Device and method for detecting pitch, diameter and forming height of preformed armor rod of connection fitting
Through the image processing method of industrial cameras and lightweight convolutional neural networks combined with the Canny edge detection algorithm, the problems of low efficiency and accuracy in pre-twisted wire detection are solved, and high-precision automated detection and classification of pre-twisted wires are achieved, which is suitable for detection needs under various working conditions.
Patent Information
- Application Number
- CN202510957997.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-11
- Publication Date
- 2025-09-16
AI Technical Summary
In the existing technology, the detection efficiency and accuracy of the pitch, diameter and forming height of pre-twisted wires are low, which makes it difficult to meet the quality control requirements of mass production. Manual measurement results are easily affected by the operator, and the two-dimensional image measurement method has projection errors and perspective deformation problems in three-dimensional spiral structures.
An image processing method combining an industrial camera with a lightweight convolutional neural network and the Canny edge detection algorithm is used. A bowl-shaped light source is used for uniform illumination. Automated detection of pre-twisted wires is achieved through image preprocessing and parameter extraction. Real-time image acquisition and parameter calculation are achieved in conjunction with a PLC system.
The accuracy and efficiency of pre-twisted wire detection are improved, the interference of lighting and background complexity on the measurement results is reduced, and high-precision automatic detection and classification of pre-twisted wires are achieved, which is suitable for detection needs under various working conditions.
Smart Images

Figure CN120651124A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of pre-twisted wire production, and in particular to a device and method for detecting the pitch, diameter and forming height of pre-twisted wires of connecting hardware. Background Art
[0002] Pre-twisted wire is a spiral protective strip used to protect bare conductors from stress concentrations, electrical corrosion, and arc burns. It is a connector used in the termination, suspension, and joints of overhead power conductors and overhead optical cables. Pre-twisted wire plays a vital role in ensuring the safe and stable operation of power systems, extending line life, and improving transmission efficiency. It is an indispensable key component in the power and communications sectors.
[0003] The consistency of parameters such as diameter, forming height, and pitch is a key indicator of product quality. During the production process of pre-twisted wire, the pitch, diameter, and forming height of the product may deviate due to the influence of various factors such as raw materials, forming process, equipment precision, and heat treatment. If these key dimensional parameters exceed the allowable tolerance range, it will seriously affect the performance of the pre-twisted wire, such as causing installation difficulties, insufficient gripping force, slippage, cable damage, and even serious accidents such as line breakage and disconnection. Therefore, in order to ensure the quality of pre-twisted wire products and the safety of line operation, accurate, efficient, and reliable testing of the pitch, diameter, and forming height of the pre-twisted wire during the production process and factory inspection is an essential quality control link.
[0004] Existing technical solution: Operators use manual measuring tools such as calipers, micrometers, pitch gauges, angle rulers, special gauges or templates with scales to measure the pre-twisted wire samples point by point or section by section. For example, use a caliper to measure the outer diameter of a certain section of the pre-twisted wire; use a pitch gauge to compare and measure the pitch; use a height gauge in combination with a platform to estimate the forming height. This method is inefficient and not suitable for full inspection or high-frequency spot checks in mass production. The disadvantage of the above-mentioned operators using related tools for measurement is that the measurement results are easily affected by the operator's proficiency, eyesight, measurement techniques, reading errors and the accuracy of the measuring tools themselves, and are highly subjective, with poor consistency and repeatability. For the pitch of the spiral structure, it is difficult to accurately measure the axial distance manually; the definition and measurement benchmark of the forming height are difficult to unify, and the manual measurement error is large.
[0005] With the advancement of machine vision technology, inspection is now performed using industrial cameras and image processing software. A camera captures images of pre-twisted wire from the top or side, and then uses image processing algorithms to calculate pitch, diameter, and other parameters. However, two-dimensional images only capture the projection of an object onto a specific plane. For complex three-dimensional spiral structures like pre-twisted wire, accurately measuring the spatial pitch and true forming height using only two-dimensional images is difficult. The aforementioned measurement method using industrial cameras combined with image processing has the disadvantage that projection errors and perspective distortion can severely affect measurement accuracy. This method also places high demands on lighting conditions, background quality, and camera calibration accuracy, otherwise image quality and measurement results will be affected. Summary of the Invention
[0006] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a device and method for detecting the pitch, diameter and forming height of pre-twisted wires of connecting hardware. The device and method can realize the online detection of pre-twisted wire diameter, forming height and pitch, improve the detection efficiency and accuracy, and thus improve product quality.
[0007] In order to achieve the above object, the present invention adopts the following technical solutions:
[0008] A device for detecting the pitch, diameter, and forming height of pre-twisted wires of connecting fittings, comprising a workbench, on which are respectively provided a feeding mechanism, a distance measuring sensor, a visual inspection module equipped with an industrial camera, a shearing mechanism, a sorting mechanism, and a display screen;
[0009] The front end of the shearing mechanism is respectively provided with a feeding mechanism, a visual detection module and a display screen. The feeding mechanism is provided on one side of the visual detection module. A distance measuring sensor is installed on the feeding mechanism. The display screen is provided on the other side of the visual detection module. The rear end of the shearing mechanism is provided with a sorting mechanism.
[0010] The feeding mechanism includes a mounting bracket, a support ring, a pull rod, an adjustment slider with adjustable inner diameter, a tightening knob, a compression spring, a rubber wheel, a rubber wheel frame, and a spring mounting plate;
[0011] A support ring is provided on the inner side of the mounting bracket, and the support ring is connected to the mounting bracket through a pull rod; four groups of anti-shake components are provided on the inner side of the support ring, and the four groups of anti-shake components are arranged perpendicular to each other in pairs, and a central hole for the pre-twisted wire to pass through is formed in the middle of the four groups of anti-shake components; each group of anti-shake components includes a rubber wheel frame, and a rubber wheel is installed at one end of the rubber wheel frame close to the middle position of the support ring, and an adjustment slider is provided at the other end of the rubber wheel frame away from the middle position of the support ring, and a tightening knob is provided on the adjustment slider, and a compression spring is provided between the rubber wheel frame and the adjustment slider, and the compression spring is installed on the spring mounting plate.
[0012] Preferably, the visual inspection module includes a CCD camera, a lens, a camera mounting bracket, a bowl light source, a light source mounting bracket, a vertical slide rail, a vertical slider, a horizontal slider and a horizontal slide rail; a horizontal slider is provided on the horizontal slide rail, a vertical slide rail is provided on the horizontal slider, the vertical slide rail is arranged perpendicular to the horizontal slider, and a vertical slider is provided on the vertical slide rail; a camera mounting bracket is connected to the vertical slider, a CCD camera is installed on the camera mounting bracket, a lens is connected to the bottom of the CCD camera, a bowl light source is provided below the lens, and the bowl light source is connected to the camera mounting bracket through the light source mounting bracket.
[0013] Preferably, the shearing mechanism includes a gantry, a hydraulic cylinder, a shearing seat, an upper shearing blade, a lower fixed blade and a push rod; a hydraulic cylinder is installed on the inner top of the gantry, and the upper shearing blade is connected to the bottom hydraulic rod of the hydraulic cylinder. A shearing seat is provided below the upper shearing blade, and a lower fixed blade is provided inside the shearing seat. The upper shearing blade and the lower fixed blade are arranged correspondingly up and down, and the lower fixed blade can slide up and down in the slide rail inside the shearing seat, and a push rod is provided at the bottom of the lower fixed blade.
[0014] Preferably, the classification mechanism includes a classification motor, a motor bracket, a coupling, a bearing support, a classification tray, a main shaft and a bearing cover; the classification motor is installed on the motor bracket, the output shaft of the classification motor is connected to the main shaft through a coupling, the end of the main shaft is installed on the bearing support, the outer side of the bearing support is provided with a bearing cover, a classification tray is provided above the main shaft, and the classification tray is connected to the main shaft.
[0015] Preferably, temporary storage boxes are provided on both sides of the classification mechanism, and the temporary storage boxes on both sides are arranged correspondingly.
[0016] The present invention also provides a method for detecting the pitch, diameter, and forming height of pre-twisted wires of a connecting fitting. The method is implemented using the above-mentioned device for detecting the pitch, diameter, and forming height of pre-twisted wires of a connecting fitting. The method comprises the following steps:
[0017] Step 1: Image acquisition: Use CCD camera trigger mode to collect images regularly to achieve consistent input;
[0018] Step 2: Image preprocessing: Use a lightweight convolutional neural network (CNN) to denoise the original image. The network structure adopts the Denoising Autoencoder. The input is the original image, and the output is the image with background noise and stripe interference removed.
[0019] The Canny edge detection algorithm is used to extract the edge of the pre-twisted wire contour. The extraction process is combined with double-threshold edge positioning to avoid false detection caused by uneven intensity in the image. The output result is a binary edge image for use by the parameter extraction module.
[0020] Step 3: Parameter extraction: Use the edge map for horizontal projection to extract the peak position of the spiral line, fit the minimum inscribed circle of the inner circle contour to calculate the vertical distance between adjacent peaks, which is the pitch;
[0021] Select the y value of a certain cross section along the axial direction of the pre-twisted wire in the image, extract the horizontal coordinates of all edge points from the cross section, and find the two valid edge points x on the left and right of the cross section. left ,x right , pixel width is D px =x right −x left , converted into physical size is diameter;
[0022] Extract the vertical grayscale change curve, and obtain the height corresponding to the grayscale difference between the bottom and the top, which is the forming height;
[0023] After the above treatment, the pre-twisted wire is qualified if its diameter, forming height and pitch are within the specified error range, otherwise it is unqualified.
[0024] Finally, each module was integrated into a complete pipeline, and a user interface was developed based on QT Creator to display the total number of inspections, the number of unqualified products, the number of qualified products, and the pass rate.
[0025] By implementing this technical solution, we achieve high-precision detection of key parameters such as pitch, diameter, and forming height during the production of pre-twisted wire for connector fittings. We obtain stable, clear images for precise measurement in real-world industrial environments, such as those with unstable lighting and complex backgrounds. This reduces the impact of image noise and distortion on measurement results, improving the accuracy of edge recognition and feature extraction. This automation and real-time nature of the pre-twisted wire detection process enhances production efficiency and intelligent classification capabilities.
[0026] Compared with the prior art, the present invention has the following beneficial effects:
[0027] 1. The present invention adopts an industrial camera with a bowl-shaped light source, which can provide uniform lighting under different working conditions, effectively suppress interference factors such as light spots, shadows, and reflections, ensure stable image quality, and improve measurement accuracy.
[0028] 2. This invention introduces a convolutional neural network to learn and repair noise, blur, and slight deformation in images, improving edge clarity and contour integrity. This significantly outperforms traditional image filtering algorithms and facilitates the precise extraction of subsequent parameters. The proposed algorithm can simultaneously extract three key parameters of the pre-twisted wire: pitch, diameter, and forming height. All of these parameters are based on the same image processing flow, simplifying the system structure and improving detection efficiency.
[0029] 3. This invention links an industrial camera with a PLC system to establish a real-time image acquisition and parameter calculation mechanism, enabling real-time updating and classification output of inspection results. This facilitates automatic rejection of defective products and reduces manual intervention costs. It utilizes a structured contour extraction method based on Canny edge detection, combined with a neural network preprocessing mechanism. This system is suitable for pre-twisted wires of various wire diameters, different forming states, and diverse background conditions, offering strong adaptability and versatility.
[0030] 4. The overall solution of the present invention is modular and easy to integrate, suitable for high-speed production lines and automated testing equipment, and meets the actual needs of new energy, electric power, rail transit and other industries for quality control of pre-twisted wires of connecting fittings. BRIEF DESCRIPTION OF THE DRAWINGS
[0031] Figure 1 It is a schematic diagram of the overall structure of the present invention;
[0032] Figure 2 It is a workflow diagram of the present invention;
[0033] Figure 3 It is a structural schematic diagram of the feeding mechanism in the present invention;
[0034] Figure 4 for Figure 3 Another perspective structural diagram;
[0035] Figure 5 Schematic diagram of the structure of the visual detection module in the present invention;
[0036] Figure 6 Schematic diagram of the structure of the shearing mechanism in the present invention;
[0037] Figure 7 It is a structural diagram of the classification mechanism in the present invention. DETAILED DESCRIPTION
[0038] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings so that those skilled in the art can better understand the advantages and features of the present invention and thus more clearly define the scope of protection of the present invention. The embodiments described in the present invention are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative work shall fall within the scope of protection of the present invention.
[0039] like Figure 1 As shown, a device for detecting the pitch, diameter and forming height of pre-twisted wires of connecting hardware includes a workbench 1, on which are respectively provided a feeding mechanism 2, a distance measuring sensor 7, a visual inspection module 3 equipped with an industrial camera, a shearing mechanism 4, a sorting mechanism 5 and a display screen 6;
[0040] The front end of the shearing mechanism 4 is respectively provided with a feeding mechanism 2, a visual detection module 3 and a display screen 6. The feeding mechanism 2 is arranged on one side of the visual detection module 3. A ranging sensor 7 is installed on the feeding mechanism 2. The display screen 6 is arranged on the other side of the visual detection module 3; the rear end of the shearing mechanism 4 is provided with a classification mechanism 5.
[0041] In this embodiment, the detection device is designed based on the characteristics of metal pre-twisted wire, and is mainly used to extract key dimensional parameters of the pre-twisted wire, such as pitch, forming height, and diameter, which directly affect the assembly accuracy of transmission line hardware, and to achieve accurate classification by judging whether each parameter is within the specified error range.
[0042] like Figure 2 As shown in the figure, after the pre-twisted wire is formed, it first passes through a distance-measuring sensor on one side of the center hole of the feed mechanism. Next, it enters the center hole of the feed mechanism, where a mechanical anti-shake structure reduces vertical and horizontal vibration of the pre-twisted wire. Next, a visual inspection device completes image acquisition, and image processing algorithms extract dimensional parameters such as the pre-twisted wire pitch, forming height, and twisted wire diameter. After image acquisition, a fixed-length closed-loop control system is constructed using a high-precision photoelectric encoder based on the measurement results of the distance-measuring sensor to cut the pre-twisted wire to a fixed length. Finally, the pre-twisted wire is classified into qualified and unqualified products based on the parameters extracted by the visual inspection system. A classification mechanism controls the motor to achieve classified storage of the pre-twisted wire.
[0043] Specifically, such as Figure 3-Figure 4 As shown, the feeding mechanism 2 includes a mounting bracket 21, a support ring 22, a pull rod 23, an adjustment slider 24 with adjustable inner diameter, a tightening knob 25, a compression spring 26, a rubber wheel 27, a rubber wheel frame 28, and a spring mounting plate 29;
[0044] A support ring 22 is provided on the inner side of the mounting bracket 21, and the support ring 22 is connected to the mounting bracket 21 through a pull rod 23; four groups of anti-shake components are provided on the inner side of the support ring 22, and the four groups of anti-shake components are arranged perpendicular to each other in pairs, and a central hole 290 for the pre-twisted wire to pass through is formed in the middle of the four groups of anti-shake components; each group of anti-shake components includes a rubber wheel frame 28, and a rubber wheel 27 is installed at one end of the rubber wheel frame 28 close to the middle position of the support ring 22, and an adjusting slider 24 is provided at the other end of the rubber wheel frame 28 away from the middle position of the support ring 22, and a tightening knob 25 is provided on the adjusting slider 24, and a compression spring 26 is provided between the rubber wheel frame 28 and the adjusting slider 24, and the compression spring 26 is installed on the spring mounting plate 29.
[0045] In this embodiment, after the pre-twisted wire is formed, it is first inserted into the center hole 290 of the anti-shake mechanism. The diameter of the center hole 290 is adjusted using the four-way adjustment slider 24 so that the outer circle of a certain point on the pre-twisted wire is tangent to the outer arc of the four rubber wheels 27. This ensures that the pre-twisted wire rolls between the four rubber wheels 27, preventing scratches on the pre-twisted wire surface. Compression springs 26 are installed between the rubber wheel frame 28 and the adjustment slider 24, allowing the four rubber wheels 27 to float in four directions. The compression springs 26 eliminate vibration caused by stress during the pre-twisted wire processing, thereby ensuring the accuracy of subsequent visual inspection and cut-to-length cutting of the pre-twisted wire.
[0046] Specifically, such as Figure 5 As shown, the visual inspection module 3 includes a CCD camera 31, a lens 32, a camera mounting bracket 33, a bowl light source 34, a light source mounting bracket 35, a vertical slide rail 36, a vertical slider 37, a horizontal slider 38 and a horizontal slide rail 39; a horizontal slider 38 is provided on the horizontal slide rail 39, and a vertical slide rail 36 is provided on the horizontal slider 38. The vertical slide rail 36 is arranged perpendicular to the horizontal slider 38, and a vertical slider 37 is provided on the vertical slide rail 36; the vertical slider 37 is connected to the camera mounting bracket 33, and the CCD camera 31 is installed on the camera mounting bracket 33, the lens 32 is connected to the bottom of the CCD camera 31, and a bowl light source 34 is provided below the lens 32, and the bowl light source 34 is connected to the camera mounting bracket 33 through the light source mounting bracket 35.
[0047] In this embodiment, a bowl light source is provided around the lens 32 of the visual inspection device. When it is necessary to inspect pre-twisted wires of different pitches, the vertical slider 37 can be moved on the vertical slide rail 36 to move the camera, lens module, and bowl light source as a whole upward or downward, thereby increasing or shortening the camera focal length, so that the camera image shows a complete appearance outline that can just detect the geometric features of the pre-twisted wire. After determining the position, use the tightening knob to fix it. The horizontal slider 38 and the horizontal slide rail 39 can adjust the horizontal position of the entire visual inspection device. When inspecting pre-twisted wires of different forming heights, the position can be adjusted to obtain an imaging photo of the complete outline.
[0048] Specifically, such as Figure 6 As shown, the shearing mechanism 4 includes a gantry 41, a hydraulic cylinder 42, a shearing seat 43, an upper shearing blade 44, a lower fixed blade 45 and a push rod 46; a hydraulic cylinder 42 is installed on the inner top of the gantry 41, and the upper shearing blade 44 is connected to the bottom hydraulic rod of the hydraulic cylinder 42, and a shearing seat 43 is provided below the upper shearing blade 44, and a lower fixed blade 45 is provided inside the shearing seat 43. The upper shearing blade 44 and the lower fixed blade 45 are arranged in correspondence with each other up and down, and the lower fixed blade 45 can slide up and down in the slide rail inside the shearing seat 43, and a push rod 46 is provided at the bottom of the lower fixed blade 45. Among them, the distance between the upper and lower shearing blades is adjusted according to the diameter of the pre-twisted wire. The upper shearing blade is installed on the cylinder rod, and the change in distance is achieved by replacing push rods of different lengths. Therefore, the lower shearing blade moves up and down on the internal slide rail.
[0049] In this embodiment, a distance sensor is installed near the center hole of the feed mechanism. This distance sensor detects the length of the pre-twisted wire passing above. When the detected length equals a fixed value input by the user, feedback is sent to the control system, triggering the shear mechanism to operate and cut the pre-twisted wire. Considering the large diameter of the pre-twisted wire and the need to operate on a high-speed production line, high shearing efficiency, convenient operation, and simple maintenance are required. A hydraulic cylinder is used in conjunction with an alloy steel cutting blade to design a small gantry-style shear mechanism. The upper shear blade 44 is mounted on a hydraulic rod and can move up and down, while the lower fixed blade 45 is mounted inside the shear seat 43 and can slide up and down within the shear seat 43's slide rails. The bottom push rod 46 ensures that the lower fixed blade 45 remains stationary during shearing mechanism operation, allowing the upper shear blade 44 to cooperate with the downward movement of the pre-twisted wire. When cutting large-diameter pre-twisted wire, the push rod 46 can be replaced with a shorter one, and the lower fixed blade 45 can be moved downward to increase the distance between the upper and lower shear blades 44, 45. The hydraulic cylinder must meet the thrust requirements required to cut steel pre-twisted wire, have a small travel range, and a high reciprocating speed. Therefore, the standard cylinder model 12000-032025 is selected, with a cylinder diameter of 32mm and a stroke of 25mm.
[0050] Specifically, such as Figure 7As shown, the classification mechanism 5 includes a classification motor 51, a motor bracket 52, a coupling 53, a bearing support 54, a classification tray 55, a main shaft 56 and a bearing cover 57; the classification motor 51 is installed on the motor bracket 52, and the output shaft of the classification motor 51 is connected to the main shaft 56 through the coupling 53. The end of the main shaft 56 is installed on the bearing support 54, and the outer side of the bearing support 54 is provided with a bearing cover 57. A classification tray 55 is provided above the main shaft 56, and the classification tray 55 is connected to the main shaft 56.
[0051] Wherein, temporary storage boxes 58 are provided on both sides of the classification mechanism 5 , and the temporary storage boxes 58 on both sides are correspondingly arranged.
[0052] In this embodiment, the pre-twisted wires cut by the shearing mechanism fall onto a sorting tray 55 in the sorting mechanism, which then sorts and stores qualified and unqualified pre-twisted wires. Upon receiving a signal from the control system, the sorting motor 51 rotates the sorting tray 55 in both directions by driving the spindle 56 in both directions, causing the pre-twisted wires on the tray to fall into corresponding temporary storage boxes 58 for classified storage.
[0053] A method for detecting the pitch, diameter, and forming height of pre-twisted wires of a connecting fitting is implemented using a device for detecting the pitch, diameter, and forming height of pre-twisted wires of a connecting fitting as described above. The method comprises the following steps:
[0054] Step 1: Image acquisition. A CCD camera trigger mode is used to capture images at regular intervals for consistent input. This CCD camera, a Basler acA2440-75µm global shutter industrial camera, boasts high resolution and zero smear, making it suitable for high-speed imaging of pre-twisted wire details. A bowl-shaped light source provides surround, low-reflection, high-contrast illumination, highlighting the pre-twisted wire's edges and minimizing shadows and background interference.
[0055] Step 2: Image preprocessing. First, a lightweight convolutional neural network (CNN) is used to denoise the original image. The network structure adopts the Denoising Autoencoder. The input is the original image, and the output is an image with background noise and stripe interference removed.
[0056] The second is to use the Canny edge detection algorithm to extract the edge of the pre-twisted wire contour. The extraction process is combined with double threshold edge positioning to avoid false detection caused by uneven intensity in the image. The output result is a binary edge image for use by the parameter extraction module.
[0057] Step 3: Parameter extraction: Use the edge map for horizontal projection to extract the peak position of the spiral line, fit the minimum inscribed circle of the inner circle contour to calculate the vertical distance between adjacent peaks, which is the pitch;
[0058] Select the y value of a certain cross section along the axial direction of the pre-twisted wire in the image, extract the horizontal coordinates of all edge points from the cross section, and find the two valid edge points x on the left and right of the cross section. left ,x right , pixel width is D px =x right −x left , converted into physical size is diameter;
[0059] Extract the vertical grayscale change curve, and obtain the height corresponding to the grayscale difference between the bottom and the top, which is the forming height;
[0060] After the above treatment, the pre-twisted wire is qualified if its diameter, forming height and pitch are within the specified error range, otherwise it is unqualified.
[0061] Finally, each module was integrated into a complete pipeline, and a user interface was developed based on QT Creator to display the total number of inspections, the number of unqualified products, the number of qualified products, and the pass rate.
[0062] In summary, the present invention can realize online detection of pre-twisted wire diameter, forming height, and pitch, thereby improving detection efficiency and accuracy, thereby improving product quality.
[0063] The present invention is applicable to the following fields:
[0064] Power transmission and distribution line field: Pre-twisted wire for tension clamps, suspension clamps, splicing bars, protective bars, repair bars, jumper clamps, and wedge clamps used in overhead transmission and distribution lines of various voltage levels. After key processes such as pre-twisted wire forming and heat treatment, we conduct rapid spot checks or full inspections on semi-finished or finished products to promptly identify and adjust production process deviations and ensure product dimensional consistency.
[0065] Railway: Pre-twisted fittings used for connecting, fixing, terminating, and protecting overhead contact wires and catenary cables. Similar to the power sector, these fittings ensure dimensional accuracy and guarantee the safe and stable operation of the overhead contact network.
[0066] Spring manufacturing industry: used for various types of coil springs to detect key geometric parameters such as spring pitch, wire diameter, outer diameter, free height, etc., for production process control and finished product inspection to ensure that the mechanical properties of the spring meet the design requirements.
[0067] Medical device manufacturing: Medical implants or devices with fine spiral structures, such as vascular stents, orthopedic screws, and guidewires. While ensuring sterility and compatibility, high-precision non-destructive testing of the pitch, diameter, and unfolded shape of these tiny spiral structures is performed to ensure functionality and safety.
[0068] The core value of this invention lies in providing an efficient, precise, and automated solution for detecting the key geometric parameters of pre-twisted wire, a specific yet important type of helical hardware. Its primary application scenarios are pre-twisted wire production quality control, factory inspection, and product development. Furthermore, based on the technical principles of machine vision and AI, it has the potential to be expanded to detect the geometric parameters of other products with similar helical geometric characteristics, potentially playing a role in a wider range of industrial manufacturing and quality inspection fields.
[0069] The descriptions and practices disclosed in this invention are easy to understand and comprehend for those skilled in the art, and modifications and refinements may be made without departing from the principles of the invention. Therefore, modifications and improvements made without departing from the spirit of the invention should also be considered within the scope of protection of this invention.
Claims
1. A device for detecting the pitch, diameter and forming height of pre-twisted wires of connecting hardware, characterized in that: The workbench (1) comprises a feeding mechanism (2), a distance sensor (7), a visual inspection module (3) equipped with an industrial camera, a shearing mechanism (4), a classification mechanism 5 and a display screen (6); The front end of the shearing mechanism (4) is provided with a feeding mechanism (2), a visual detection module (3) and a display screen (6), respectively; the feeding mechanism (2) is provided on one side of the visual detection module (3); a distance sensor (7) is installed on the feeding mechanism (2); and the display screen (6) is provided on the other side of the visual detection module (3); the rear end of the shearing mechanism (4) is provided with a classification mechanism (5); The feeding mechanism (2) includes a mounting bracket (21), a support ring (22), a pull rod (23), an adjusting slider (24) with an adjustable inner diameter, a tightening knob (25), a compression spring (26), a rubber wheel (27), a rubber wheel frame (28), and a spring mounting plate (29); A support ring (22) is provided on the inner side of the mounting bracket (21), and the support ring (22) is connected to the mounting bracket (21) through a pull rod (23); four groups of anti-shake components are provided on the inner side of the support ring (22), and the four groups of anti-shake components are arranged perpendicular to each other in pairs, and a central hole (290) for the pre-twisted wire to pass through is formed in the middle of the four groups of anti-shake components; each group of anti-shake components includes a rubber wheel frame (28), and a rubber wheel (27) is installed at one end of the rubber wheel frame (28) close to the middle position of the support ring (22), and an adjustment slider (24) is provided at the other end of the rubber wheel frame (28) away from the middle position of the support ring (22), and a tightening knob (25) is provided on the adjustment slider (24), and a compression spring (26) is provided between the rubber wheel frame (28) and the adjustment slider (24), and the compression spring (26) is installed on the spring mounting plate (29).
2. The device for detecting the pitch, diameter and forming height of pre-twisted wires of a connection fitting according to claim 1, characterized in that: The visual inspection module (3) comprises a CCD camera (31), a lens (32), a camera mounting bracket (33), a bowl light source (34), a light source mounting bracket (35), a vertical slide rail (36), a vertical slider (37), a horizontal slider (38) and a horizontal slide rail (39); the horizontal slide rail (39) is provided with a horizontal slider (38), the horizontal slider (38) is provided with a vertical slide rail (36), the vertical slide rail (36) and the horizontal slider (38) are arranged perpendicularly, and the vertical slide rail (36) is provided with a vertical slider (37); the vertical slider (37) is connected to the camera mounting bracket (33), the camera mounting bracket (33) is mounted with a CCD camera (31), the lower part of the CCD camera (31) is connected with a lens (32), the lower part of the lens (32) is provided with a bowl light source (34), and the bowl light source (34) is connected to the camera mounting bracket (33) through the light source mounting bracket (35).
3. The device for detecting the pitch, diameter and forming height of pre-twisted wires of a connection fitting according to claim 1, characterized in that: The shearing mechanism (4) comprises a gantry (41), a hydraulic cylinder (42), a shearing seat (43), an upper shearing blade (44), a lower fixed blade (45) and a push rod (46); a hydraulic cylinder (42) is installed on the inner top of the gantry (41); an upper shearing blade (44) is connected to a hydraulic rod at the bottom of the hydraulic cylinder (42); a shearing seat (43) is provided below the upper shearing blade (44); a lower fixed blade (45) is provided inside the shearing seat (43); the upper shearing blade (44) and the lower fixed blade (45) are arranged in correspondence with each other in the upper and lower directions; the lower fixed blade (45) can slide up and down in a slide rail inside the shearing seat (43); and a push rod (46) is provided at the bottom of the lower fixed blade (45).
4. The device for detecting the pitch, diameter and forming height of pre-twisted wires of a connection fitting according to claim 1, characterized in that: The classification mechanism (5) comprises a classification motor (51), a motor bracket (52), a coupling (53), a bearing support (54), a classification tray (55), a main shaft (56) and a bearing cover (57); the classification motor (51) is mounted on the motor bracket (52); the output shaft of the classification motor (51) is connected to the main shaft (56) through the coupling (53); the end of the main shaft (56) is mounted on the bearing support (54); the outer side of the bearing support (54) is provided with a bearing cover (57); a classification tray (55) is provided above the main shaft (56); and the classification tray (55) is connected to the main shaft (56).
5. The device for detecting the pitch, diameter and forming height of pre-twisted wires of a connection fitting according to claim 4, characterized in that: Temporary storage boxes (58) are provided on both sides of the classification mechanism (5), and the temporary storage boxes (58) on both sides are correspondingly arranged.
6. A method for detecting the pitch, diameter, and forming height of pre-twisted wires of a connection fitting, the method being implemented using a device for detecting the pitch, diameter, and forming height of pre-twisted wires of a connection fitting according to any one of claims 1 to 5, characterized in that: The method comprises the following steps: Step 1: Image acquisition: Use CCD camera trigger mode to collect images regularly to achieve consistent input; Step 2: Image preprocessing: Use a lightweight convolutional neural network (CNN) to denoise the original image. The network structure adopts the Denoising Autoencoder. The input is the original image, and the output is an image without background noise and stripe interference. The Canny edge detection algorithm is used to extract the edge of the pre-twisted wire contour. The extraction process is combined with double-threshold edge positioning to avoid false detection caused by uneven intensity in the image. The output result is a binary edge image for use by the parameter extraction module. Step 3: Parameter extraction: Use the edge map for horizontal projection to extract the peak position of the spiral line, fit the minimum inscribed circle of the inner circle contour to calculate the vertical distance between adjacent peaks, which is the pitch; Select the y value of a certain cross section along the axial direction of the pre-twisted wire in the image, extract the horizontal coordinates of all edge points from the cross section, and find the two valid edge points x on the left and right of the cross section. left ,x right , pixel width is D px =x right −x left , converted into physical size is diameter; Extract the vertical grayscale change curve, and obtain the height corresponding to the grayscale difference between the bottom and the top, which is the forming height; After the above treatment, the pre-twisted wire is qualified if its diameter, forming height and pitch are within the specified error range, otherwise it is unqualified. Finally, each module was integrated into a complete pipeline, and a user interface was developed based on QT Creator to display the total number of inspections, the number of unqualified products, the number of qualified products, and the pass rate.