Part shaft hole deviation detection method

Abstract

The invention discloses a part shaft hole shaft deviation detection method, and relates to the technical field of detection, and the detection method comprises the steps: S1, placing a part to be detected on a positioning assembly, positioning and fixing the part to be detected through the positioning assembly, and installing a linear module driven by a servo motor at the bottom of the positioning assembly, the position of a part positioning hole is shot through an industrial camera, the distance needing to be adjusted of a positioning pin is automatically calculated, and a linear module drives the positioning pin to complete position self-adaptive correction; s2, carrying out image acquisition based on a camera, and obtaining image data of a to-be-inspected part; the linear module driven by the servo motor is matched with the industrial camera, the position of the positioning pin is automatically corrected, errors caused by manual or fixed positioning are eliminated, an accurate reference is laid for follow-up inspection, whether the positioning pin is qualified or not is rapidly distinguished, traceability is achieved when the positioning pin is unqualified, the direction is provided for process adjustment, precise quality control is assisted, and circulation of unqualified products is reduced.

Description

Technical Field

[0001] This invention relates to the field of testing technology, specifically to a method for inspecting the axial deviation of a component's shaft hole. Background Technology

[0002] In the field of mechanical manufacturing, the accuracy of the shaft and hole fit of components directly determines the assembly quality and operational reliability of equipment. Shaft and hole misalignment is one of the core defects affecting fit accuracy. Therefore, efficient and accurate inspection of shaft and hole misalignment of components is crucial. Current mainstream methods for inspecting shaft and hole misalignment of components generally suffer from insufficient accuracy in the positioning stage. Traditional positioning components often use fixed-size positioning pins or manual adjustment structures. When dealing with components of different specifications and batches, repeated manual disassembly, replacement of positioning components, and recalibration are required. This is not only cumbersome and time-consuming, but also prone to positioning deviations due to human error, thus affecting the accuracy of subsequent inspection results. Furthermore, it is difficult to address positioning challenges caused by machining errors or surface condition differences in component positioning holes. Therefore, we propose a new method for inspecting shaft and hole misalignment of components. Summary of the Invention

[0003] To solve the above-mentioned technical problems, a method for inspecting shaft and hole misalignment of components is provided. This technical solution solves the above-mentioned problems.

[0004] To achieve the above objectives, the technical solution adopted by this invention is: a method for inspecting shaft misalignment of a component's shaft hole, wherein the inspection method is as follows:

[0005] S1. Place the parts to be inspected on the positioning assembly. The positioning assembly positions and fixes the parts to be inspected. The bottom of the positioning assembly is equipped with a linear module driven by a servo motor. The position of the positioning hole of the parts is photographed by an industrial camera. The distance that the positioning pin needs to be adjusted is automatically calculated. The linear module drives the positioning pin to complete the position adaptive correction.

[0006] S2. Based on the camera, image data of the parts to be inspected is acquired, and the image data is preprocessed to construct a dataset;

[0007] S3. Obtain the dataset and generate a hologram of the hole and shaft area of ​​the component based on holographic technology. Analyze the interference fringes of the hologram and extract feature parameters, including the center coordinates of the hole, the axis coordinates of the shaft, and the diameters of the hole and the shaft.

[0008] S4. Calculate the offset of the hole and shaft based on the center coordinates of the extracted hole and the axis center coordinates; compare the calculated offset with the preset offset threshold. If the offset is less than or equal to the threshold, the hole and shaft of the component are deemed to be qualified; if the offset is greater than the threshold, the hole and shaft of the component are deemed to be unqualified.

[0009] Preferably, the adaptive correction step in step S1 is as follows:

[0010] Place the parts to be inspected on the positioning component worktable, initially align them with the rough positioning marks, and trigger the industrial camera to capture a panoramic image of the parts.

[0011] Upload the image, focus and identify the positioning hole area on the component, perform edge detection on the acquired image, extract the actual contour of the positioning hole, calculate the geometric center coordinates of the positioning hole, and record the position of the coordinates in the image coordinate system;

[0012] Retrieve the standard CAD data of this model of component, obtain the theoretical center coordinates of the positioning hole, and automatically calculate the deviation between the actual coordinates and the theoretical coordinates;

[0013] If the deviation exceeds the allowable range of positioning accuracy, the automatic correction program is started, and pulse signals are sent to the servo motor to drive the X and Y axes of the linear module to move along the planned path. When the positioning pin moves to the theoretical coordinate position, the motor stops moving, and the initial correction is completed.

[0014] The industrial camera takes another picture of the positioning hole to check the alignment status of the positioning pin and the positioning hole after correction.

[0015] If a slight deviation still exists, send a fine-tuning command to drive the positioning pin at a low speed to achieve the required positioning accuracy.

[0016] After calibration, the locking device of the positioning component locks the position of the positioning pin to fix it in place.

[0017] Preferably, the deviation value is obtained by calculating the deviation in the X-axis and Y-axis directions and then combining the results; the formula for calculating the deviation in the X-axis direction is: ;in The X coordinates of the actual center of the positioning hole obtained from image recognition. The X-coordinate of the theoretical center of the positioning hole in standard CAD data, where ΔX is the deviation in the X-axis direction;

[0018] The formula for calculating the Y-axis deviation is: ,in The Y-coordinate of the actual center of the positioning hole obtained from image recognition. The Y-coordinate of the theoretical center of the positioning hole in the standard CAD data, where ΔY is the deviation in the Y-axis direction;

[0019] The formula for calculating the overall deviation distance is: Where D is the deviation of the straight-line distance between the actual center and the theoretical center, used to determine whether it exceeds the allowable range of accuracy.

[0020] Preferably, the data preprocessing in step S2 includes image denoising, contrast enhancement, geometric correction, image segmentation, and data standardization.

[0021] Preferably, the hologram generation step in step S3 is as follows:

[0022] Using a Mach-Zehnder interferometric optical path, the laser output from the helium-neon laser is divided into an object beam and a reference beam. The object beam illuminates the hole axis region of the component to be inspected, and after reflection and transmission from the hole axis surface, it carries geometric information to the holographic plate. The reference beam directly illuminates the holographic plate and interferes with the object beam on the plate. After exposure, a hologram of the hole axis region is generated.

[0023] An industrial camera is used to capture holographic plates, converting the optical hologram into a digital image; a grayscale correction algorithm is used to eliminate grayscale deviations caused by uneven camera sensor response; and a Fourier transform is performed on the digital hologram.

[0024] Preferably, in step S3, the interference fringe analysis separates the fringes from the background using an adaptive threshold segmentation method, extracts the centerline of each interference fringe using a skeleton extraction algorithm, and obtains a continuous fringe centerline network. The edges of the aperture and the axis correspond to specific fringe curvature features. The circular edge of the aperture causes the optical path difference of the object beam to change periodically, forming annular interference fringes centered on the aperture. The fringe density decreases as the distance from the aperture center increases. The cylindrical surface of the axis corresponds to parallel and uniformly dense interference fringes. Changes in the diameter of the axis cause changes in the fringe spacing. Curvature fitting is performed on the annular fringes to determine the center position of each annular fringe. Linear fitting is performed on the parallel fringes to obtain the fringe tilt angle and spacing.

[0025] Preferably, in step S3, the aperture axis characteristic parameters are obtained by selecting three annular interference fringes with different radii, fitting the center of each fringe, and taking the average coordinates of the three centers as the actual center coordinates of the aperture to eliminate single-point deviations caused by fringe noise. The distance Δd between two adjacent annular fringes is measured. Based on the relationship between the interference fringe distance and the optical path difference of the object light, Δd = λ / (2sinθ), where θ is the incident angle of the object light and λ is the laser wavelength. Combined with the aperture radius position corresponding to the fringe, the distance from the edge of the aperture to the center is calculated, i.e., the radius of the aperture, to obtain the diameter. radius;

[0026] Three sets of parallel interference fringes corresponding to the axis are selected, and sampling points are taken for each set of fringes. The straight line equation of each fringe is obtained by fitting. The perpendicular bisectors of two adjacent parallel fringes are calculated, and the intersection of the three perpendicular bisectors is the coordinate of the axis center.

[0027] The spacing Δs of the parallel interference fringes corresponding to the measured axis is determined. Based on the imaging principle of the cylindrical surface of the axis, the fringe spacing Δs is related to the diameter of the axis. Satisfying Relationship: Where n is the air refractive index and α is the reference light incident angle, the diameter of the shaft is calculated by substituting the known parameters.

[0028] Preferably, in step S4, the offset calculation is performed by using a unified coordinate reference to transform the coordinates of the hole center and the shaft center to the same coordinate system, decompose the offset components of each axis, calculate the horizontal and vertical offsets, clarify the degree of offset and positional relationship in each direction, calculate the comprehensive offset according to the matching scenario, verify the results, calculate the offset direction angle as required, and organize the obtained data storage.

[0029] Preferably, in step S4, the matching threshold is retrieved from the system database based on the component model, assembly scenario, and design requirements; the comprehensive offset is compared with the threshold; when the offset is less than or equal to the threshold, the component is deemed qualified, its status is marked, and the value is recorded for traceability; when the offset is greater than the threshold, the component is deemed unqualified, and an anomaly prompt is triggered; for components that are qualified but whose offset is close to the threshold, they are marked for attention, and the deviation trend is tracked in subsequent batch inspections.

[0030] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0031] This invention utilizes a servo motor-driven linear module in conjunction with an industrial camera to automatically correct the position of positioning pins, eliminating errors from manual or fixed positioning and establishing a precise benchmark for subsequent inspection. Feature extraction employs holographic technology to analyze interference fringes and obtain hole axis parameters with sub-micron precision, far exceeding traditional methods. It accurately captures minute geometric features, adapting to the inspection needs of precision components. By comparing offset with a threshold, the standard is clear and quantifiable, and the threshold can be flexibly adjusted according to the component's context, quickly distinguishing between qualified and unqualified products. When unqualified products fail, the problem can be traced, providing direction for process adjustments, facilitating precise quality control, and reducing the flow of non-conforming products. Attached Figure Description

[0032] Figure 1 This is a flowchart of the testing steps of the present invention. Detailed Implementation

[0033] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0034] Reference Figure 1 As shown, a method for inspecting the axial deviation of a component's shaft hole is as follows:

[0035] S1. Place the parts to be inspected on the positioning assembly. The positioning assembly positions and fixes the parts to be inspected. The bottom of the positioning assembly is equipped with a linear module driven by a servo motor. The position of the positioning hole of the parts is photographed by an industrial camera. The distance that the positioning pin needs to be adjusted is automatically calculated. The linear module drives the positioning pin to complete the position adaptive correction.

[0036] S2. Based on the camera, image data of the parts to be inspected is acquired, and the image data is preprocessed to construct a dataset;

[0037] S3. Obtain the dataset and generate a hologram of the hole and shaft area of ​​the component based on holographic technology. Analyze the interference fringes of the hologram and extract feature parameters, including the center coordinates of the hole, the axis coordinates of the shaft, and the diameters of the hole and the shaft.

[0038] S4. Calculate the offset of the hole and shaft based on the center coordinates of the extracted hole and the axis center coordinates; compare the calculated offset with the preset offset threshold. If the offset is less than or equal to the threshold, the hole and shaft of the component are deemed to be qualified; if the offset is greater than the threshold, the hole and shaft of the component are deemed to be unqualified.

[0039] This application features full-process automation to improve efficiency. Servo motors automatically correct positioning and collect, analyze, and judge data, reducing manual intervention and shortening single-piece inspection time by over 50%, thus lowering labor costs. It also offers higher precision, with holographic technology combined with image preprocessing enabling parameter measurement down to the micrometer level, quantifying judgment standards, avoiding subjective human error, and reducing misjudgment rates. Operation is simplified with modular processes, requiring no complex technical skills from personnel, and data is automatically stored for easy traceability and review, reducing training and recording costs. Furthermore, it boasts strong adaptability, allowing for on-demand inspection upon production line integration. Adjusting parameters adapts it to various parts, meeting the needs of both batch and flexible production, making it an inspection solution perfectly suited for intelligent manufacturing.

[0040] The adaptive correction step in step S1 is as follows:

[0041] Place the parts to be inspected on the positioning component worktable, initially align them with the rough positioning marks, and trigger the industrial camera to capture a panoramic image of the parts.

[0042] Upload the image, focus and identify the positioning hole area on the component, perform edge detection on the acquired image, extract the actual contour of the positioning hole, calculate the geometric center coordinates of the positioning hole, and record the position of the coordinates in the image coordinate system;

[0043] Retrieve the standard CAD data of this model of component, obtain the theoretical center coordinates of the positioning hole, and automatically calculate the deviation between the actual coordinates and the theoretical coordinates;

[0044] If the deviation exceeds the allowable range of positioning accuracy, the automatic correction program is started, and pulse signals are sent to the servo motor to drive the X and Y axes of the linear module to move along the planned path. When the positioning pin moves to the theoretical coordinate position, the motor stops moving, and the initial correction is completed.

[0045] The industrial camera takes another picture of the positioning hole to check the alignment status of the positioning pin and the positioning hole after correction.

[0046] If a slight deviation still exists, send a fine-tuning command to drive the positioning pin at a low speed to achieve the required positioning accuracy.

[0047] After calibration, the locking device of the positioning component locks the position of the positioning pin to fix it in place.

[0048] This application process first captures a panoramic image using an industrial camera, then extracts the actual contour and geometric center of the positioning hole through edge detection. Next, it compares the actual contour with standard CAD theoretical coordinates to calculate the deviation, ensuring accurate deviation identification. When the deviation exceeds the limit, a servo motor drives a linear module to move along a planned path. After correction, a second image is captured for inspection. Minor deviations can be fine-tuned at low speeds, ultimately achieving the required positioning accuracy. This provides a precise benchmark for image acquisition, parameter extraction, and pass / fail judgment, avoiding misjudgments in subsequent inspections due to initial positioning deviations. It reduces labor costs and operational barriers, eliminating the need for manual intervention from initial alignment triggering to image analysis, deviation calculation, automatic correction, and locking. The system automatically measures deviations and adjusts positioning pins. Even when fine adjustments are needed, the system automatically sends instructions, requiring operators to simply place the parts and initially align the marks, reducing manual intervention and lowering the skill requirements for personnel. This also avoids the errors and inefficiencies of manual operation. The calibration process forms a closed loop: image recognition, deviation calculation, drive calibration, secondary detection, fine adjustment, and locking. Each step is supported by data and verified by results, ensuring that positioning deviations are fully corrected. After calibration, the locking device locks the positioning pins to prevent parts from shifting during subsequent inspections, ensuring stable positioning and further improving the reliability and consistency of the overall inspection process.

[0049] The deviation value is obtained by calculating the deviation in the X-axis and Y-axis directions and then combining the results; the formula for calculating the deviation in the X-axis direction is: ;in The X coordinates of the actual center of the positioning hole obtained from image recognition. The X-coordinate of the theoretical center of the positioning hole in standard CAD data, where ΔX is the deviation in the X-axis direction;

[0050] The formula for calculating the Y-axis deviation is: ,in The Y-coordinate of the actual center of the positioning hole obtained from image recognition. The Y-coordinate of the theoretical center of the positioning hole in the standard CAD data, where ΔY is the deviation in the Y-axis direction;

[0051] The formula for calculating the overall deviation distance is: Where D is the deviation of the straight-line distance between the actual center and the theoretical center, used to determine whether it exceeds the allowable range of accuracy.

[0052] This application has been approved. Calculating the X and Y axis deviations separately allows for direct location of the deviation in the horizontal or vertical direction, facilitating targeted adjustment of the corresponding axis of the servo motor driving the linear module. This avoids the low correction efficiency and repeated debugging issues caused by the fuzzy adjustment of the overall deviation in traditional methods. Separate axis calculations also reduce the possibility of single-axis deviation calculations masking instances where a single axis deviation exceeds the limit, resulting in more detailed deviation analysis. The comprehensive deviation quantification is more complete, and the accuracy judgment is more reliable. Calculating the comprehensive deviation between the actual and theoretical centers using the straight-line distance formula fully reflects the spatial offset of the positioning hole. Even if the individual X and Y axis deviations are within limits, the comprehensive deviation may still exceed the limit due to the superposition of the two axis deviations. This calculation method avoids the loophole of judging a single axis as qualified, ensuring that the positioning accuracy fully meets the inspection requirements and providing a more reliable benchmark for subsequent image acquisition and parameter extraction. The judgment criteria are objective and unified, reducing subjective errors. Both separate axis deviations and comprehensive deviations are calculated using fixed formulas based on actual and theoretical coordinates, eliminating the need for manual estimation or experience-based judgment. The deviation calculation logic and judgment criteria remain completely consistent when inspecting different operators and different batches of parts, completely eliminating differences in deviation judgment caused by subjective operations and ensuring the uniformity of positioning correction standards and the reliability of inspection results.

[0053] The data preprocessing in step S2 includes image denoising, contrast enhancement, geometric correction, image segmentation, and data standardization.

[0054] The hologram generation steps in step S3 are as follows:

[0055] Using a Mach-Zehnder interferometric optical path, the laser output from the helium-neon laser is divided into an object beam and a reference beam. The object beam illuminates the hole axis region of the component to be inspected, and after reflection and transmission from the hole axis surface, it carries geometric information to the holographic plate. The reference beam directly illuminates the holographic plate and interferes with the object beam on the plate. After exposure, a hologram of the hole axis region is generated.

[0056] An industrial camera is used to capture holographic plates, converting the optical hologram into a digital image; a grayscale correction algorithm is used to eliminate grayscale deviations caused by uneven camera sensor response; and a Fourier transform is performed on the digital hologram.

[0057] By employing a Mach-Zehnder interferometric optical path, the laser beam is divided into an object beam and a reference beam. When the object beam illuminates the aperture axis region, it carries three-dimensional geometric information such as the surface morphology, aperture size, and axis position of the aperture axis. This information is then used to interfere with the reference beam on a holographic plate to form an image. This method differs from traditional two-dimensional images, which only record planar information. It can completely preserve the spatial structural details of the aperture axis. Even tiny aperture deviations and axis offsets can be reflected through differences in the distribution of interference fringes, providing comprehensive data support for the subsequent extraction of the center coordinates and key diameter parameters.

[0058] Achieving the digitization of optical holography improves processing convenience and accuracy. By converting optical holograms into digital images using industrial cameras, the limitations of traditional optical holograms, which rely on physical plates and are difficult to analyze later, are overcome. This allows for precise processing using computer algorithms. Furthermore, the grayscale correction algorithm eliminates grayscale deviations caused by uneven camera sensor response and avoids image brightness imbalances due to hardware differences. This ensures that the grayscale distribution of the digital hologram truly reflects the interference fringe characteristics, providing a high-quality digital image source for subsequent Fourier transforms.

[0059] In step S3, interference fringe analysis separates the fringes from the background using an adaptive threshold segmentation method. A skeleton extraction algorithm is then used to extract the centerline of each interference fringe, resulting in a continuous network of fringe centerlines. The edges of the aperture and axis correspond to specific fringe curvature features. The circular edge of the aperture causes a periodic change in the object beam path difference, forming annular interference fringes centered on the aperture. The fringe density decreases as the distance from the aperture center increases. The cylindrical surface of the axis corresponds to parallel and uniformly dense interference fringes, and changes in the axis diameter alter the fringe spacing. Curvature fitting is performed on the annular fringes to determine the center position of each fringe. Linear fitting is performed on the parallel fringes to obtain the fringe tilt angle and spacing.

[0060] This application employs an adaptive threshold segmentation method, which dynamically adjusts the segmentation threshold based on the grayscale difference between the interference fringes and the background. This avoids the problem that a fixed threshold is difficult to adapt to different regions, efficiently separating the fringes from the complex background. This ensures that subsequent analysis focuses only on the effective fringes information, eliminates the interference of background noise on feature recognition, and improves the accuracy of fringes analysis.

[0061] In step S3, the aperture axis characteristic parameters are obtained by selecting three annular interference fringes with different radii, fitting the center of each fringe, and taking the average coordinates of the three centers as the actual center coordinates of the aperture to eliminate single-point deviations caused by fringe noise. The distance Δd between two adjacent annular fringes is measured. Based on the relationship between the interference fringe spacing and the optical path difference of the object beam, Δd = λ / (2sinθ), where θ is the incident angle of the object beam and λ is the laser wavelength. Combined with the aperture radius positions corresponding to the fringes, the distance from the edge of the aperture to the center is calculated, i.e., the aperture radius, thus obtaining the diameter. radius;

[0062] Three sets of parallel interference fringes corresponding to the axis are selected, and sampling points are taken for each set of fringes. The straight line equation of each fringe is obtained by fitting. The perpendicular bisectors of two adjacent parallel fringes are calculated, and the intersection of the three perpendicular bisectors is the coordinate of the axis center.

[0063] The spacing Δs of the parallel interference fringes corresponding to the measured axis is determined. Based on the imaging principle of the cylindrical surface of the axis, the fringe spacing Δs is related to the diameter of the axis. Satisfying Relationship: Where n is the air refractive index and α is the reference light incident angle, the diameter of the shaft is calculated by substituting the known parameters.

[0064] When obtaining the center coordinates of the hole, this application selects three annular stripes with different radii to fit the center of the hole, and then takes the average coordinates. This effectively offsets the center deviation caused by noise and local interference anomalies of a single stripe, and avoids the random error of single-point fitting. When obtaining the axis center coordinates, the axis center coordinates are determined by the intersection of the perpendicular bisectors of three sets of parallel stripes. The geometric logic of the three lines being at the same point reduces the influence of the fitting deviation of a single set of stripes on the axis center positioning, making the core coordinate parameters of the hole and the axis closer to the actual values.

[0065] In step S4, the offset calculation uses a unified coordinate reference to transform the coordinates of the hole center and the shaft center to the same coordinate system, decomposes the offset components of each axis, calculates the horizontal and vertical offsets, clarifies the degree of offset and positional relationship in each direction, calculates the comprehensive offset according to the matching scenario, verifies the results, calculates the offset direction angle according to the requirements, and organizes the obtained data storage.

[0066] This application transforms the coordinates of the hole center and the axis center to the same coordinate system, completely solving the problem of miscalculation of offset caused by the hole and axis coordinates belonging to different dimensions. It avoids the miscalculation of offset caused by the difference in coordinate systems in the early stage of image acquisition and parameter extraction, and ensures that the subsequent decomposition and comprehensive offset calculation of horizontal and vertical offset components are based on the same spatial dimension, thus ensuring data comparability and calculation accuracy from the source.

[0067] In step S4, the matching threshold is retrieved from the system database based on the component model, assembly scenario, and design requirements. The overall offset is compared with the threshold. When the offset is less than or equal to the threshold, the component is deemed qualified, its status is marked, and the value is recorded for traceability. When the offset is greater than the threshold, the component is deemed unqualified, and an anomaly prompt is triggered. For components that are qualified but whose offset is close to the threshold, they are marked for attention, and the deviation trend is tracked in subsequent batch inspections.

[0068] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A method for inspecting shaft offset of a component's shaft hole, characterized in that, The testing method is as follows: S1. Place the parts to be inspected on the positioning assembly. The positioning assembly positions and fixes the parts to be inspected. The bottom of the positioning assembly is equipped with a linear module driven by a servo motor. The position of the positioning hole of the parts is photographed by an industrial camera. The distance that the positioning pin needs to be adjusted is automatically calculated. The linear module drives the positioning pin to complete the position adaptive correction. S2. Based on the camera, image data of the parts to be inspected is acquired, and the image data is preprocessed to construct a dataset; S3. Obtain the dataset and generate a hologram of the hole and shaft area of ​​the component based on holographic technology. Analyze the interference fringes of the hologram and extract feature parameters, including the center coordinates of the hole, the axis coordinates of the shaft, and the diameters of the hole and the shaft. S4. Calculate the offset of the hole and shaft based on the center coordinates of the extracted hole and the axis center coordinates; compare the calculated offset with the preset offset threshold. If the offset is less than or equal to the threshold, the hole and shaft of the component are deemed to be qualified; if the offset is greater than the threshold, the hole and shaft of the component are deemed to be unqualified.

2. The method for inspecting shaft deviation of a component's shaft hole according to claim 1, characterized in that, The adaptive correction step in step S1 is as follows: Place the parts to be inspected on the positioning component worktable, initially align them with the rough positioning marks, and trigger the industrial camera to capture a panoramic image of the parts. Upload the image, focus and identify the positioning hole area on the component, perform edge detection on the acquired image, extract the actual contour of the positioning hole, calculate the geometric center coordinates of the positioning hole, and record the position of the coordinates in the image coordinate system; Retrieve the standard CAD data of this model of component, obtain the theoretical center coordinates of the positioning hole, and automatically calculate the deviation between the actual coordinates and the theoretical coordinates; If the deviation exceeds the allowable range of positioning accuracy, the automatic correction program is started, and pulse signals are sent to the servo motor to drive the X and Y axes of the linear module to move along the planned path. When the positioning pin moves to the theoretical coordinate position, the motor stops moving, and the initial correction is completed. The industrial camera takes another picture of the positioning hole to check the alignment status of the positioning pin and the positioning hole after correction. If a slight deviation still exists, send a fine-tuning command to drive the positioning pin at a low speed to achieve the required positioning accuracy. After calibration, the locking device of the positioning component locks the position of the positioning pin to fix it in place.

3. The method for inspecting shaft deviation of a component's shaft hole according to claim 2, characterized in that: The deviation value is obtained by calculating the X-axis and Y-axis directions and then combining them. The formula for calculating the X-axis deviation is: ;in The X coordinates of the actual center of the positioning hole obtained from image recognition. The X-coordinate of the theoretical center of the positioning hole in standard CAD data, where ΔX is the deviation in the X-axis direction; The formula for calculating the Y-axis deviation is: ,in The Y-coordinate of the actual center of the positioning hole obtained from image recognition. The Y-coordinate of the theoretical center of the positioning hole in the standard CAD data, where ΔY is the deviation in the Y-axis direction; The formula for calculating the overall deviation distance is: Where D is the deviation of the straight-line distance between the actual center and the theoretical center, used to determine whether it exceeds the allowable range of accuracy.

4. The method for inspecting shaft deviation of a component's shaft hole according to claim 1, characterized in that: The data preprocessing in step S2 includes image denoising, contrast enhancement, geometric correction, image segmentation, and data standardization.

5. The method for inspecting shaft deviation of a component's shaft hole according to claim 1, characterized in that, The hologram generation steps in step S3 are as follows: Using a Mach-Zehnder interferometric optical path, the laser output from the helium-neon laser is divided into an object beam and a reference beam. The object beam illuminates the hole axis region of the component to be inspected, and after reflection and transmission from the hole axis surface, it carries geometric information to the holographic plate. The reference beam directly illuminates the holographic plate and interferes with the object beam on the plate. After exposure, a hologram of the hole axis region is generated. An industrial camera is used to capture holographic plates, converting the optical hologram into a digital image; a grayscale correction algorithm is used to eliminate grayscale deviations caused by uneven camera sensor response; and a Fourier transform is performed on the digital hologram.

6. The method for inspecting shaft offset of a component's shaft hole according to claim 1, characterized in that: In step S3, the interference fringe analysis separates the fringes from the background using an adaptive threshold segmentation method, and extracts the center line of each interference fringe using a skeleton extraction algorithm to obtain a continuous fringe center line network, where the edges of the holes and axes correspond to specific fringe bending features. The circular edge of the aperture causes the optical path difference of the object beam to change periodically, forming annular interference fringes centered on the aperture. The fringe density decreases as the distance from the aperture center increases. The cylindrical surface of the shaft corresponds to parallel and uniformly dense interference fringes. Changes in the diameter of the shaft will cause changes in the fringe spacing. Curvature fitting is performed on the annular stripes to determine the center position of each annular stripe; linear fitting is performed on the parallel stripes to obtain the inclination angle and spacing of the stripes.

7. The method for inspecting shaft deviation of a component's shaft hole according to claim 1, characterized in that: In step S3, the aperture axis characteristic parameters are obtained by selecting three annular interference fringes with different radii, fitting the center of each fringe, and taking the average coordinates of the three centers as the actual center coordinates of the aperture to eliminate single-point deviations caused by fringe noise. The distance Δd between two adjacent annular fringes is measured. Based on the relationship between the interference fringe spacing and the optical path difference of the object beam, Δd = λ / (2sinθ), where θ is the incident angle of the object beam and λ is the laser wavelength. Combined with the aperture radius positions corresponding to the fringes, the distance from the edge of the aperture to the center is calculated, i.e., the aperture radius, thus obtaining the diameter. radius; Three sets of parallel interference fringes corresponding to the axis are selected, and sampling points are taken for each set of fringes. The straight line equation of each fringe is obtained by fitting. The perpendicular bisectors of two adjacent parallel fringes are calculated, and the intersection of the three perpendicular bisectors is the coordinate of the axis center. The spacing Δs of the parallel interference fringes corresponding to the measured axis is determined. Based on the imaging principle of the cylindrical surface of the axis, the fringe spacing Δs is related to the diameter of the axis. Satisfying Relationship: Where n is the air refractive index and α is the reference light incident angle, the diameter of the shaft is calculated by substituting the known parameters.

8. The method for inspecting shaft deviation of a component's shaft hole according to claim 1, characterized in that: In step S4, the offset calculation uses a unified coordinate reference to transform the coordinates of the hole center and the axis center to the same coordinate system, decomposes the offset components of each axis, calculates the horizontal and vertical offsets, clarifies the degree of offset and positional relationship in each direction, calculates the comprehensive offset according to the matching scenario, verifies the results, calculates the offset direction angle as required, and organizes and stores the obtained data.

9. The method for inspecting shaft deviation of a component's shaft hole according to claim 1, characterized in that: In step S4, the matching threshold is retrieved from the system database based on the component model, assembly scenario, and design requirements. The overall offset is compared with the threshold. When the offset is less than or equal to the threshold, the system is deemed qualified, the status is marked, and the value is recorded for traceability. When the offset is greater than the threshold, the system is deemed unqualified, and an exception prompt is triggered. For qualified parts but whose deviation is close to the threshold, mark them for attention and track the deviation trend in subsequent batch inspections.

Images