional image filtering and full-image search increase computational load; on the other hand, if sporadic erroneous centers caused by occlusion, blurring, or local noise are directly incorporated into angle calculations, angle spikes or even erroneous phases may form, while simply smoothing the final angle sequence may weaken the true dynamic response. Therefore, it is necessary to judge the validity of features in different directions at the center observation level and distinguish between abnormal observations and real rapid motion.
[0006] Furthermore, when the visual target is mounted off-center from the rotation axis, although its center forms a closed or nearly closed trajectory with rotation, the actual trajectory in the image is usually not an ideal circle due to factors such as target mounting eccentricity, camera observation direction deviation, rotation axis tilt, end face runout, and lens distortion residuals. If the rotation angle is still directly calculated using a fixed center and geometric angle, the error of the actual trajectory deviating from the ideal model will be transformed into a periodic angular systematic error. Although existing visual angle measurement schemes can handle target motion through camera models or calibration models, it is still necessary to solve the problem of how to stably convert the actual two-dimensional center trajectory into a continuous one-dimensional mechanical angular phase without pre-setting an ideal circular trajectory.
[0007] Therefore, the practical challenge for high-precision continuous visual angular displacement measurement lies in forming a unified measurement chain that links the target structure, sub-pixel center extraction, multi-directional redundant geometric constraints, abnormal center quality control, and non-ideal closed trajectory phase inversion. This ensures that the output of the previous stage has sufficient accuracy and quality information that can be used for the next stage to make judgments, and ultimately avoids directly converting non-ideal trajectory errors and abnormal center errors into angular errors. Summary of the Invention
[0008] The technical problem to be solved by the present invention is to provide a high-precision angular displacement measurement method and system based on a phase target, which addresses the shortcomings of the prior art. This method solves the technical problems in existing visual angular displacement measurement, such as the target center positioning being easily affected by imaging quality, abnormal positioning during continuous measurement causing angle jumps, and non-ideal factors such as eccentric installation and perspective projection causing trajectory distortion and reducing the accuracy of angular displacement inversion.
[0009] The present invention adopts the following technical solution: A high-precision angular displacement measurement method based on a phase target includes the following steps: A phase target with multiple radial positioning stripes is placed at an off-center position on the rotating object, and the target image is acquired. The target image is preprocessed and coarsely located using positioning strips to obtain the coarse center and the normals of each positioning strip; Complex Gabor filtering is performed along the normal of each positioning strip. Sub-pixel points of the strip centerline are determined based on local phase zero crossings. Robust straight line fitting is performed on the sub-pixel points. The multiple principal axes formed by the relative positioning strips are weighted and intersected to obtain the original center observation. The quality of center observations is determined based on Gabor amplitude, effective zero-crossing ratio, effective strip number, robust fitting residual, and center intersection residual. Combining the consistency between the original center observations and the time-series predictions, a selection or fusion is made between the original center observations and the centers based on the state predictions of the previous frame and corrected by the effective principal axis normal constraint to obtain the fused center. During the calibration phase, the fusion center sequence of at least one complete rotation cycle is obtained and the corresponding calibration phase is determined. A Fourier eccentric trajectory model of the horizontal and vertical coordinates as a function of the calibration phase is established. During the measurement phase, the current fusion center is back-projected onto the Fourier eccentric trajectory model. The phase corresponding to the minimum distance between the center and the trajectory point is taken as the current phase. The current phase of the continuous frames is unwound, and the angular displacement is obtained based on the change of the unwound phase relative to the initial frame phase.
[0010] Furthermore, the phase target includes radial positioning strips, a central reference area, an arc-shaped coding area, and direction markers; The radial positioning strips are used to provide center positioning constraints in multiple directions, the center reference area is used to determine the target center reference area, the arc-shaped coding area is used to represent target identity information, and the direction mark is used to determine the starting direction of coding reading.
[0011] Furthermore, the radial positioning strips include six rectangular strips, with a preset angle between adjacent positioning strips, and the relative positioning strips form multiple central positioning axes; Based on the directional differences, positional symmetry, and length consistency among multiple positioning strips, the positioning strips are paired to form multi-directional geometric constraints for center calculation.
[0012] Furthermore, the determination of the sub-pixel position of the midline of the positioning strip based on complex Gabor filtering includes: The direction parameters of the complex Gabor filter are set according to the direction of the positioning strip so that the complex Gabor filter can produce the maximum response to the grayscale changes of the cross-section of the positioning strip. A one-dimensional grayscale profile is established along the normal direction of the positioning strip. The zero-crossing interval is determined based on the local phase sign change corresponding to adjacent sampling positions, and the sub-pixel midline position is obtained through linear interpolation.
[0013] Furthermore, the obtained sub-pixel midline positions are evaluated for reliability, which is determined based on at least the following parameters: The amplitude of the complex Gabor response, the zero-crossing slope of the phase, the uniqueness of the candidate zero-crossing, and the consistency between the midline position and the coarse localization results; When the confidence level is lower than a preset threshold, the weight of the corresponding sub-pixel midline position is reduced or the corresponding midline position is removed.
[0014] Furthermore, the robust straight-line fitting uses the Huber loss function to iteratively weight and fit the set of midline points of the positioning strip. The weights of outlier points are adjusted based on the normal residual from the point to the fitted straight line to reduce the impact of local occlusion, noise, and strip degradation on the center calculation.
[0015] Furthermore, the quality score also includes the encoded confidence level and the consistency between the current central observation and the time-series predicted value; When the quality score is higher than the preset threshold and the central innovation quantity is within the statistical threshold range, the current frame center observation value is used. When the quality score is lower than the preset threshold or the central innovation exceeds the statistical threshold, the fusion center is determined by combining the predicted center and the effective principal axis constraint.
[0016] Furthermore, the Fourier eccentric trajectory model includes: Based on the calibration phase, establish Fourier series models corresponding to the abscissa and ordinate of the fusion center respectively; The Bayesian information criterion is calculated for the orders of multiple candidate models, and the final trajectory model order is determined according to the minimum principle of the Bayesian information criterion.
[0017] Furthermore, the projection of the fusion center onto the Fourier eccentric trajectory model includes: Perform trajectory search within the preset phase range to determine the phase corresponding to the trajectory point with the smallest distance from the current fusion center; The current phase search range is predicted based on the phase result after unwinding in the previous frame, and the current phase search window is limited by the maximum angular velocity. The reliability of the current phase is evaluated based on the backprojection error between the current fusion center and the trajectory model.
[0018] Secondly, embodiments of the present invention provide a high-precision angular displacement measurement system based on a phase target, comprising: Industrial cameras are used to acquire images of composite-coded phase targets mounted on rotating objects; A composite coded phase target is set at an off-center position on the rotating object, including radial positioning strips, a central reference area, an arc-shaped coding area, and direction markers; An illumination unit is used to provide stable illumination for the composite coded phase target; The processing terminal is communicatively connected to the industrial camera and is used to execute the method to obtain the continuous angular displacement of the rotating object.
[0019] Thirdly, a computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described high-precision angular displacement measurement method based on a phase target.
[0020] Fourthly, embodiments of the present invention provide a computer-readable storage medium including a computer program, which, when executed by a processor, implements the steps of the above-described high-precision angular displacement measurement method based on a phase target.
[0021] Compared with the prior art, the present invention has at least the following beneficial effects: A high-precision angular displacement measurement method based on a phase target is proposed. This method constructs a complete angular displacement measurement process, including phase target image acquisition, sub-pixel center localization, quality-aware correction, actual trajectory modeling, and phase inversion. The angular displacement of a rotating object is converted into phase changes in the two-dimensional motion trajectory of an eccentric target for solution. The method improves target center localization accuracy through complex Gabor local phase zero-crossing, reduces positioning errors caused by single-feature degradation through multi-principal-axis constraints, suppresses the influence of abnormal observations on angle calculation through quality discrimination and prediction correction, and reduces systematic errors caused by eccentric installation, perspective projection, and periodic trajectory distortion by replacing the ideal circular trajectory model with a Fourier eccentric trajectory model. This improves the accuracy, stability, and reliability of continuous angular displacement measurement in complex imaging environments.
[0022] Furthermore, by setting up a composite coded phase target that includes radial positioning strips, a central reference area, an arc-shaped coding area, and direction markers, the same visual target can simultaneously perform target recognition, center positioning, and direction determination functions. This avoids the inconsistency in position references caused by using multiple independent markers, improves the consistency between coded recognition information and geometric positioning information, and provides a unified visual reference for subsequent multi-directional center calculation and continuous phase determination, thereby increasing the applicability and engineering deployment flexibility of the measurement system.
[0023] Furthermore, by setting multiple radial positioning strips and forming multi-directional principal axis constraints based on the directional relationship and spatial distribution between the positioning strips, the target center can be determined without relying on a single edge or single region feature. This allows the use of geometric constraints in multiple independent directions to offset positioning deviations caused by local occlusion, brightness variations, partial regional blurring, and imaging noise, improving the redundancy and fault tolerance of the center calculation process and enhancing the stability of target center positioning.
[0024] Furthermore, by employing a complex Gabor filter to perform local phase analysis on the positioning strip and utilizing the phase zero-crossing position to determine the sub-pixel position of the strip centerline, the problem of traditional methods based on grayscale thresholds, edge gradients, or centroid calculations being easily affected by illumination changes and edge blurring is avoided. Since local phase analysis has good robustness to brightness changes, it can obtain a more stable strip center position under low contrast, small-sized targets, and complex background conditions, improving the input accuracy of subsequent geometric calculations.
[0025] Furthermore, by comprehensively evaluating the amplitude of the complex Gabor response, phase zero-crossing characteristics, and geometric consistency, the reliability of the obtained sub-pixel centerline positions is screened. This approach can proactively identify unreliable positioning results caused by noise interference, local occlusion, or texture anomalies, preventing low-quality feature points from entering the centering calculation process, reducing the impact of anomalous features on measurement results, and improving the reliability and anti-interference capability of the centering positioning process.
[0026] Furthermore, by employing the Huber robust loss function to perform linear fitting on the midline points of the positioning strip and using an iterative weighting method to reduce the influence of outliers, the obtained principal axis parameters are made more stable. This effectively suppresses fitting offsets caused by local error points, edge defects, and occlusion areas, improves the accuracy of principal axis direction and position estimation, and provides more reliable geometric constraints for subsequent multi-principal axis intersection calculations.
[0027] Furthermore, by establishing a quality discrimination mechanism based on image quality parameters and center innovation quantity, and combining time-series prediction results with effective principal axis constraints to correct abnormal centers, dynamic optimization of center observation results is achieved. This can distinguish between real rotational motion changes and abrupt positioning changes caused by abnormal images, avoid erroneous centers directly participating in angle calculations, and avoid dynamic response lag caused by simple filtering, thereby improving the continuity and stability of angular displacement measurement during high-speed continuous rotation.
[0028] Furthermore, by collecting center motion data within the complete rotation cycle and establishing an eccentric trajectory model based on Fourier series, the non-ideal two-dimensional trajectory formed during the actual rotation is described. Compared to the method of pre-setting an ideal circular trajectory, this method can characterize trajectory deviations caused by installation eccentricity, rotation axis tilt, perspective projection, and periodic mechanical errors. Simultaneously, by utilizing the Bayesian information criterion to select the model order, it can ensure the accuracy of trajectory description while avoiding excessive model complexity, thus improving the accuracy and generalization ability of trajectory modeling.
[0029] Furthermore, by back-projecting the current fusion center onto the actual Fourier eccentric trajectory model, the corresponding phase is determined based on the distance between the center and the trajectory points. Combined with continuous phase constraints for unwinding, the two-dimensional spatial position is converted into a continuous rotational phase. This avoids the systematic errors generated when calculating angles based on a fixed center and geometric angle, improving the accuracy of angular displacement inversion under non-ideal installation conditions. Simultaneously, through phase continuity constraints and back-projection error evaluation, the resistance to jumps during continuous rotation measurement is improved.
[0030] A high-precision angular displacement measurement system based on a phase target is proposed. This system integrates image acquisition, visual positioning, data processing, and angular displacement calculation into a complete angular displacement measurement system comprising an industrial camera, a composite coded phase target, an illumination unit, and a processing terminal. It eliminates the need for a mechanical connection with the rotating object, enabling angular displacement detection visually. The system features non-contact measurement, flexible installation, recordable measurement process, and adaptability to complex spatial environments, making it applicable to scenarios such as precision turntables, rotary actuators, optical loads, and inertial platforms.
[0031] In summary, this invention combines composite coded phase target, Gabor local phase sub-pixel localization, multi-principle robust center calculation, quality-aware temporal optimization, and non-ideal eccentric trajectory back projection to transform rotational angular displacement into a two-dimensional trajectory phase solving problem. This improves the center positioning accuracy and continuous measurement stability under complex imaging conditions, and reduces the impact of installation errors, perspective distortion, and abnormal observations on angular displacement results.
[0032] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description
[0033] Figure 1 This is a schematic diagram of the angular displacement measurement system of the present invention; Figure 2 This is a schematic diagram of the structure of the composite coded phase target of the present invention; Figure 3 This is a flowchart illustrating the target center localization and mass determination process of the present invention; Figure 4This is a schematic diagram of the eccentric trajectory modeling and back projection of the center to be measured in this invention; Figure 5 This is a flowchart illustrating the overall process of the angular displacement measurement method of the present invention. Figure 6 A schematic diagram of a computer device provided in an embodiment of the present invention; Figure 7 This is a block diagram of an electronic device according to an embodiment of the present invention.
[0034] Among them, 1. Industrial camera; 2. Composite coded phase target; 3. Rotating object; 4. Illumination unit; 5. Processing terminal; 6. Drive or reference device; 21. Radial positioning strip; 22. Central reference area; 23. Arc-shaped coded area; 24. Direction mark; 60. Computer equipment; 61. Processor; 62. Memory; 63. Computer program; 600. Electronic device; 610. Processing unit; 620. Storage unit; 6201. Random access memory unit; 6202. Cache memory unit; 6203. Read-only memory unit; 6204. Program / utility; 6205. Program module; 630. Bus; 640. Display unit; 650. Input / output interface; 660. Network adapter; 700. External device. Detailed Implementation
[0035] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0036] In the description of this invention, it should be understood that the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.
[0037] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.
[0038] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. Additionally, the character " / " in this invention generally indicates that the preceding and following objects have an "or" relationship.
[0039] It should be understood that although terms such as first, second, third, etc., may be used in the embodiments of the present invention to describe the preset range, these preset ranges should not be limited to these terms. These terms are only used to distinguish the preset ranges from one another. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.
[0040] Depending on the context, the word "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."
[0041] The accompanying drawings illustrate various structural schematic diagrams according to embodiments disclosed in this invention. These drawings are not to scale, and some details have been enlarged for clarity, and some details may have been omitted. The shapes of the various regions and layers shown in the drawings, as well as their relative sizes and positional relationships, are merely exemplary and may deviate from reality due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art can design regions / layers with different shapes, sizes, and relative positions as needed.
[0042] This invention provides a high-precision angular displacement measurement method based on a phase target. A phase target positioned off-center from the rotating object serves as a unified visual reference. The sub-pixel centerline position is obtained using the zero-crossing of local phases in the positioning strips. The target center is formed through multi-strip redundancy constraints and robust principal axis convergence. For continuous images, the center observation quality is evaluated based on Gabor response, zero-crossing validity, geometric fitting, and center convergence results. In abnormal frames, the center is corrected using temporal prediction results and still-valid principal axis constraints, preventing abnormal centers from directly entering the angle calculation. Furthermore, the actual center trajectory of the complete rotation cycle is acquired. A Fourier model is used to describe the non-ideal periodic trajectory formed by factors such as off-center installation, perspective projection, and end-face runout. The center to be measured is back-projected onto the trajectory to obtain the phase. This improves the stability and continuity of target center positioning while reducing systematic errors caused by using ideal circular trajectories for angle calculation.
[0043] Please see Figure 5 The present invention discloses a high-precision angular displacement measurement method based on a phase target, comprising the following steps: S1. Constructing a measurement system and a composite coded phase target Please see Figure 1 The measurement system includes an industrial camera 1, a composite coded phase target 2 mounted at an off-center position on a rotating object 3, an illumination unit 4, and a processing terminal 5. A drive or reference device 6 is used to control the motion of the rotating object or provide experimental reference values, but it is not a necessary condition for implementing this method. The optical axis of the industrial camera 1 is oriented towards the composite coded phase target 2, and the illumination unit 4 provides stable, uniform, and unexposed illumination. The camera exposure time is preferably less than the time corresponding to the occurrence of significant motion blur in a single frame for the object under test, and the sampling frame rate is determined based on the highest angular velocity and the expected angular displacement bandwidth.
[0044] Please see Figure 2 The composite coded phase target 2 includes radial positioning strips 21, a central reference area 22, an arc-shaped coding area 23, and a direction marker 24. The radial positioning strips 21 preferably consist of six rectangular strips, with a nominal included angle of 60° between adjacent strips, forming three principal axes. The length of each strip is greater than three times its imaging width to allow for the arrangement of multiple normal profiles along the tangential direction. The arc-shaped coding area 23 indicates the target number through the presence or absence of arc segments, their length, or radial hierarchy; the direction marker 24 may be an external small pattern used to determine the starting point for coding readout.
[0045] Composite coded phase targets can be fabricated using printing, spraying, screen printing, laser etching, or photolithography. Sufficient grayscale contrast should exist between the target background and the positioning stripes; when used on reflective surfaces, diffuse coatings or polarized illumination can be employed. An offset distance should be set between the target center and the rotation axis to ensure that the center trajectory covers a sufficient pixel radius within one complete rotation cycle. A larger offset distance generally results in a greater image displacement per unit angle, but it is crucial to ensure that the trajectory remains within the field of view and does not cause occlusion.
[0046] S2. Preprocess the acquired images, perform coarse localization of candidate bands, and determine the coding direction. S201, For the grayscale image of frame t Perform grayscale normalization to obtain Subsequently, an adaptive thresholding method maximizing inter-class variance was used to obtain binary candidate regions, and candidate strips, arc-shaped coded regions, and direction markers were extracted through 8-connected domain analysis.
[0047] in, and These are the minimum and maximum gray values in the current image or region of interest, respectively, and ε is a small positive number to prevent the denominator from being zero.
[0048] The criteria for selecting connected components should include at least area, aspect ratio, principal axis length, compactness, and distance from the prediction center.
[0049] S202. Perform principal component analysis on the pixel coordinate set of the i-th candidate strip, and take the eigenvector corresponding to the largest eigenvalue as the strip tangential unit vector. The unit vector perpendicular to it is taken as the normal. Relative strip pairing is performed based on the direction difference, positional symmetry, and length consistency among candidate strips to obtain multiple principal axes. Coarse centers are obtained by the ordinary least-squares intersection of these principal axes. This is used to establish the subsequent local search area; S203. Polar coordinate sampling is performed on the arc-shaped coding area 23, and the zero-angle direction is determined by the direction marker 24. The average gray level, arc length, or connectivity state of each coding sector is converted into binary or multi-valued codes to obtain the target identification number. Coding recognition and center localization are completed in parallel; when the coding confidence is insufficient, the identity of the previous frame can be used and the quality score of the current frame can be reduced.
[0050] S3. Sub-pixel position of the strip centerline based on Gabor local phase extraction. S301. Establish a two-dimensional complex Gabor filter, the expression of which is:
[0051]
[0052] in, Let θ be the spatial wavelength and θ be the carrier direction. For phase offset, Using Gaussian window scale, The aspect ratio of the space. The unit is the imaginary unit. For strip centerline detection, Set to the strip normal direction to maximize the filter's response to grayscale changes in the strip cross-section.
[0053] S302. Convolve the image with the complex Gabor filter to obtain the complex response. :
[0054]
[0055] in, For local amplitude, This is a local phase. For each positioning strip, at several tangential positions... A one-dimensional profile is established by offsetting along the normal direction by q. For a grayscale profile approximately symmetrical about the strip centerline, when the filter center is near the strip centerline, the phase experiences stable positive and negative crossovers near zero. Let adjacent sampling positions... and satisfy Then, the sub-pixel zero-crossing position is obtained through linear interpolation:
[0056] S303, Thus, the centerline sampling point is obtained. ; To avoid false zeros caused by background texture and phase wrapping, only zero-crossing points located near the thick midline, with amplitudes above the threshold, phase crossing slopes reaching the threshold, and unique candidates are retained in each profile.
[0057] S304. Calculate the confidence weight of a single midline point based on local amplitude, phase zero-crossing slope, candidate uniqueness, and coarse geometric consistency:
[0058] in, For magnitude weighting, For phase zero cross slope weights, As the uniqueness weight of the candidates, Weights are assigned to align with the coarse geometric position. Each weight is normalized to 0-1. Candidate points with low magnitude, small slope, multiple competing zeros, or those far from the predicted strip position are weighted less or eliminated.
[0059] S4. The target center is obtained by robust linear fitting and multi-principle weighted intersection.
[0060] S401. Apply the normal line model to the set of centerline points of the i-th positioning strip. Robust fitting is performed using Huber loss:
[0061]
[0062] in, The normal residual from the point to the line. The threshold can be 1 to 3 times the residual scale estimated based on the median absolute deviation. The midline parameters of each strip are obtained through iterative reweighted least squares. The relative strip fitting results are further synthesized into principal axis constraints to reduce the impact of unilateral occlusion on the center solution.
[0063] S402. Suppose there are a total of m effective principal axes, and the unit normal of the i-th principal axis is... The normal distance to the origin Z is Credibility is The target center c is then obtained by the following formula:
[0064]
[0065] in, Composed of the principal axis normal vectors arranged row by row. The diagonal matrix with main axis weights This is the normal distance column vector. The principal axis weights are determined by the number of effective median points, the average Gabor magnitude, the Huber fit residual, relative strip consistency, and the intersection condition number. When If the number of ill-conditioned or effective principal axes is insufficient, the current frame center observation quality is determined to be insufficient.
[0066] S5. Perform real-time computation optimization, frame quality discrimination, and abnormal frame timing correction. S501. To reduce the computational cost of complete 2D Gabor dense convolution, the following combination strategy is preferred: First, the two-dimensional Gabor kernel that satisfies the separability condition is decomposed into two one-dimensional kernels, and then convolved sequentially along the local coordinate direction; Second, normal profiles are established only within the neighborhood of the strips obtained by coarse positioning, and profiles are sparsely selected along the tangential direction; Third, update the region of interest based on the center, velocity, and target bounding box of the previous frame; Fourth, multiple cross-sections in the same direction are organized into a matrix for batch calculation, reducing point-by-point iterations.
[0067] S502, Calculate the quality score of the t-th frame. :
[0068] in, At a minimum, this includes the average Gabor amplitude, effective zero-crossing ratio, effective stripe count, Huber fit residual, center intersection residual, coding confidence, and consistency with temporal predictions. The quality threshold can be set to 0.4~0.8, specifically determined by the normal frame distribution of the calibration sequence.
[0069] S503, when When the centrality exceeds the threshold but the centrality innovation is within the statistical threshold, the original centrality observation is used directly. ;when When the value is below the threshold or the innovation level is abnormal, the center value predicted from the previous frame's state is used. And utilize the currently valid principal axis normal offset to Perform geometric corrections.
[0070] The integration center is:
[0071] in, The fusion coefficient increases monotonically with the mass fraction. Time series prediction can employ a constant velocity or constant acceleration state model, and can be achieved through Kalman filtering, fixed interval smoothing, or equivalent robust filtering. The purpose of this step is not to excessively low-pass process the true angular displacement, but to suppress non-physical jumps caused by anomalous positioning while preserving the effective dynamic response.
[0072] S6. Collect the complete rotation trajectory and establish a Fourier eccentric trajectory model.
[0073] S601. During the calibration phase, ensure the rotating object completes at least one full rotation cycle and simultaneously acquire the target center sequence. When a reference encoder or drive angle exists, the reference angle is directly used as the calibration phase. When there is no external reference, the polar angle between the center and the coarse trajectory center can be used to form a monotonic initial phase, and then the calibration phase can be obtained by uniformizing the trajectory arc length and unwinding the phase. S602. The x and y coordinates of the calibration trajectory are described using K-order Fourier series respectively:
[0074]
[0075] in, , , , , and These are the model coefficients obtained through least squares or weighted least squares. When =1, the model can describe an elliptical trajectory; When the value is >1, it can be further described that the non-ideal trajectory is caused by installation error, perspective projection, end face runout, and periodic structural disturbance.
[0076] S603, Regarding candidate order Calculate the Bayesian Information Criterion (BIC) from 1 to 10 to avoid underfitting and overfitting.
[0077] in, To determine the sample size, For the sum of squared residuals of the K-order model, This represents the number of model parameters.
[0078] Choose the order with the smallest BIC. As the final trajectory model order, after the model is completed, the model coefficients, effective phase range, and calibration residual distribution are saved for quality assessment during the measurement phase.
[0079] S7. Back-project the center to be measured onto the calibration trajectory and output continuous angular displacement. S701. During the measurement phase, the above center localization and quality judgment process is repeated for the t-th frame image to obtain the center observation. ;Will Back projection onto the calibration trajectory Above, optimal phase It can be obtained from the following formula:
[0080] The preferred approach is to first perform a coarse search using a uniform grid within the range of 0 to 2π, and then perform local one-dimensional optimization using the phase corresponding to the minimum distance as the initial value. For continuous video, the unwound phase prediction value of the previous frame can also be used as a search prior to restrict the local search to a phase window that meets the maximum angular velocity.
[0081] S702. Calculate the back projection error corresponding to the center to be measured:
[0082] Back projection error is used to evaluate the consistency between the current center point and the calibration trajectory. If... If the error significantly exceeds the calibrated residual upper quantile or the set threshold, the confidence level of the current frame phase is reduced, and a time-series predicted phase, neighboring frame interpolation, or repositioning strategy is adopted.
[0083] S703, due to Constrained within the range of 0 to 2π, a phase jump occurs when crossing the periodic boundary. Therefore, the phase sequence is unwound to obtain continuous phase. Taking the initial frame as the zero point, the angular displacement is:
[0084]
[0085] S704, depending on application requirements, can be modified... Angular velocity and angular acceleration can be obtained by performing finite difference or state estimation; for step drive, rise time, overshoot, steady-state error and hysteresis can be further calculated; for periodic motion, the dominant frequency and harmonic characteristics can be obtained through spectrum or time-frequency analysis.
[0086] S8. Determine the engineering implementation parameters and applicable boundaries based on the target imaging size, lens characteristics, sampling frame rate, angular velocity, and expected accuracy.
[0087] S801. All key parameters are set using range values, and the final values are determined through calibration sequences. Recommended ranges are shown in Table 1.
[0088] Table 1 Recommended range of engineering implementation parameters
[0089] S802. When the target rotates at high speed, causing motion blur in a single frame to exceed the strip width, the target is completely occluded for a long period, the trajectory intersects itself, or there is unmodeled relative motion between the camera and the target, the frame rate should be increased, the exposure time shortened, the target size increased, the trajectory recalibrated, or camera motion compensation introduced. For multi-target scenes, arc coding can be used to distinguish each target, and the trajectory model and quality parameters can be stored independently for each target.
[0090] This invention provides a terminal device comprising a processor and a memory. The memory stores a computer program, which includes program instructions. The processor executes the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, graphics processing units (GPUs), tensor processing units (TPUs), digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions to achieve a corresponding method flow or function. The processor described in this embodiment can be used for the operation of a high-precision angular displacement measurement method based on a phase target, including: A phase target with multiple radial positioning stripes is placed at an off-center position on the rotating object, and the target image is acquired. The target image is preprocessed and coarsely located using positioning strips to obtain the coarse center and the normals of each positioning strip; Complex Gabor filtering is performed along the normal of each positioning strip. Sub-pixel points of the strip centerline are determined based on local phase zero crossings. Robust straight line fitting is performed on the sub-pixel points. The multiple principal axes formed by the relative positioning strips are weighted and intersected to obtain the original center observation. The quality of center observations is determined based on Gabor amplitude, effective zero-crossing ratio, effective strip number, robust fitting residual, and center intersection residual. Combining the consistency between the original center observations and the time-series predictions, a selection or fusion is made between the original center observations and the centers based on the state predictions of the previous frame and corrected by the effective principal axis normal constraint to obtain the fused center. During the calibration phase, the fusion center sequence of at least one complete rotation cycle is obtained and the corresponding calibration phase is determined. A Fourier eccentric trajectory model of the horizontal and vertical coordinates as a function of the calibration phase is established. During the measurement phase, the current fusion center is back-projected onto the Fourier eccentric trajectory model. The phase corresponding to the minimum distance between the center and the trajectory point is taken as the current phase. The current phase of the continuous frames is unwound, and the angular displacement is obtained based on the change of the unwound phase relative to the initial frame phase.
[0091] Please see Figure 6 The terminal device is a computer device. In this embodiment, the computer device 60 includes a processor 61, a memory 62, and a computer program 63 stored in the memory 62 and executable on the processor 61. When executed by the processor 61, the computer program 63 implements the high-precision angular displacement measurement method based on a phase target as described in this embodiment. To avoid repetition, these details are not elaborated here. Alternatively, when executed by the processor 61, the computer program 63 implements the functions of each model / unit in the high-precision angular displacement measurement system based on a phase target as described in this embodiment. To avoid repetition, these details are not elaborated here.
[0092] Computer device 60 can be a desktop computer, laptop, handheld computer, cloud server, or other computing device. Computer device 60 may include, but is not limited to, a processor 61 and a memory 62. Those skilled in the art will understand that... Figure 6 This is merely an example of computer device 60 and does not constitute a limitation on computer device 60. It may include more or fewer components than shown, or combine certain components, or different components. For example, computer device may also include input / output devices, network access devices, buses, etc.
[0093] The processor 61 may be a Central Processing Unit (CPU), or other general-purpose processors, graphics processing units (GPUs), tensor processing units (TPUs), digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.
[0094] The memory 62 can be an internal storage unit of the computer device 60, such as a hard disk or memory of the computer device 60. The memory 62 can also be an external storage device of the computer device 60, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. equipped on the computer device 60.
[0095] Furthermore, the memory 62 may include both internal storage units of the computer device 60 and external storage devices. The memory 62 is used to store computer programs and other programs and data required by the computer device. The memory 62 can also be used to temporarily store data that has been output or will be output.
[0096] Please see Figure 7 The terminal device is an electronic device 600, which is manifested in the form of a general-purpose computing device. The components of the electronic device may include, but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different platform components (including storage unit 620 and processing unit 610), a display unit 640, etc.
[0097] The storage unit stores program code, which can be executed by the processing unit 610 to perform the steps described in the method section of this specification according to various exemplary embodiments of the present invention. For example, the processing unit 610 can perform actions such as... Figure 5 The steps are shown in the figure.
[0098] Storage unit 620 may include readable media in the form of volatile storage units, such as random access memory (RAM) 6201 and / or cache memory 6202, and may further include read-only memory (ROM) 6203.
[0099] Storage unit 620 may also include a program / utility 6204 having a set (at least one) program module 6205, such program module 6205 including but not limited to: operating system, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.
[0100] Bus 630 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the multiple bus structures.
[0101] Electronic device 600 can also communicate with one or more external devices 700 (e.g., keyboard, pointing device, Bluetooth device, etc.), and with one or more devices that enable a user to interact with electronic device 600, and / or with any device that enables electronic device 600 to communicate with one or more other computing devices (e.g., router, modem). This communication can be performed via input / output interface 650. Furthermore, electronic device 600 can also communicate with one or more networks (e.g., local area network, wide area network, and / or public network, such as the Internet) via network adapter 660. Network adapter 660 can communicate with other modules of electronic device 600 via bus 630. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms.
[0102] This invention also provides a storage medium, specifically a computer-readable storage medium, which is a memory device in a terminal device for storing programs and data. It is understood that the computer-readable storage medium here can include both built-in storage media in the terminal device and extended storage media supported by the terminal device; it can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, the storage space also stores one or more instructions suitable for loading and execution by a processor, which can be one or more computer programs (including program code). More specific examples of the computer-readable storage medium include: an electrical connection with one or more wires, a portable disk, a hard disk, random access memory, read-only memory, erasable programmable read-only memory, optical fiber, portable compact disk read-only memory, optical storage device, magnetic storage device, or any suitable combination thereof.
[0103] Computer-readable storage media also include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable storage medium can also be any readable medium other than a readable storage medium that can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium can be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, radio frequency, etc., or any suitable combination thereof.
[0104] Program code for performing the operations of this invention can be written in any combination of one or more programming languages, including object-oriented programming languages such as Java and C++, and conventional procedural programming languages such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).
[0105] One or more instructions stored in a computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the high-precision angular displacement measurement method based on a phase target in the above embodiments; one or more instructions in the computer-readable storage medium are loaded and executed by the processor in the following steps: A phase target with multiple radial positioning stripes is placed at an off-center position on the rotating object, and the target image is acquired. The target image is preprocessed and coarsely located using positioning strips to obtain the coarse center and the normals of each positioning strip; Complex Gabor filtering is performed along the normal of each positioning strip. Sub-pixel points of the strip centerline are determined based on local phase zero crossings. Robust straight line fitting is performed on the sub-pixel points. The multiple principal axes formed by the relative positioning strips are weighted and intersected to obtain the original center observation. The quality of center observations is determined based on Gabor amplitude, effective zero-crossing ratio, effective strip number, robust fitting residual, and center intersection residual. Combining the consistency between the original center observations and the time-series predictions, a selection or fusion is made between the original center observations and the centers based on the state predictions of the previous frame and corrected by the effective principal axis normal constraint to obtain the fused center. During the calibration phase, the fusion center sequence of at least one complete rotation cycle is obtained and the corresponding calibration phase is determined. A Fourier eccentric trajectory model of the horizontal and vertical coordinates as a function of the calibration phase is established. During the measurement phase, the current fusion center is back-projected onto the Fourier eccentric trajectory model. The phase corresponding to the minimum distance between the center and the trajectory point is taken as the current phase. The current phase of the continuous frames is unwound, and the angular displacement is obtained based on the change of the unwound phase relative to the initial frame phase.
[0106] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.
[0107] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.
[0108] Theoretically, this invention utilizes the closed periodic trajectory formed by the center of the eccentric target in the image plane to transform the solution of rotational angular displacement into a phase inversion problem on a two-dimensional center trajectory; the propagation relationship of center positioning error to angular error is as follows.
[0109] Assume the true center is located on the trajectory Above, the center positioning error is ε, and the observation point satisfies By linearizing equation (17) to the vicinity of the true phase, the phase error can be approximated as:
[0110] As can be seen from equation (21), the component of the center positioning error along the tangential direction of the trajectory will directly form the phase error. For an ideal circular eccentric trajectory with pixel radius , the magnitude of the tangential derivative of the trajectory with respect to the phase satisfies Therefore, the standard deviation of the phase error can be approximately expressed as: ,in, The standard deviation for tangential center positioning is given. Therefore, within the limits of the camera's field of view and the target's imaging size, increasing the pixel radius of the eccentric trajectory, improving the target center positioning accuracy, and reducing the jitter of the center position along the tangential direction of the trajectory can all reduce phase estimation errors, thereby improving the accuracy of angular displacement measurement.
[0111] For non-ideal eccentric trajectories, during the calibration phase, the trajectory and its tangential derivative magnitude are... A check is performed. When the modulus is too small within a certain phase interval, the same center positioning error will produce a large phase estimation error, which is not conducive to the stable inversion of angular displacement. Therefore, the trajectory changes within each phase interval can be kept sufficiently obvious by adjusting the target eccentricity distance, the target installation position, or the camera observation direction, and local overlap or self-intersection of trajectories can be avoided.
[0112] When performing the back projection solution from the current center position to the calibration trajectory, the search range of the current phase can be limited by combining the phase result of the previous moment, or the constraint of continuous phase change at adjacent moments can be added to reduce local multiple solutions, incorrect matching and phase abrupt change, thereby obtaining continuous and stable angular displacement measurement results.
[0113] Therefore, within the permissible field of view, increasing the effective pixel radius of the eccentric trajectory, improving the local phase center positioning accuracy, and compensating for installation eccentricity, perspective projection, and periodic trajectory distortion in the trajectory model can reduce angular displacement inversion errors; quality discrimination and timing optimization are used to prevent abnormal center points from being directly transmitted to the phase solution results.
[0114] To verify the beneficial effects of the method of the present invention, experiments or simulations were conducted to verify the accuracy of long-distance center positioning, real-time computing efficiency, continuity under image degradation conditions, and the ability to invert eccentric trajectory angular displacement.
[0115] (1) Verification of long-distance micro-displacement and center positioning accuracy.
[0116] A micro-displacement experimental system was constructed using an industrial camera, an electric displacement stage, a composite coded phase target, and an optical support platform, with a working distance set at 2.239 m. The displacement stage sequentially performed multi-level steps of 0.5 mm, 0.2 mm, 0.05 mm, and 0.02 mm, continuously acquiring images during each pause. The algorithm output the target center frame by frame and converted it into physical displacement based on the pixel-to-actual displacement ratio.
[0117] The residuals of the dwell interval were evaluated using standard deviation, root mean square error, mean absolute error, 95th percentile absolute error, and maximum absolute error. The statistical results are shown in Table 2.
[0118] Table 2. Statistics of errors during long-distance micro-displacement dwell phases
[0119] This embodiment demonstrates that phase centerline extraction and multi-principle intersection of composite targets can stably output the center trajectory under conditions of long distance, small imaging size, and weak pixel step. The above results only correspond to the camera, lens, illumination, working distance, and displacement stage conditions of this embodiment and do not constitute a limitation on the accuracy of other systems.
[0120] (2) Continuity verification under occlusion and image degradation conditions.
[0121] Local occlusion, slight blurring, low contrast, local noise, and stripe interference were applied to the original image, and the results of directly using single-frame center trajectories were compared with those obtained by employing quality discrimination and temporal optimization. The quality discrimination module used Gabor amplitude, phase zero-crossing efficiency, ray fitting residual, effective principal axis count, and center intersection residual to identify anomalous frames.
[0122] When the quality score falls below a threshold, the system does not directly adopt the abnormal center. Instead, it uses the predicted state center as the initial value and performs robust correction using the still-valid principal axis normal constraint. The original observation-dominated system is then restored upon the arrival of the next normal frame. In the experiment, the root mean square value of jitter in the dwell segment of the continuous sequence decreased by 32.03% compared to the unoptimized result, indicating that quality-aware switching can improve the continuity of the center trajectory.
[0123] (3) Simulation of eccentric trajectory angular displacement and verification by rotary table.
[0124] To verify the ability to invert two-dimensional center trajectory to one-dimensional angular displacement, a real target image was first acquired as the background. Test images were then generated according to a given eccentric trajectory and angular displacement sequence. The simulation included six angular displacement steps, each held for 15 frames, with an angular interval of 0.005° between adjacent steps. After completing center localization, Fourier trajectory modeling, backprojection phase calculation, phase dewinding, and zero-point alignment, the algorithm was able to distinguish adjacent steps.
[0125] Furthermore, a composite coded phase target is mounted at an eccentric position on a high-precision electric rotary table. During the calibration phase, complete rotational images are acquired and a Fourier eccentric trajectory model is established; during the measurement phase, the rotary table performs an angular step, and an industrial camera continuously acquires images. Under actual measurement conditions, this method can stably identify angular displacement steps of 0.01°. This result demonstrates that replacing the ideal circular model with an actual trajectory model can achieve feasible angle inversion under realistic optical imaging and installation error conditions.
[0126] Example 1: Center trajectory recovery under occlusion and image degradation conditions.
[0127] Local occlusion, slight blurring, low contrast, local noise, and stripe interference are applied to the original image, and the frame quality score is calculated according to step S5. When the quality score is lower than the threshold obtained by normal frame calibration, robust correction is performed using the state prediction center as the initial value and the currently effective principal axis normal constraint, and the original observation dominance is restored after the next normal frame arrives. In this embodiment, the root mean square value of jitter in the continuous sequence dwell segment is reduced by 32.03% compared with the unoptimized result.
[0128] Example 2: Eccentric trajectory angular displacement inversion.
[0129] First, using real target images as the background, test images are generated according to a given eccentric trajectory and angular displacement sequence. Six angular displacement steps are set, each step is maintained for 15 frames, and the angular interval between adjacent steps is 0.005°. Fourier trajectory modeling, back-projection phase calculation, phase unwinding, and zero-point alignment are completed according to steps S6 and S7. Then, the composite coded phase target is installed at the eccentric position of a high-precision electric rotary table. After calibrating the complete rotation trajectory, an angular step is performed. Actual measurements show that it can stably identify angular displacement steps of 0.01°.
[0130] Example 3: Measurement of dynamic angular displacement of a small servo motor.
[0131] A composite coded phase target is fixed at an eccentric position on the output end of a small servo motor. The servo motor control program sends angular displacement commands, and an industrial camera simultaneously acquires image sequences. During the calibration phase, the servo motor or auxiliary turntable covers the entire angular range and establishes an eccentric trajectory model. During the operation phase, the target center is extracted in real time, and the angular displacement is inverted.
[0132] This embodiment obtains continuous servo motor angular displacement response curves, which can be used to observe non-ideal dynamic characteristics such as start-up hysteresis, backlash, control dead zone, and speed fluctuations. This demonstrates that this method can not only be used for angular displacement measurement but also provide graphical data for dynamic performance evaluation and control parameter optimization of servos and other rotary actuators.
[0133] In summary, this invention presents a high-precision angular displacement measurement method and system based on a phase target. By integrating target structure design, sub-pixel center positioning, center quality control, and trajectory phase inversion, it achieves a complete measurement chain from image feature acquisition to angular displacement output. Compared to traditional measurement methods that rely on photoelectric encoders, gratings, or ideal geometric models, this invention utilizes complex Gabor local phase zero-crossing to improve positioning accuracy, reduces the impact of local occlusion, noise, and imaging degradation through multi-directional positioning strips and robust geometric intersections, avoids abnormal center observations from directly entering angle calculations through quality discrimination and timing correction mechanisms, and improves stability during continuous measurement. By acquiring the real rotation trajectory and establishing a Fourier eccentric trajectory model, it compensates for errors caused by installation eccentricity, perspective projection, rotation axis tilt, and periodic trajectory disturbances, enabling the two-dimensional center trajectory to be mapped to the rotation phase more accurately.
[0134] Therefore, this invention does not require a rigid mechanical connection with the rotating mechanism under test. It can complete non-contact angular displacement measurement using only a camera, a phase target, and a processing terminal. It has the advantages of high positioning accuracy, strong anti-interference ability, good continuity, wide applicability, and traceability of the measurement process. It can meet the high-precision angular displacement measurement needs in scenarios such as aerospace, precision manufacturing, optical payloads, and rotating actuator status detection.
[0135] The above content is only for illustrating the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made to the technical solution based on the technical concept proposed in this invention shall fall within the scope of protection of the claims of this invention.
Claims
1. A high-precision angular displacement measurement method based on a phase target, characterized in that, Includes the following steps: A phase target with multiple radial positioning stripes is placed at an off-center position on the rotating object, and the target image is acquired. The target image is preprocessed and coarsely located using positioning strips to obtain the coarse center and the normals of each positioning strip; Complex Gabor filtering is performed along the normal of each positioning strip. Sub-pixel points of the strip centerline are determined based on local phase zero crossings. Robust straight line fitting is performed on the sub-pixel points. The multiple principal axes formed by the relative positioning strips are weighted and intersected to obtain the original center observation. The quality of center observations is determined based on Gabor amplitude, effective zero-crossing ratio, effective strip number, robust fitting residual, and center intersection residual. Combining the consistency between the original center observations and the time-series predictions, a selection or fusion is made between the original center observations and the centers based on the state predictions of the previous frame and corrected by the effective principal axis normal constraint to obtain the fused center. During the calibration phase, the fusion center sequence of at least one complete rotation cycle is obtained and the corresponding calibration phase is determined. A Fourier eccentric trajectory model of the horizontal and vertical coordinates as a function of the calibration phase is established. During the measurement phase, the current fusion center is back-projected onto the Fourier eccentric trajectory model. The phase corresponding to the minimum distance between the center and the trajectory point is taken as the current phase. The current phase of the continuous frames is unwound, and the angular displacement is obtained based on the change of the unwound phase relative to the initial frame phase.
2. The high-precision angular displacement measurement method based on a phase target according to claim 1, characterized in that, The phase target includes radial positioning strips, a central reference area, an arc-shaped coding area, and direction markers; The radial positioning strips are used to provide center positioning constraints in multiple directions, the center reference area is used to determine the target center reference area, the arc-shaped coding area is used to represent target identity information, and the direction mark is used to determine the starting direction of coding reading.
3. The high-precision angular displacement measurement method based on a phase target according to claim 2, characterized in that, The radial positioning strips include six rectangular strips, with a preset angle between adjacent positioning strips, and the relative positioning strips form multiple central positioning axes; Based on the directional differences, positional symmetry, and length consistency among multiple positioning strips, the positioning strips are paired to form multi-directional geometric constraints for center calculation.
4. The high-precision angular displacement measurement method based on a phase target according to claim 1, characterized in that, The determination of the sub-pixel position of the midline of the positioning strip based on complex Gabor filtering includes: The direction parameters of the complex Gabor filter are set according to the direction of the positioning strip so that the complex Gabor filter can produce the maximum response to the grayscale changes of the cross-section of the positioning strip. A one-dimensional grayscale profile is established along the normal direction of the positioning strip. The zero-crossing interval is determined based on the local phase sign change corresponding to adjacent sampling positions, and the sub-pixel midline position is obtained through linear interpolation.
5. The high-precision angular displacement measurement method based on a phase target according to claim 4, characterized in that, The reliability of the obtained sub-pixel midline positions is evaluated, and the reliability is determined based on at least the following parameters: The amplitude of the complex Gabor response, the zero-crossing slope of the phase, the uniqueness of the candidate zero-crossing, and the consistency between the midline position and the coarse localization results; When the confidence level is lower than a preset threshold, the weight of the corresponding sub-pixel midline position is reduced or the corresponding midline position is removed.
6. The high-precision angular displacement measurement method based on a phase target according to claim 1, characterized in that, The robust straight-line fitting uses the Huber loss function to iteratively weight and fit the set of midline points of the positioning strip. The weights of outlier points are adjusted based on the normal residual from the point to the fitted straight line to reduce the impact of local occlusion, noise, and strip degradation on the center calculation.
7. The high-precision angular displacement measurement method based on a phase target according to claim 1, characterized in that, The quality score also includes the coding confidence and the consistency between the current central observation and the time-series prediction; When the quality score is higher than the preset threshold and the central innovation quantity is within the statistical threshold range, the current frame center observation value is used. When the quality score is lower than the preset threshold or the central innovation exceeds the statistical threshold, the fusion center is determined by combining the predicted center and the effective principal axis constraint.
8. The high-precision angular displacement measurement method based on a phase target according to claim 1, characterized in that, The Fourier eccentric trajectory model includes: Based on the calibration phase, establish Fourier series models corresponding to the abscissa and ordinate of the fusion center respectively; The Bayesian information criterion is calculated for the orders of multiple candidate models, and the final trajectory model order is determined according to the minimum principle of the Bayesian information criterion.
9. The high-precision angular displacement measurement method based on a phase target according to claim 1, characterized in that, The step of projecting the fusion center onto the Fourier eccentric trajectory model includes: Perform trajectory search within the preset phase range to determine the phase corresponding to the trajectory point with the smallest distance from the current fusion center; The current phase search range is predicted based on the phase result after unwinding in the previous frame, and the current phase search window is limited by the maximum angular velocity. The reliability of the current phase is evaluated based on the backprojection error between the current fusion center and the trajectory model.
10. A high-precision angular displacement measurement system based on a phase target, characterized in that, include: An industrial camera (1) is used to acquire images of a composite coded phase target mounted on a rotating object; The composite coded phase target (2) is set at the off-center position of the rotating object (3), including radial positioning strips (21), central reference area (22), arc-shaped coding area (23) and direction mark (24). Illumination unit (4) is used to provide stable illumination for the composite coded phase target; The processing terminal (5) is communicatively connected to the industrial camera (1) and is used to execute the method described in any one of claims 1 to 9 to obtain the continuous angular displacement of the rotating object.
