A double-pull wire detection device and method based on visual and photoelectric composite technology

Abstract

The application discloses a double-pull wire detection device and method based on vision and photoelectric composite technology, and belongs to the technical field of detection. The device adopts a vision and photoelectric fusion mode, and can detect a series of quality problems such as whether the pull wire exists, the pull wire joint, transparent paper splicing, double-pull wire overall or local deviation, whether the pull wire is in the middle of the U-shaped pull head or not, regardless of the pull wire category, such as transparent pull wire, gold pull wire, laser silvering and the like. The method solves the problem that the pull wire cannot be accurately removed due to the lack of phase information, realizes package statistics, facilitates intelligent data management, provides strong support for production, and has a wide application space.

Description

Technical Field

[0001] The present invention belongs to the field of detection technology, and in particular relates to a double-pull wire detection device and method based on vision and photoelectric composite technology. Background Art

[0002] To attract consumers, the trend toward customized cigarettes is becoming mainstream. In addition to diverse packaging styles, cigarette packs also feature unique opening methods, including the common flip-top, full-open, side-open, bottom-open, and diagonal-open styles. Many high-end cigarettes use full-open or side-open packaging. Unlike traditional flip-top packaging, these require two pull cords to open the outer transparent paper. Compared to single-pull cords, double-pull cords present not only quality issues such as missing cords, cord splices, and transparent paper splicing, but can also cause overall cord offset and incorrect spacing.

[0003] Currently, there's no testing equipment specifically designed for double-pull cables on the market. Many devices are simply modified versions of existing single-pull cable testing, providing only the ability to detect cable presence and splice joints. However, they can't effectively verify the proper spacing between the double-pull cables or whether the cable is centered within the U-shaped puller. Therefore, the market urgently needs a dedicated double-pull cable testing device to comprehensively address these quality issues. Summary of the Invention

[0004] In response to the above technical problems existing in the prior art, the present invention proposes a double-wire detection device and method based on vision and photoelectric composite technology, which has a reasonable design, overcomes the shortcomings of the prior art, and has good effects.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A double-pull wire detection device based on vision and photoelectric composite technology, including an integrated dedicated controller, an image collector and an encoder;

[0007] An all-in-one dedicated controller configured for human-machine interactive display, camera light source control, image and optoelectronic data processing, I / O control, and communication with the electrical cabinet; it includes a display module, core processing module, embedded ARM control module, power module, and I / O module;

[0008] An image collector configured to capture images of the wires and detect splicing joints; comprising a black and white camera, a low-distortion lens, an LED strip light source, and a through-beam fiber optic sensor;

[0009] The encoder is configured to determine the working status of the machine and accurately count and eliminate wire defects;

[0010] When the encoder captures the machine working signal and the machine is not running with empty materials, the integrated dedicated controller controls the light source to flash, the camera to take pictures, and the photoelectric sensor to collect data. It then processes the image data and sends the detection results to the embedded ARM control module. The embedded ARM control module controls the I / O output and communicates with the electrical cabinet. If a defect is detected, the electrical cabinet is controlled to remove it at the corresponding workstation.

[0011] Preferably, the display module uses a liquid crystal color touch screen as a window for interacting with the user to display the human-computer interaction program;

[0012] The core processing module is configured to process the cable images captured by the black and white camera, calculate and determine whether the cable is missing, offset, or in the middle of the U-shaped pull head, and send the calculation results to the embedded ARM control module;

[0013] The embedded ARM control module is configured to control the strobe light source and the black and white camera, receive processing results from the host computer program, control I / O input and output, communicate with the electrical cabinet, and send the detection results to the electrical cabinet system to achieve online precise rejection;

[0014] The power module is configured to provide power to the display module, the core processing module, and the embedded ARM control module;

[0015] The I / O module is configured to be responsible for external input and output.

[0016] Preferably, the black and white camera is configured to clearly capture the image of the wire;

[0017] The LED strip light source is configured to focus light on the pull wire area through a focusing lens so as to clearly distinguish the pull wire from the transparent paper; the LED strip light source is composed of multiple lamp beads.

[0018] Preferably, the low-distortion lens is a 6mm low-distortion lens.

[0019] In addition, the present invention also provides a double-pull wire detection method based on a visual and photoelectric composite technology. The method uses the double-pull wire detection device based on a visual and photoelectric composite technology as described above, and specifically includes the following steps:

[0020] Step S1: vertically projecting the collected wireline image to obtain one-dimensional curve data;

[0021] Step S2: Find all candidate peak points in the calculation curve;

[0022] Step S3: determining the exact position of the edge of the pull wire and the U-shaped pull head from the candidate peak points;

[0023] Step S4: learning and calculating the center position templateCenter of the double-pull wire of the template;

[0024] Step S5: Calculate the center distance between the real-time double pull wires and the center distance between the template double pull wires, and calculate the center position deviation between the pull wires and the corresponding U-shaped sliders to determine whether the pull wires are missing, offset as a whole or in part, and whether the pull wires are in the middle of the U-shaped sliders;

[0025] Step S6: Photoelectric technology is used to detect defects including wire joints and transparent paper splicing.

[0026] Preferably, in step S2, all candidate peak points in the curve are calculated as follows:

[0027]

[0028] Where x is a one-dimensional curve array and x.Length is the length of the array. The curve value of the candidate peak point is greater than the threshold 100 and is a local peak.

[0029] Preferably, in step S3, the calculation formula for determining the exact position pos of the edge of the pull line and the U-shaped pull head from the candidate peak points is as follows:

[0030]

[0031] Where list.Length is the length of the candidate peak array.

[0032] Preferably, in step S4, the center position of the template double-pull wire templateCenter is:

[0033]

[0034] Among them, templatepos[0] stores the left edge position of the first pull wire of the template double pull wire, templatepos[1] stores the right edge position of the first pull wire of the template double pull wire, templatepos[2] stores the left edge position of the second pull wire of the template double pull wire, and templatepos[3] stores the right edge position of the second pull wire of the template double pull wire.

[0035] Preferably, in step S5, during real-time detection, when there is only a pull wire without a U-shaped pull head, the edge position of the pull wire is stored in POS, and the center distance CenterSpace of the double pull wire is calculated to be equal to 6mm to determine whether the pull wire is broken or there is a local offset defect; if CenterSpace meets the requirements, then calculate whether the center position realtimeCenter of the double pull wire and the center position templateCenter of the template double pull wire are within the set deviation Offset range to determine the overall offset defect;

[0036]

[0037]

[0038] When there are pull cords and U-shaped sliders, the POS stores the edge positions of the pull cords and the U-shaped sliders. The center positions of each pull cord (lineCenter1, lineCenter2) and the positions of each U-shaped slider (paperCenter1, paperCenter2) are calculated to determine whether the pull cord is in the middle of the U-shaped slider.

[0039]

[0040] The beneficial technical effects brought about by the present invention are:

[0041] The device, which integrates vision and optoelectronics, can detect a wide range of quality issues, including the presence of wires, wire joints, transparent paper splicing, overall or partial offset of double wires, and whether the wire is centered in the U-shaped puller, regardless of wire type. This device solves the problem of accurately rejecting wires due to a lack of phase information, implements per-package statistics, facilitates intelligent data management, and provides strong support for production, with a wide range of applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is the electrical principle block diagram of the present invention;

[0043] Figure 2 This is a schematic structural diagram of the integrated dedicated controller of the present invention;

[0044] Among them, 1-USB interface; 2-power switch; 3-input and output interface; 4-power interface;

[0045] Figure 3 Schematic diagram of the structure of the image collector of the present invention;

[0046] 5- black and white camera; 6- low distortion lens; 7- LED strip light source; 8- focusing lens; 9- through-beam sensor fiber head;

[0047] Figure 4 This is the normal double pull line without U-shaped pull head and the corresponding projection curve;

[0048] Figure 5 This is the double pull line and corresponding projection curve of a normal U-shaped pull head;

[0049] Figure 6 It is the missing tension line and the corresponding projection curve;

[0050] Figure 7 The double-pull wire deviates from the U-shaped pull head and the corresponding projection curve;

[0051] Figure 8 The local offset of the double tension wire and the corresponding projection curve. DETAILED DESCRIPTION

[0052] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments:

[0053] A double-wire detection device based on vision and photoelectric composite technology, the principle is as follows Figure 1 As shown, the detection device is composed of an integrated dedicated controller, an image acquisition system, and an encoder.

[0054] An integrated dedicated controller, the structure of which is as follows Figure 2 As shown, it consists of a display module, a core processing module, an embedded ARM control module, a power module, and an I / O module. The display module uses an 8.4-inch LCD color touch screen as a window for user interaction, displaying the human-computer interaction program. The core processing module processes the pull wire images captured by the camera, calculates and determines whether the pull wire is missing, offset, or in the middle of the U-shaped pull head, and sends the calculation results to the embedded ARM control module. The embedded ARM control module controls the strobe light source and the black and white camera, receives the processing results of the host computer program, controls the input and output of the I / O, communicates with the electrical cabinet, and sends the detection results to the electrical cabinet system to achieve online precise rejection. The power module is responsible for providing power to the display module, core processing module, and embedded ARM control module. The I / O module is responsible for external input and output.

[0055] The structure of the image collector is as follows Figure 3 As shown, it consists of a black and white camera, a low-distortion lens, an LED strip light source, a focusing lens and a beam-type fiber optic sensor. The present invention adopts an oblique lighting method to highlight the brightness of the edge of the wire so as to identify the position of the edge of the wire. Therefore, it is applicable to various types of wires without the need to distinguish. Compared with a color camera, the use of a black and white camera can capture the image of the wire more clearly, the contrast of the wire edge is more obvious, the sensitivity is higher, and the effect is more significant. The LED strip light source is installed vertically along the direction of the wire, and the focusing lens focuses the light on the wire area through the focusing lens so that the wire can be clearly distinguished from the transparent paper. In addition, the continuous and uninterrupted working characteristics of the wire require high-frequency lighting and shooting to ensure comprehensive coverage of the detection. Even if the light source adopts stroboscopic technology, its working intensity is much higher than that of other workstation detection products, and the decay rate is relatively fast. The LED strip light source is composed of multiple lamp beads. Even if a certain path fails, the entire light source will only dim slightly and can continue to provide lighting, and will not cause it to be unusable. Considering that the width of the wire is only 2.5mm, the present invention uses a 6mm low-distortion lens.

[0056] An encoder, mounted on the machine's moving gear, converts mechanical position into phase angle. The embedded ARM control module introduces the encoded signal, which serves three purposes. First, it accurately determines the machine's operating status, effectively reducing the idle time of the light source camera. Second, it precisely locates defect information on each pack of cigarettes, avoiding the previous problem of over-rejection or missed rejection due to a lack of phase information, thereby achieving precise rejection of string lines. Third, it enables accurate counting by pack, avoiding multiple, ineffective counts of the same pack, which hinders production management and problem identification.

[0057] A double-pull wire detection method based on visual and photoelectric composite technology uses a double-pull wire detection device based on visual and photoelectric composite technology. It can not only detect the presence of pull wires, pull wire joints, and transparent paper splicing, but also identify whether the double pull wires are offset as a whole or locally, and whether the pull wires are in the middle of the U-shaped pull head. It is characterized by including the following implementation steps:

[0058] S1. Project the collected wireline image vertically to obtain a one-dimensional curve, such as Figure 4 As shown. You can see that there are four peaks in the curve, which are the four edges of the two pull lines, and the rest is very flat. When the U-shaped pull head of each pack of cigarettes is illuminated, you can see that the pull line is in the middle of the U-shaped pull head, so two small peaks will be formed on both sides of each pull line, as shown in the figure. Figure 5 shown.

[0059] S2. Find all candidate peak points in the calculation curve. The calculation method is as follows:

[0060]

[0061] Where x is a one-dimensional curve array and x.Length is the length of the array. The curve value of the candidate peak point must be greater than the threshold of 100 and be a local peak.

[0062] S3. The width of the pull cord corresponds to 52 pixels. The distance between the edge of the U-shaped slider and the edge of the pull cord is approximately 30 to 45 pixels. The edge of the pull cord and the U-shaped slider must be the maximum value within the local region [-25, 25]. The location of the peak point is determined from the candidate peak points. The calculation formula is as follows:

[0063]

[0064] Where list.Length is the length of the candidate peak array.

[0065] S4. Learn the standard template double-pull line and calculate the template pull line position templatepos according to the above formula. The template double-pull line center position templateCenter is:

[0066]

[0067] Wherein, templatepos[0] stores the left edge position of the first pull line of the template double pull line, templatepos[1] stores the right edge position of the first pull line of the template double pull line, templatepos[2] stores the left edge position of the second pull line of the template double pull line, and templatepos[3] stores the right edge position of the second pull line of the template double pull line.

[0068] S5. When only the pull line is detected in real time without the U-shaped pull head, the edge position of the pull line is stored in pos, and whether the center distance CenterSpace of the double pull line is equal to 6mm is calculated to determine whether the pull line is broken, locally offset, or the like; if the CenterSpace meets the requirements, whether the overall offset of the double pull line center position realtimeCenter and the center position templateCenter of the template double pull line is within the set deviation Offset range is calculated to determine whether the overall offset is defective or the like.

[0069]

[0070]

[0071] When there is a pull line and a U-shaped pull head, the edge position of the pull line and the U-shaped pull head is stored in pos, and whether the center position lineCenter1 and lineCenter2 of each pull line and the position paperCenter1 and paperCenter2 of each U-shaped pull head is within the set deviation Offset2 range is calculated to determine whether the pull line is in the middle of the U-shaped pull head.

[0072]

[0073] Figure 6 For the missing pull line 2, only two wave crests are found, and the center distance of the template pull line 2 and the center distance of the U-shaped pull head 2 are not within the set deviation range; Figure 7 For the overall offset of the pull line, the center distance of the template pull line 1 and the pull line 2, and the center distance of the U-shaped pull head 1 and the U-shaped pull head 2 are not within the set deviation range; Figure 8 For the pull line 2 edge curl, the center distance of the template pull line 2 is not within the set deviation range. Therefore, the detection results of Figures 6-8 are all unqualified.

[0074] S6. There are two ways to splice transparent paper: one is to use tape, and the other is to use a soldering iron. Soldering iron splicing is equivalent to using two layers of transparent paper. For a visual camera, the difference between one and two layers of transparent paper is not obvious. Therefore, a through-beam sensor fiber head can be installed above the lens and on the reflector. This allows for both wire splice detection and transparent paper splicing.

[0075] Vision technology accurately detects quality defects in double-pull cables, including missing wires, overall or partial offset, and whether they are centered within the U-shaped puller. Photoelectric technology detects quality defects such as wire joints and transparent paper splicing, overcoming the inaccurate visual recognition issues associated with soldering irons. Combining these two technologies effectively detects all quality defects in double-pull cables and enables precise online removal.

[0076] Of course, the above description is not a limitation of the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by technicians in this technical field within the essential scope of the present invention should also fall within the scope of protection of the present invention.

Claims

1. A double-pull wire detection method based on vision and photoelectric composite technology, characterized by: A double-wire detection device based on vision and photoelectric composite technology is used, which includes an integrated dedicated controller, image collector and encoder; An integrated dedicated controller configured for human-machine interactive display, camera light source control, image and optoelectronic data processing, I / O control, and communication with the electrical cabinet; Including display module, core processing module, embedded ARM control module, power module and I / O module; An image collector configured to capture images of the wires and detect splicing joints; comprising a black and white camera, a low-distortion lens, an LED strip light source, and a through-beam fiber optic sensor; The encoder is configured to determine the working status of the machine and accurately count and eliminate wire defects; When the encoder captures the machine's operating signal and the machine is running without empty materials, the integrated dedicated controller controls the light source to flash, the camera to take pictures, and the through-beam fiber optic sensor to collect data. It then processes the image data and sends the detection results to the embedded ARM control module. The embedded ARM control module controls the I / O output and communicates with the electrical cabinet. If a defect is detected, the electrical cabinet is controlled to remove the defect at the corresponding workstation. The display module uses a liquid crystal color touch screen as a window for interaction with the user and displays the human-computer interaction program; The core processing module is configured to process the cable images captured by the black and white camera, calculate and determine whether the cable is missing, offset, or in the middle of the U-shaped pull head, and send the calculation results to the embedded ARM control module; The embedded ARM control module is configured to control the strobe light source and the black and white camera, receive processing results from the host computer program, control I / O input and output, communicate with the electrical cabinet, and send the detection results to the electrical cabinet system to achieve online precise rejection; The power module is configured to provide power to the display module, the core processing module, and the embedded ARM control module; The I / O module is configured to be responsible for external input and output; a black and white camera configured to clearly capture an image of the pull line; The LED strip light source is configured to focus light on the pull-cord area through a condenser lens so as to clearly distinguish the pull-cord from the transparent paper; the LED strip light source is composed of multiple lamp beads; the low-distortion lens is a 6mm low-distortion lens; The detection method specifically comprises the following steps: Step S1: vertically projecting the collected wireline image to obtain one-dimensional curve data; Step S2: Find all candidate peak points in the calculation curve; Step S3: determining the exact position of the edge of the pull wire and the U-shaped pull head from the candidate peak points; Step S4: Learn to calculate the center position of the double tension wire of the template ; Step S5: Calculate the center distance between the real-time double pull wires and the center distance between the template double pull wires, and calculate the center position deviation between the pull wires and the corresponding U-shaped sliders to determine whether the pull wires are missing, offset as a whole or in part, and whether the pull wires are in the middle of the U-shaped sliders; In step S5, during real-time detection, when there is only a pull line but no U-shaped pull head, The edge position of the wire is stored here, and the center distance of the double wire is calculated. Is it equal to 6mm to judge whether the wire is broken or has local deviation defects; if If it meets the requirements, calculate the center position of the double pull wire Center position of double tension line with template Is the deviation being set? The overall offset defect is determined within the specified range; When there is a pull cord and a U-shaped pull head, The position of the edge of the pull wire and the U-shaped pull head is stored in it, and the center position of each pull wire is calculated 、 The position of each U-shaped slider 、 Is the deviation being set? The range is used to determine whether the pull line is in the middle of the U-shaped pull head; ； Step S6: Photoelectric technology is used to detect defects including wire joints and transparent paper splicing.

2. The double-pull wire detection method based on vision and photoelectric composite technology according to claim 1 is characterized in that: In step S2, all candidate peak points in the curve are calculated as follows: in, is a one-dimensional curve array, The length of the array; the curve value of the candidate peak point is greater than the threshold 100 and is a local peak.

3. The double-pull wire detection method based on vision and photoelectric composite technology according to claim 1 is characterized in that: In step S3, the exact position of the edge of the pull line and the U-shaped pull head is determined from the candidate peak points. The calculation formula is as follows: in, is the length of the candidate peak array.

4. The double-pull wire detection method based on vision and photoelectric composite technology according to claim 1 is characterized in that: In step S4, the center position of the double-pull wire of the template for: Among them, templatepos[0] stores the left edge position of the first pull wire of the template double pull wire, templatepos[1] stores the right edge position of the first pull wire of the template double pull wire, templatepos[2] stores the left edge position of the second pull wire of the template double pull wire, and templatepos[3] stores the right edge position of the second pull wire of the template double pull wire.

Images