Telecentric shadow imaging and adaptive fine-tuned cable dimension measurement apparatus and method

Abstract

The application belongs to the field of precise optical measurement and machine vision, and specifically discloses a telecentric shadow imaging and self-adaptive fine adjustment cable size measuring device, which comprises: a mounting seat unit comprising a base and a rotating table rotatably arranged on the base, the rotating table being used for placing a cable to be measured; an optical imaging unit comprising a double-telecentric lens, a backlight source and a CMOS sensor, which is used for collecting image information of the cable to be measured; and an image processing unit electrically connected with the mounting seat unit and the optical imaging unit, the image processing unit being configured to calculate current posture information of the cable to be measured according to the image information, and to control the rotating table to rotate to the optical imaging unit to collect image information of a standard normal section according to the current posture information, and further configured to calculate structural size parameters of the cable to be measured according to the image information of the standard normal section. The application can solve the problems of poor precision, low efficiency and poor adaptability existing in the prior cable size measuring method.

Description

Technical Field

[0001] This application belongs to the field of precision optical measurement and machine vision, and more specifically, relates to a cable size measurement device and method with telecentric shadow imaging and adaptive fine-tuning. Background Technology

[0002] In the quality inspection of wires and cables, outer diameter, cross-sectional dimensions, and appearance defects are core control indicators. However, existing measurement technologies still have the following technical problems that urgently need to be solved in practical applications.

[0003] 1. Conventional image measuring instruments often employ ordinary backlights or non-telecentric optical lenses, which suffer from ambient light interference and thermal drift. On one hand, changes in ambient light in the workshop can cause shifts in the grayscale gradient at the edge of the cable under test; on the other hand, the thermal expansion and contraction of the lens and body materials can easily lead to instability in the system calibration parameters. These factors combined result in poor measurement repeatability, with errors typically reaching ±50μm, making it difficult to meet the requirements of precision testing.

[0004] 2. Cable workpieces, especially stranded conductors or stranded wires, have inherent helical textures on their surfaces. The stranded structure can easily cause measurement artifacts. In ordinary image measurement mode, the peak-valley structure of the helix is ​​easily misjudged as the true contour edge, leading to significant fluctuations in the outer diameter measurement value. To obtain the standard cross-section of the cable, operators need to repeatedly rotate the sample manually to find the optimal observation posture. This process heavily relies on manual experience, resulting in low inspection efficiency and difficulty in ensuring consistency.

[0005] 3. Traditional edge extraction algorithms often rely on integer pixel grayscale jumps for edge detection, which is easily affected by surface defects. In actual working conditions, tiny burrs or oil stains on the cable surface can cause local grayscale abrupt changes. Such interference is easily misidentified by the algorithm as valid edges, leading to inaccurate measurements and reducing the accuracy and reliability of the measurements.

[0006] 4. Before inspection, general-purpose image measuring instruments usually need to perform complex pre-processing operations such as dimming, focusing, and stage space calibration. The cumbersome operation process is in stark contrast to the "one-click" rapid and batch inspection requirements of the production line, making it difficult to directly embed into the efficient production and inspection chain. Therefore, there is a problem of insufficient production line adaptability of general-purpose equipment.

[0007] Based on the shortcomings of the existing technology, this application makes targeted innovations from three dimensions: optical system design, mechanical structure improvement and image processing algorithm optimization, aiming to provide a cable-specific size measurement device and method that combines high-precision measurement, high anti-interference stability and high-efficiency detection. Summary of the Invention

[0008] In response to the deficiencies or improvement needs of existing technologies, this application provides a cable size measurement device and method with telecentric shadow imaging and adaptive fine-tuning, aiming to solve the problems of poor accuracy, low efficiency and poor adaptability of existing cable size measurement methods.

[0009] The above-mentioned technical objectives of this application are mainly achieved through the following technical solutions.

[0010] On one hand, this application provides a cable size measurement device with telecentric shadow imaging and adaptive fine-tuning, comprising: The mounting unit includes a base and a rotating platform rotatably mounted on the base, the rotating platform being used to place the cable to be tested; An optical imaging unit, comprising a dual telecentric lens, a backlight, and a CMOS sensor, wherein the CMOS sensor is used to acquire image information of the cable under test through the dual telecentric lens when the backlight illuminates the cable under test. An image processing unit is electrically connected to the mounting base unit and the optical imaging unit. The image processing unit is configured to calculate the current attitude information of the cable under test based on the image information transmitted by the optical imaging unit, and to control the rotary table to rotate to the image information of the standard cross-section of the cable under test acquired by the optical imaging unit based on the current attitude information. The image processing unit is also configured to calculate the structural dimension parameters of the cable under test based on the image information of the standard cross-section.

[0011] In a preferred embodiment of this application, the backlight source uses green LED parallel light.

[0012] In a preferred embodiment of this application, the mounting base unit further includes a lifting seat disposed on the base, and the rotary table is mounted and fixed on the lifting seat.

[0013] In a preferred embodiment of this application, the rotating platform is provided with a V-shaped or semi-circular bracket for placing the cable under test.

[0014] In a preferred embodiment of this application, the cable size measuring device further includes a display unit, which is connected to the optical imaging unit and the image processing unit. The display unit is used to display the image information acquired by the optical imaging unit and the structural size parameters of the cable under test calculated by the image processing unit.

[0015] On the other hand, this application also provides a cable size measurement method based on telecentric shadow imaging and adaptive fine-tuning, used to measure the cross-sectional dimensions of a cylindrical cable. The cable size measurement method includes: The cable under test is placed on a rotating stage, and a cross-sectional image of the cable under test is acquired using an optical imaging unit. The contour of the cross-sectional image is extracted using the Canny operator. A Hough line transform is then performed on the contour to extract the slope k of the dominant feature line in the contour, and the initial helix angle is calculated. ; Constructing a roundness evaluation function ,in, and Let θ be the projected lengths of the cable's profile in the X and Y directions of the image at the current angle θ, and λ be the weighting coefficient. The mean square error is obtained by fitting the profile of the cable under test to a standard ellipse at the current angle θ. use The characteristic of obtaining the minimum value when the cable under test is in the standard normal cross-section orientation, with the initial angle Using the midpoint of the initial interval as the reference point, the rotary table is driven to approximate the point containing the minimum value within the initial interval using the golden section method. The convergence interval is defined, and the midpoint of the convergence interval is taken as... ; Drive the rotary table to rotate from its initial position by an angle Then, the optical imaging unit is used to acquire a cross-sectional image of the cable under test, and the structural dimension parameters of the cable under test are obtained based on the processing results of the cross-sectional image.

[0016] In a preferred embodiment of this application, the golden section method is used to approximate the value to the minimum point. The convergence criterion for the convergence interval is: or ; in, This represents the boundary angle of the previous interval in the golden section process. This represents the boundary angle of the next interval in the golden section process.

[0017] In a preferred embodiment of this application, the optical imaging unit uses green LED parallel light as a backlight and acquires a cross-sectional image of the cable under test through a dual telecentric lens.

[0018] In a preferred embodiment of this application, obtaining the structural dimensional parameters of the cable under test based on the processing results of the cross-sectional image includes: Traverse the grayscale sequence of the cross-sectional image, solve for the first derivative of the grayscale, and take the pixel position x0 corresponding to the maximum value of the first derivative as the coarse positioning point of the edge integer pixel; Centered on the coarse positioning point x0, select the grayscale values ​​g of three adjacent pixels. 1. g0 and g +1 Calculate subpixel offset ; The final coordinates of the cross-sectional image edge are calculated based on the coarse positioning point x0 and the sub-pixel offset δ. .

[0019] In a preferred embodiment of this application, obtaining the structural dimensional parameters of the cable under test based on the processing results of the cross-sectional image further includes: Calculate the curvature K of the edge points of the cross-sectional image i and curvature threshold K th =1.5 / R, where R is the theoretical radius of the cable under test. If K i >K th If the neighborhood grayscale variance of the edge point is less than twice the background variance of the cross-sectional image, then the edge point is determined to be a spur. A cubic Bézier curve is then used to fit the surrounding normal edge points, and the fitted values ​​are used to replace the coordinates of the edge point confirmed as a spur. The grayscale difference ΔG between the two sides of the edge point of the cross-sectional image is detected on the 8-bit grayscale image. If ΔG < 30, the edge point is determined to be an oil stain defect. The edge point is marked as a defect and is not included in the calculation of structural dimension parameters.

[0020] In summary, compared with the prior art, the technical solutions conceived in this application have the following main technical advantages: 1. The cable size measurement device and method with telecentric shadow imaging and adaptive fine adjustment described in this application uses a rotary table with adaptive helix angle fine adjustment function to place the cable to be measured, and controls the rotation of the rotary table based on the recognition and analysis results of the image recognition algorithm to find the "minimum circumcircle" posture of the cable cross section, automatically compensates for the helix tilt angle of the cable, eliminates the need for manual alignment, and reduces the diameter measurement error caused by the threads on the cable.

[0021] 2. The cable size measurement device and method with telecentric shadow imaging and adaptive fine adjustment described in this application uses green LED parallel light in conjunction with dual telecentric lenses to acquire cross-sectional image information of the cable under test. It only receives light rays parallel to the optical axis, which can filter out ambient stray light and reduce the influence of ambient light. Even under normal lighting, it can obtain high-contrast contours and improve measurement accuracy. At the same time, the optical path structure in this application is symmetrical, and thermal expansion has minimal impact on the magnification, enabling "automatic calibration upon power-on without manual calibration".

[0022] 3. The cable size measurement device and method with telecentric shadow imaging and adaptive fine-tuning described in this application uses a subpixel edge subdivision and denoising algorithm based on ROI to identify image edges and filter burrs and oil stains for the image information of the standard cross-section of the cable under test, while retaining continuous contours. Automatic correction of edge contours can be achieved without manual intervention, thereby improving the accuracy of the measurement of the cable size parameters. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the internal architecture of the cable size measurement device with telecentric shadow imaging and adaptive fine-tuning described in this application; Figure 2 This is a flowchart of the cable size measurement method with telecentric shadow imaging and adaptive fine-tuning described in this application. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0025] Implementation Method 1: This application provides a cable size measurement device with telecentric shadow imaging and adaptive fine-tuning, such as... Figure 1 As shown, the cable size measuring device includes a mounting base unit, an optical imaging unit, and an image processing unit. The mounting base unit includes a base and a rotating stage rotatably mounted on the base, the rotating stage being used to place the cable under test. The optical imaging unit includes dual telecentric lenses, a backlight, and a CMOS sensor. The CMOS sensor is used to acquire image information of the cable under test through the dual telecentric lenses when the backlight illuminates the cable under test. The image processing unit is electrically connected to the mounting base unit and the optical imaging unit. The image processing unit is configured to calculate the current attitude information of the cable under test based on the image information transmitted by the optical imaging unit, and to control the rotating stage to rotate to the image information of the standard cross-section of the cable under test acquired by the optical imaging unit based on the current attitude information. The image processing unit is also configured to calculate the structural dimensional parameters of the cable under test based on the image information of the standard cross-section.

[0026] The cable size measuring device described in this application uses a rotary table with adaptive helix angle fine adjustment function to place the cable to be measured, and controls the rotation of the rotary table based on the image recognition analysis results to find the "minimum circumcircle" posture of the cable cross-section, automatically compensating for the helix tilt angle of the cable, eliminating the need for manual alignment, and reducing the diameter measurement error caused by the threads on the cable.

[0027] The following section will provide a detailed description of the specific structure of each part of the cable size measuring device described in this application, as well as the position and connection relationship between each part.

[0028] The cable size measuring device includes a mounting base unit, such as... Figure 1 As shown, the mounting unit is used to place the cable under test and can adaptively adjust the position and orientation of the cable under test.

[0029] Specifically, the mounting unit includes a base and a rotating platform. The base serves as the mounting foundation for the entire device. The rotating platform is mounted on the base and can rotate around its own axis under the drive of a motor. The rotation of the rotating platform is controlled according to the current posture information of the cable under test on the rotating platform, thereby enabling the optical imaging unit to acquire image information of the standard cross-section of the cable under test. In this embodiment, the rotating platform is controlled by an electronic control module. The control commands of the electronic control module come from the optical imaging unit. The rotating platform can rotate within a range of -10° to +10° with the initial zero point position as the origin, and the rotation control accuracy of the rotating platform is less than or equal to 0.1°.

[0030] Better, such as Figure 1 As shown, the mounting unit also includes a lifting seat mounted on the base. The rotary table is fixedly mounted on the lifting seat. Driven by a motor, the lifting seat can move the rotary table up and down, thereby raising and lowering the cable under test, allowing the cable to accurately enter the center of the optical imaging unit. In this embodiment, the lifting seat is controlled by an electronic control module. The control commands from the electronic control module come from the operator. The lifting seat can move up and down within a certain range with the initial zero point as the origin, and the lifting control accuracy of the lifting seat is less than or equal to 0.1 mm.

[0031] Better, such as Figure 1 As shown, the rotating platform is equipped with a V-shaped or semi-circular bracket for placing the cable under test. The cable under test is clamped and fixed by the bracket, thereby ensuring that the position of the cable under test relative to the rotating platform is relatively fixed.

[0032] The cable size measuring device also includes an optical imaging unit, such as... Figure 1 As shown, the optical imaging unit is used to acquire cross-sectional image information of the cable under test and transmit the image information to the image processing unit.

[0033] Specifically, the optical imaging unit includes a dual telecentric lens, a backlight, and a CMOS sensor. The object-side and image-side optical paths of the dual telecentric lens are parallel to the optical axis, possessing characteristics such as constant magnification, no perspective parallax, ultra-low distortion, and large depth of field. The dual telecentric lens can eliminate imaging errors caused by the position and installation deviation of the cable under test, ensuring accurate and distortion-free imaging of contours, angles, and dimensions. In this embodiment, the telecentricity of the dual telecentric lens is less than 0.1°. The backlight can use parallel light from a green LED with high transmittance. In this embodiment, the backlight uses parallel light with a wavelength of 520nm. The image information obtained by the green light source combined with the dual telecentric lens has a smooth grayscale curve, which can fully utilize the accuracy of sub-pixel positioning, contour detection, and angle optimization algorithms in the image recognition and analysis process. The CMOS sensor, as a photosensitive element, can convert light signals into electrical signals carrying image information. In this embodiment, a CMOS sensor with more than 2 megapixels and low distortion is used.

[0034] The cable size measuring device also includes an image processing unit, such as... Figure 1 As shown, the image processing unit is used to analyze the image information acquired by the optical imaging unit and control the rotation of the rotary table to adjust the posture of the cable under test based on the analysis results. At the same time, it is used to analyze the image information of the standard cross section to calculate the structural dimension parameters of the cable under test.

[0035] Specifically, after the cable under test is placed on the rotary table, its posture may deviate to some extent, resulting in the image captured by the optical imaging unit not being the standard cross-section of the cable under test. Therefore, the position of the cable under test needs to be fine-tuned. First, the optical imaging unit captures an image of the cable under test in its initial position, and this image is analyzed to confirm the initial deviation of the cable under test. Then, the rotary table rotates the cable under test within a certain range, and the adjusted image is acquired in real time. The adjusted image is then analyzed again to further determine the deviation of the cable under test. The above process is repeated until the optical imaging unit can capture the standard cross-section of the cable under test. In this embodiment, to improve the efficiency of the above adjustment process, a specific strategy can be adopted to find the optimal position. For example, a roundness evaluation function can be constructed based on image information, and the standard cross-section of the cable under test can be quickly found using the golden section method. After the standard cross-section of the cable under test is found by the rotating stage, the standard cross-section image of the cable under test is captured by the optical imaging unit. Then, the image processing unit processes the image to obtain the structural dimension parameters of the cable under test, such as the diameter. In order to improve the image analysis results, edge subdivision algorithm and denoising algorithm can be used to process the image to improve the accuracy of the structural dimension parameters of the cable under test.

[0036] According to one embodiment of this application, such as Figure 1As shown, the cable size measuring device also includes a display unit, which is connected to the optical imaging unit and the image processing unit. The display unit is used to display the image information acquired by the optical imaging unit and the structural size parameters of the cable under test calculated by the image processing unit.

[0037] Implementation Method Two: This application also provides a cable size measurement method based on telecentric shadow imaging and adaptive fine-tuning, used to measure the cross-sectional dimensions of cylindrical cables, such as... Figure 2 As shown, the cable size measurement method includes the following steps: Step S10: Place the cable under test on a rotating stage and use an optical imaging unit to acquire a cross-sectional image of the cable under test.

[0038] Step S20: Use the Canny operator to extract the contour of the cross-sectional image, perform a Hough line transform on the contour to extract the slope k of the dominant feature line in the contour, and calculate the initial helix angle. .

[0039] Step S30: Construct the roundness evaluation function ,in, and Let θ be the projected lengths of the cable's profile in the X and Y directions of the image at the current angle θ, and λ be the weighting coefficient. The mean square error is the result of fitting the profile of the cable under test to a standard ellipse at the current angle θ.

[0040] Step S40: Utilize The characteristic of obtaining the minimum value when the cable under test is in the standard normal cross-section orientation, with the initial angle Using the midpoint of the initial interval as the reference point, the rotary table is driven to approximate the point containing the minimum value within the initial interval using the golden section method. The interval of convergence is defined, and the midpoint of the interval of convergence is taken as the boundary. .

[0041] Step S50: Drive the rotary table to rotate from the initial position by an angle. Then, the cross-sectional image of the cable under test is acquired using an optical imaging unit, and the structural dimensional parameters of the cable under test are obtained based on the processing results of the cross-sectional image.

[0042] The cable size measurement method described in this application constructs a roundness evaluation function and uses the golden section method to control the rotation of the rotary table based on the image recognition analysis results in order to find the "minimum circumcircle" posture of the cable cross-section, automatically compensate for the helical tilt angle of the cable, eliminates the need for manual alignment, and reduces the diameter measurement error caused by the cable threads.

[0043] The following section will provide a detailed description of the specific steps involved in the cable size measurement method described in this application.

[0044] In step S10, the cable to be tested is placed and an initial interface image is obtained.

[0045] Specifically, the cable under test is placed on and fixed on a rotating platform that can rotate around its own axis. The rotating platform can be the mounting base unit described in Embodiment 1. Then, an initial cross-sectional image of the cable under test is acquired using an imaging device, which can be the optical imaging unit described in Embodiment 1.

[0046] Preferably, the optical imaging unit uses green LED parallel light as a backlight and acquires the cross-sectional image of the cable under test through a dual telecentric lens; the image information obtained by the green light source combined with the dual telecentric lens has a smooth grayscale curve, which can give full play to the accuracy of sub-pixel positioning, contour detection and angle optimization algorithms in the image recognition and analysis process.

[0047] In step S20, the contour of the cross-sectional image is extracted and the initial helix angle is calculated.

[0048] The image contour extraction and coarse helix angle estimation were completed to provide an initial starting point for subsequent iterative optimization. The specific process is as follows.

[0049] First, the Canny edge detection algorithm is used to extract the effective contours from the cross-sectional image of the cable under test. On an 8-bit grayscale image (grayscale value range 0-255), two thresholds are set: a low threshold T... low =50, High threshold T high =150; High threshold filters strong edges (the true outline of the cable under test), low threshold connects weak edges and completes broken outlines; the combination of dual thresholds can suppress image noise and false edges, while ensuring the integrity and continuity of the spiral feature outline.

[0050] Next, a Hough line transform is performed on the extracted contour to identify the dominant feature line in the contour, and the slope k of the line is calculated. The spatial tilt angle of the spiral structure is directly reflected in the slope of the image line, serving as a bridge between the geometric angle and the pixel image. The initial spiral angle is calculated from the line slope using the arctangent function. , The helix angle is the coarse measurement value calculated by the image processing unit under the initial posture of the cable under test, and it is also the initial angle for subsequent iterative searches.

[0051] In step S30, a roundness evaluation function is constructed.

[0052] Specifically, for the objective of "rotating the cable under test to a standard cross-section," a quantitative evaluation function F(θ) is designed to convert the standard orientation into a calculable numerical index. In this embodiment, the roundness evaluation function F(θ) is: (1); In the above formula (1), F(θ) is the roundness evaluation function; and These are the projected lengths of the cable profile under test in the two orthogonal directions X and Y of the image at the current angle θ, and their values ​​are related to the image recognition results at the current angle θ. The mean square error is obtained by fitting the extracted profile of the cable under test into a standard ellipse, which characterizes the degree of deviation between the profile and the ideal ellipse; λ is the weighting coefficient, and in this embodiment, λ is taken as 0.5.

[0053] in, The difference in the projection length of the profile is such that when the cable under test has a standard cross-section (circular cross-section), the projection lengths in the two orthogonal directions are approximately equal, and the difference approaches 0. The greater the attitude deflection and helix angle offset, the more obvious the difference in the projection lengths in the X and Y directions, and the larger this value is. The term is used to evaluate the "circular symmetry" of the cross-sectional image. Under normal cross-sectional conditions, the profile of the cable under test approximates a standard perfect circle (a perfect circle is a special ellipse), and the fitting error is... The minimum value is achieved when the cable is tilted; however, when the cable is tilted, the profile is stretched into an irregular ellipse, significantly increasing the error. The λ term is used to balance the numerical magnitudes of the two indicators, "projection length difference" and "ellipse fitting error," to avoid a single indicator dominating the evaluation result. F(θ) is a comprehensive evaluation index that integrates profile symmetry and profile regularity. The smaller the value, the closer the cable's posture is to the ideal cross-section. When the cable is in a standard cross-section posture, F(θ) reaches its global minimum value.

[0054] The aforementioned roundness evaluation function relies on the advantages of a dual telecentric lens. This lens has no perspective distortion, and the magnification of near and far objects is consistent. The image size is only related to the actual geometric size of the cable under test and will not be deformed due to position or angle, thus ensuring the effectiveness of the roundness evaluation function.

[0055] In step S40, the golden section method is used to iteratively search for the optimal angle.

[0056] With initial angle Starting from the given angle range, the golden section method is used to quickly find the minimum point that minimizes the evaluation function F(θ). .

[0057] Specifically, set search interval constraints. Considering the actual working conditions, the initial placement deviation and initial angle error of the cable under test will not exceed ±10°, thus limiting the initial search range to [ ]. -10°, +10°], the total length of the interval is 20°; in the process of confirming the initial interval, reducing the interval length as much as possible can greatly reduce the amount of calculation, improve real-time performance, and avoid invalid global search.

[0058] Based on the golden ratio, two initial sampling angles are set in the above initial search range x1= -2.36°, x2= +2.36° (wherein, 0.618×20°=2.36°), that is, the rotating table is successively rotated 2.36° to the left and right respectively from the initial position, the optical imaging unit is used to obtain the cross-sectional image information of the cable to be tested at the two positions above, and at the same time the method in the above step S20 is used to extract the image contour, and the corresponding , and , which are substituted into the above formula (1) to calculate F(x1) and F(x2) respectively. Compare F(x1) and F(x2), if F(x1)<F(x2), the minimum value is in the left half area, and the new interval is -10°, +2.36°], if F(x1)≥F(x2), the minimum value is in the right half area, and the new interval is -2.36°, +10°].

[0059] Based on the confirmed new interval, the golden section method is used again to generate new interior points, calculate the corresponding F(θ), compare the angles of two adjacent iterations and the function values of two adjacent iterations, so as to complete further interval contraction until the interval convergence condition is reached, and obtain the corresponding convergence interval. The convergence interval contains the minimum point of F(θ) , and the midpoint of the convergence interval is taken as .

[0060] Preferably, in the process of using the golden section method for interval iteration to approximate the minimum point , the interval convergence condition is: or (2); In the above formula (2), is the boundary angle of the previous interval in the golden section process, is the boundary angle of the next interval in the golden section process.

[0061] During the interval iteration process, if the angle difference between two consecutive iterations is less than 0.1°, it indicates that the angle has essentially stopped changing and the angle accuracy requirement has been met; or, if the difference in the roundness evaluation function F(θ) value between two consecutive iterations is less than 0.001 pixels, it represents that the contour shape and roundness index have stabilized and the visual features have not changed significantly. When any of the above conditions are met during the iteration process, the angle relative to the boundary is used as the reference. The next corresponding interval is taken as the convergence interval.

[0062] In step S50, attitude compensation and precision measurement are performed.

[0063] The optimal angle is obtained through iteration in step S40. Then, calculate the compensation angle. The cable under test is rotated from its initial position to the optimal angle using a rotary table; whereby... This represents the angle the rotary table needs to rotate; a positive number indicates forward rotation, and a negative number indicates reverse rotation.

[0064] Furthermore, after the rotary table drives the cable under test to rotate into position, the optical imaging unit acquires a cross-sectional image (a standard positive cross-section image) of the cable under test, and obtains the structural dimensional parameters of the cable under test based on the processing results of the cross-sectional image.

[0065] The process of processing the cross-sectional image to obtain the structural dimension parameters of the cable under test is mainly as follows: For the high signal-to-noise ratio edge (SNR>45dB) formed by the green LED backlight, a three-point Gaussian interpolation algorithm is used to achieve 0.01 pixel-level positioning. The specific process is as follows.

[0066] First, the grayscale sequence of the cross-sectional image is traversed, the first derivative of the grayscale is calculated, and the pixel position x0 corresponding to the maximum value of the first derivative is used as the coarse positioning point of the integer pixel of the edge. The contour edge is located at the ideal step edge, where the grayscale change rate is the fastest and the first derivative takes the maximum value. This is used to lock the integer pixel position of the edge and provide a benchmark for subsequent sub-pixel correction.

[0067] Secondly, using the coarse positioning point x0 as the center, select the grayscale values ​​g of three adjacent pixels. 1. g0 and g +1 Calculate subpixel offset : (3); The above formula (3) is derived from fitting the gray distribution of a single pixel neighborhood using a Gaussian function. The edge gray value approximately follows a Gaussian distribution. Only when the gray value curve is continuous, smooth, and without drastic fluctuations will the denominator be stable and not zero. This formula relies on the smooth gray value curve provided by the narrow green light. Ordinary white light dispersion will cause the denominator to oscillate and fail.

[0068] Finally, the final coordinates of the cross-sectional image edge are calculated based on the coarse positioning point x0 and the sub-pixel offset δ. This enables positioning at the 0.01 pixel level, improving the contour positioning accuracy during the measurement of the cable dimensions under test.

[0069] According to one embodiment of this application, in step S50, after obtaining the high-precision edge of the cross-sectional image using three-point Gaussian interpolation, the edge points are denoised to remove abnormal points. The specific process is as follows.

[0070] First, a curvature constraint is used for filtering to remove contour burrs. The curvature K of the edge points in the cross-sectional image is calculated. i and curvature threshold K th =1.5 / R, where R is the theoretical radius of the cable under test. If K i >K th If the neighborhood grayscale variance of the edge point is less than twice the variance of the image background, then the edge point is determined to be a burr. A cubic Bézier curve is used to fit the surrounding normal edge points, and the fitted value is used to replace the coordinates of the edge points confirmed as burrs to ensure that the edge contour is continuous and smooth, and to avoid the contour breakage caused by directly deleting points.

[0071] Where, when K i >K th When the local contour is curved far beyond the standard arc, the shape is abrupt; when the gray-level variance of the neighborhood of the edge point is less than twice the gray-level variance of the background, the local gray level is uniform and does not belong to the real bright-dark step edge, which can further prove that it is noise or burrs.

[0072] Secondly, a secondary filtering process is performed using grayscale consistency constraints to remove oil stains or surface defects. The grayscale difference ΔG between the two sides of the edge point of the cross-sectional image is detected on an 8-bit grayscale image (grayscale value range 0-255). If ΔG < 30, the edge point is determined to be an oil stain defect, and the edge point is marked as a defect and is not included in the calculation of structural dimension parameters.

[0073] The above two-stage filtering process can filter out burrs and oil stains in the profile, retain the continuous profile, and achieve automatic correction of the edge profile without manual intervention, thereby further improving the measurement accuracy of the dimensional parameters of the cable under test.

[0074] It should be understood that expressions such as "comprising" and "may include" as used in this application indicate the existence of the disclosed functions, operations, or constituent elements, and do not limit one or more additional functions, operations, and constituent elements. In this application, terms such as "comprising" and / or "having" may be interpreted as indicating a specific characteristic, number, operation, constituent element, component, or combination thereof, but should not be interpreted as excluding the existence or possibility of adding one or more other characteristics, numbers, operations, constituent elements, components, or combinations thereof.

[0075] It should be understood that the terms “center,” “upper,” “lower,” “front,” “rear,” “left,” “right,” “vertical,” “horizontal,” “inner,” “outer,” “clockwise,” “counterclockwise,” “axial,” “radial,” and “circumferential” indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0076] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0077] In this application, unless otherwise expressly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0078] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A cable size measurement device with telecentric shadow imaging and adaptive fine-tuning, characterized in that, The cable size measuring device includes: The mounting unit includes a base and a rotating platform rotatably mounted on the base, the rotating platform being used to place the cable to be tested; An optical imaging unit, comprising a dual telecentric lens, a backlight, and a CMOS sensor, wherein the CMOS sensor is used to acquire image information of the cable under test through the dual telecentric lens when the backlight illuminates the cable under test. An image processing unit is electrically connected to the mounting base unit and the optical imaging unit. The image processing unit is configured to calculate the current attitude information of the cable under test based on the image information transmitted by the optical imaging unit, and to control the rotary table to rotate to the image information of the standard cross-section of the cable under test acquired by the optical imaging unit based on the current attitude information. The image processing unit is also configured to calculate the structural dimension parameters of the cable under test based on the image information of the standard cross-section.

2. The cable size measuring device with telecentric shadow imaging and adaptive fine-tuning according to claim 1, characterized in that, The backlight uses green LED parallel light.

3. The cable size measuring device with telecentric shadow imaging and adaptive fine-tuning according to claim 1, characterized in that, The mounting unit also includes a lifting seat disposed on the base, and the rotary table is mounted and fixed on the lifting seat.

4. The cable size measuring device with telecentric shadow imaging and adaptive fine-tuning according to claim 1, characterized in that, The rotating platform is equipped with a V-shaped or semi-circular bracket for placing the cable under test.

5. The cable size measuring device with telecentric shadow imaging and adaptive fine-tuning according to claim 1, characterized in that, It also includes a display unit, which is connected to the optical imaging unit and the image processing unit. The display unit is used to display the image information acquired by the optical imaging unit and the structural dimension parameters of the cable under test calculated by the image processing unit.

6. A method for measuring cable dimensions using telecentric shadow imaging and adaptive fine-tuning, for measuring the cross-sectional dimensions of cylindrical cables, characterized in that, The cable size measurement method includes: The cable under test is placed on a rotating stage, and a cross-sectional image of the cable under test is acquired using an optical imaging unit. The contour of the cross-sectional image is extracted using the Canny operator. A Hough line transform is then performed on the contour to extract the slope k of the dominant feature line in the contour, and the initial helix angle is calculated. ; Constructing a roundness evaluation function ,in, and Let θ be the projected lengths of the cable's profile in the X and Y directions of the image at the current angle θ, and λ be the weighting coefficient. The mean square error is obtained by fitting the profile of the cable under test to a standard ellipse at the current angle θ. use The characteristic of obtaining the minimum value when the cable under test is in the standard normal cross-section orientation, with the initial angle Using the midpoint of the initial interval as the reference point, the rotary table is driven to approximate the point containing the minimum value within the initial interval using the golden section method. The convergence interval is defined, and the midpoint of the convergence interval is taken as... ; Drive the rotary table to rotate from its initial position by an angle Then, the optical imaging unit is used to acquire a cross-sectional image of the cable under test, and the structural dimension parameters of the cable under test are obtained based on the processing results of the cross-sectional image.

7. The cable size measurement method with telecentric shadow imaging and adaptive fine-tuning according to claim 6, characterized in that, Approximating to the minimum point using the golden section method The convergence criterion for the convergence interval is: or ; in, This represents the boundary angle of the previous interval in the golden section process. This represents the boundary angle of the next interval in the golden section process.

8. The cable size measurement method with telecentric shadow imaging and adaptive fine-tuning according to claim 6, characterized in that, The optical imaging unit uses green LED parallel light as a backlight and acquires cross-sectional images of the cable under test through a dual telecentric lens.

9. The cable size measurement method with telecentric shadow imaging and adaptive fine-tuning according to claim 8, characterized in that, The structural dimensional parameters of the cable under test obtained based on the processing results of the cross-sectional image include: Traverse the grayscale sequence of the cross-sectional image, solve for the first derivative of the grayscale, and take the pixel position x0 corresponding to the maximum value of the first derivative as the coarse positioning point of the edge integer pixel; Centered on the coarse positioning point x0, select the grayscale values ​​g of three adjacent pixels.

1. g0 and g +1 Calculate subpixel offset ; The final coordinates of the cross-sectional image edge are calculated based on the coarse positioning point x0 and the sub-pixel offset δ. .

10. The cable size measurement method with telecentric shadow imaging and adaptive fine-tuning according to claim 9, characterized in that, The structural dimensional parameters of the cable under test obtained based on the processing results of the cross-sectional image also include: Calculate the curvature K of the edge points of the cross-sectional image i and curvature threshold K th =1.5 / R, where R is the theoretical radius of the cable under test. If K i >K th Furthermore, if the neighborhood grayscale variance of the edge point is less than twice the background variance of the cross-sectional image, then the edge point is determined to be a spur. A cubic Bézier curve is then used to fit the surrounding normal edge points, and the fitted values ​​are used to replace the coordinates of the edge point confirmed as a spur. The grayscale difference ΔG between the two sides of the edge point of the cross-sectional image is detected on the 8-bit grayscale image. If ΔG < 30, the edge point is determined to be an oil stain defect. The edge point is marked as a defect and is not included in the calculation of structural dimension parameters.

Images