Broken wire detection system based on three-dimensional measurement

Abstract

The invention is suitable for the technical field of broken yarn detection, and provides a broken yarn detection system based on three-dimensional measurement, the broken yarn detection system comprises a laser mechanism, a camera array, an encoder, a single-chip microcomputer and a host, the laser mechanism is used for emitting a linear laser to irradiate a running yarn group, a highlight bright spot is formed on each yarn of the yarn group, and the camera array is used for detecting the broken yarn. The camera array is used for collecting bright spot images of a silk yarn group, the encoder is installed on a main shaft motor of the textile equipment and used for generating continuous pulse signals along with rotation of the main shaft motor, and the single chip microcomputer is in communication connection with the encoder and generates trigger signals of the images collected by the camera. High-brightness bright spot features are formed on each silk thread through linear laser, the three-dimensional space position of the silk thread is reconstructed in combination with image information collected by a camera array, the running state of the multiple rows of silk threads sequentially arranged in the depth direction of the optical axis of the laser can be recognized, and the possibility of missing detection caused by mutual shielding of the silk threads in the front row and the rear row is reduced.

Description

Technical Field

[0001] This invention relates to the field of wire breakage detection technology, and more specifically, to a wire breakage detection system based on three-dimensional measurement. Background Technology

[0002] As the textile industry rapidly develops towards high speed, intelligence, and high quality, the operating speed of mainstream textile equipment such as double needle bed warp knitting machines, KS warp knitting machines, and jacquard water jet looms continues to increase, and the control precision requirements for fabric weaving density and yarn fineness are constantly rising.

[0003] In the textile production process, yarn breakage is the most common production failure. If yarn breakage failure cannot be identified and handled in a timely and accurate manner, it will directly lead to quality problems such as weft breakage, holes, and defects in the fabric, resulting in a large waste of raw materials and production time. In severe cases, it can also cause secondary failures such as equipment entanglement and jamming, which will significantly reduce production efficiency and product qualification rate.

[0004] Therefore, high-performance yarn breakage detection technology is a core and crucial link in ensuring the continuous and stable operation of textile equipment and improving product quality.

[0005] Currently, the yarn breakage detection technologies used in the textile industry are mainly divided into two categories: contact detection and non-contact detection. Traditional contact yarn breakage detection is represented by tension-based and contact-bar-based detection schemes. Their core principle is to directly contact the yarn through a mechanical structure, utilizing the tension of the yarn during normal operation to maintain the steady state of the detection mechanism. When the yarn breaks, the tension disappears, and the detection mechanism triggers a switch signal to identify the yarn breakage. In non-contact detection schemes, through-beam and reflective photoelectric detection are among the earliest applied technologies. They determine the continuity or disconnection status of the yarn by the yarn's ability to block or reflect light.

[0006] However, contact detection structures generate continuous friction with high-speed running yarns, which can easily cause secondary damage such as wear and stretching to high-quality raw materials such as ultra-fine denier yarns, making them unsuitable for the production needs of high-end fabrics. While non-contact detection avoids direct contact with the yarns, it can only determine the presence or absence of broken yarns and cannot identify early anomalies such as yarn positional deviations or tension fluctuations. Therefore, a three-dimensional measurement-based broken yarn detection system is proposed to improve the existing problems. Summary of the Invention

[0007] In view of the shortcomings of the existing technology, the purpose of this invention is to provide a wire breakage detection system based on three-dimensional measurement.

[0008] To achieve the above objectives, the present invention provides the following technical solution: The three-dimensional measurement-based wire breakage detection system includes: a laser mechanism, a camera array, an encoder, a microcontroller, and a host computer.

[0009] The laser mechanism is used to emit a linear laser beam to irradiate the running filament assembly, creating a bright spot on each filament of the assembly.

[0010] The camera array includes at least one set of A cameras and / or at least one set of B cameras, the camera array being used to acquire bright spot images of the silk thread group.

[0011] The encoder is mounted on the main shaft motor of the textile equipment, and the encoder is used to generate continuous pulse signals as the main shaft motor rotates.

[0012] The microcontroller is connected to the encoder and is used to receive continuous pulse signals and generate image acquisition instructions according to a preset acquisition interval.

[0013] The host computer is communicatively connected to the microcontroller and each camera in the camera array. The host computer is used to receive image information collected by each camera in the camera array. By analyzing the presence, brightness and position changes of bright spots in the image, it determines whether there are broken threads and whether the three-dimensional position of the threads has changed abruptly, and outputs corresponding control signals to control the action of the actuator.

[0014] The present invention is further configured such that: the wire group includes wire one, wire two, and wire three arranged sequentially along the longitudinal direction of the optical axis of the laser mechanism; wire one, wire two, and wire three all form bright spots under the irradiation of the laser mechanism; and the brightness of the bright spots on wire one, wire two, and wire three decreases sequentially along the longitudinal direction as their distance from the laser mechanism increases.

[0015] The present invention is further configured such that each camera in the camera array is arranged at equal intervals along the fabric width direction to realize the three-dimensional position reconstruction of the yarn, and each camera in the camera array is an industrial camera.

[0016] The present invention is further configured such that: the angle between the light output axis of the laser mechanism and the horizontal reference line is F, the value range of the angle F is [-75°, +75°], and the horizontal distance between the laser mechanism and the first wire is D.

[0017] The horizontal distance between the optical center of the industrial camera lens and the first wire is C, the vertical distance from the optical center of the industrial camera lens to the optical axis of the laser mechanism is E, and the angle between the imaging optical axis of the industrial camera and the vertical reference line is G, with the value range of angle G being [10°, 75°].

[0018] The relative positions of the industrial camera with wires two and three are the same as the relative positions of the industrial camera with wire one.

[0019] The present invention is further configured such that: the host includes an image analysis module, which divides the brightness feature region of the bright spot in the image acquired by the camera array into detection block units according to a preset size; the image analysis module is used to extract the center point coordinates of the bright spot in the detection block unit, and reconstruct the three-dimensional position of the thread based on the center point coordinates and brightness value, thereby analyzing the operating status of the thread group and whether there is a broken thread.

[0020] The present invention is further configured such that: the image analysis module is also used to re-examine and analyze N consecutive images of the same thread, wherein N is a positive integer that can be set by the user, and the host outputs a thread breakage signal only when the N consecutive analysis results are all determined to be thread breakage.

[0021] The present invention is further configured such that: the yarn group passes through the comb of the textile equipment, and the breakage position of the yarn corresponds to the internal or external area of ​​the comb.

[0022] The present invention is further configured such that: the image analysis module of the host can distinguish between broken filament one, whose break point is located inside the comb, and broken filament two, whose break point is located outside the comb, based on the three-dimensional position change of the bright spot after the filament breaks.

[0023] The present invention is further configured such that: the image acquisition frequency of the camera array is synchronized with the rotation speed of the main shaft motor of the textile equipment; for each rotation of the main shaft motor, the comb completes one reciprocating motion; the encoder generates continuous pulse signals synchronously with the main shaft motor; and the microcontroller sends an acquisition command to the camera array to trigger image acquisition based on the received pulse signals and the preset acquisition interval.

[0024] In summary, this application includes at least one of the following beneficial technical effects: (1) The microcontroller communicates with the encoder and generates a trigger signal for the camera to acquire images. A bright spot is formed on each thread by a line laser. The three-dimensional spatial position of the thread is reconstructed by combining the image information acquired by the camera array. The running status of multiple rows of threads arranged in sequence along the longitudinal direction of the laser optical axis can be identified, reducing the possibility of missed detection caused by mutual occlusion of the front and rear rows of threads. At the same time, the present invention can identify the hidden broken threads inside the guide hole of the comb by the sudden change amplitude and offset direction of the three-dimensional position after the thread breaks. It realizes full-area coverage detection of the broken position inside and outside the comb, reducing the blind area of ​​hidden broken thread detection.

[0025] (2) By introducing a continuous N-times re-inspection mechanism, the broken yarn signal is only output when the same yarn is identified as broken in N consecutive image acquisitions. This can effectively filter out single misjudgments caused by environmental interference such as equipment vibration, yarn jumping, flying dust, and laser spot flicker in the textile workshop, reduce the probability of accidental shutdown, and improve the continuity of production.

[0026] (3) By adopting a non-contact detection scheme that combines laser irradiation and visual acquisition, the detection process does not require direct contact with the high-speed running threads, reducing secondary damage to the threads such as friction, stretching, and snagging. At the same time, there is no need to configure an independent detection unit for each thread. By arranging a camera array along the width of the fabric, the synchronous detection of the entire width of the threads can be achieved, which greatly simplifies the equipment installation structure.

[0027] (4) Based on the determination of whether there is a broken wire fault, the sudden change characteristics of the three-dimensional position of the bright spot can be used to distinguish whether the break point is inside or outside the comb, providing clear fault location information for on-site maintenance personnel. There is no need to check each wire after stopping the machine, which greatly shortens the fault handling and downtime maintenance time. At the same time, the present invention can identify early potential wire breakage problems such as wire deviation and abnormal tension through the change of the three-dimensional position of the wire, realize preventive maintenance of faults, reduce the occurrence of broken wire faults from the source, and further improve production efficiency and product qualification rate. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the overall structure of the three-dimensional measurement-based wire breakage detection system in this invention.

[0029] Figure 2 This is a schematic diagram of the positional distribution structure of the industrial camera and the wire in this invention.

[0030] Figure 3 This is a schematic diagram of the arrangement of the silk thread group in this invention.

[0031] Figure 4 This is a schematic diagram illustrating the interrupted wire situation of the present invention.

[0032] Figure 5 This is a flowchart of the interrupted wire detection system of the present invention.

[0033] Figure 6 This is a structural block diagram of the wire breakage detection system of the present invention.

[0034] Explanation of reference numerals in the attached diagram: 1. Laser mechanism; 2. Silk thread set; 21. Silk thread one; 22. Silk thread two; 23. Silk thread three; 3. Highlights; 4. Industrial camera; 5. Broken wire one; 6. Broken wire two; 7. Comb. Detailed Implementation

[0035] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0036] It should be noted that, unless otherwise specified, all technical and scientific terms used in this application have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0037] Please see Figures 1-6 The present invention provides the following technical solutions: Example 1, see Figure 6 A three-dimensional measurement-based wire breakage detection system includes: a laser mechanism 1, a camera array, an encoder, a microcontroller, and a host computer.

[0038] See Figure 1 The laser mechanism 1 is used to emit a linear laser to irradiate the running wire group 2, forming a bright spot 3 on each wire of the wire group 2.

[0039] Among them, the laser mechanism 1 emits a linear laser to illuminate the running silk thread group 2. Utilizing the reflective properties of the cylindrical surface of the silk thread, a bright spot 3 corresponding to the laser optical axis is formed on each silk thread, providing a unique feature for the spatial positioning of the silk thread.

[0040] See Figure 6 The camera array includes at least one set of A cameras and / or at least one set of B cameras, which are used to acquire images of the bright spot 3 of the silk thread group 2.

[0041] Each camera in the camera array is arranged at equal intervals along the width of the fabric to achieve three-dimensional position reconstruction of the yarn. Each camera in the camera array is an industrial camera.

[0042] In this embodiment, camera group A and camera group B are arranged opposite each other along the depth direction of the thread group 2. The shooting direction of camera group A is towards the first side of the thread group 2, and the shooting direction of camera group B is towards the second side of the thread group 2. The field of view of both camera group A and camera group B covers the entire width of the thread group 2. For the production scenario of single-sided fabric, only camera group A is needed to complete the image acquisition of the entire thread. For the production scenario of double-sided fabric, camera group A and camera group B are activated simultaneously to acquire images of both sides of the thread group 2 respectively, avoiding detection blind spots caused by the threads occluding each other.

[0043] Multiple industrial cameras 4 of the camera array simultaneously acquire images of the filament group 2 containing bright spots 3, providing multi-view visual basis data for three-dimensional position reconstruction.

[0044] The encoder is installed on the spindle motor of the textile equipment. The encoder is used to generate continuous pulse signals as the spindle motor rotates.

[0045] The encoder rotates synchronously with the spindle motor of the textile equipment, generating continuous pulse signals that correspond one-to-one with the spindle speed, thereby synchronizing the operation cycle of the detection system with that of the textile equipment.

[0046] The microcontroller communicates with the encoder. The microcontroller receives continuous pulse signals and generates image acquisition instructions according to the preset acquisition interval. This setting allows the acquisition timing to match the operating cycle of the textile equipment.

[0047] The host computer communicates with the microcontroller and each camera in the camera array. The host computer receives image information collected by each camera in the camera array. By analyzing the presence, brightness and position changes of bright spots 3 in the image, it determines whether there are broken threads and whether the three-dimensional position of the threads has changed abruptly, and outputs corresponding control signals to control the action of the actuator.

[0048] In this embodiment, the textile equipment includes all processing equipment that transforms yarn into fabric, including but not limited to double-needle bed warp knitting machines, KS warp knitting machines, or jacquard water-jet / air-jet looms. This system can be directly adapted to the installation space, operating cycle, and testing requirements of the above-mentioned equipment without requiring modification to the main structure of the equipment, thus improving the system's versatility.

[0049] In this embodiment, the actuator includes a main control shutdown unit and an audible and visual alarm unit for the textile equipment. The host is communicatively connected to the actuator. When a broken yarn is detected, the host outputs a corresponding control signal to trigger the audible and visual alarm unit to issue an alarm prompt. At the same time, it can send a shutdown signal to the main control shutdown unit of the textile equipment according to preset rules to control the equipment to stop.

[0050] The specific structure of thread group 2 is as follows: See Figure 3 The wire group 2 includes wire 1 21, wire 22, and wire 3 23 arranged sequentially along the longitudinal direction of the optical axis of the laser mechanism 1. Wire 1 21, wire 22, and wire 3 23 all form bright spots 3 under the illumination of the laser mechanism 1. The brightness of the bright spots 3 on wire 1 21, wire 22, and wire 3 23 decreases sequentially along the longitudinal direction as their distance from the laser mechanism 1 increases.

[0051] Multiple filaments are arranged sequentially along the longitudinal direction of the light output axis of the laser mechanism 1. When the linear laser irradiates along the optical axis, the laser energy decreases as the propagation distance increases. Under the premise that the filaments are of the same material and diameter, the filament closest to the laser mechanism 1 has the highest brightness (bright spot 3), while the filament furthest away (bright spot 3) has the lowest brightness, forming a brightness gradient that decreases sequentially along the longitudinal direction.

[0052] In this embodiment, a fitting formula for the brightness value of the bright spot 3 of the filament and the depth distance under different depth distances is established in advance through calibration experiments. That is, the brightness value is used as the depth distance coordinate. In actual detection, the extracted brightness value of the bright spot 3 is substituted into the fitting formula to obtain the depth direction coordinate corresponding to the filament. This serves as an auxiliary basis for the depth dimension of the three-dimensional position reconstruction. At the same time, it can quickly distinguish filaments at different depth positions and avoid confusion of the features of bright spot 3 of multiple filaments.

[0053] See Figure 2 The angle between the light output axis of laser mechanism 1 and the horizontal reference line is F, and the horizontal distance between laser mechanism 1 and wire 21 is D.

[0054] See Figure 2 The horizontal distance between the optical center of the lens of the industrial camera 4 and the filament 21 is C, the vertical distance from the optical center of the lens of the industrial camera 4 to the optical axis of the laser mechanism 1 is E, and the angle between the imaging optical axis of the industrial camera 4 and the vertical reference line is G.

[0055] The included angle F ranges from [-75°, +75°], and the included angle G ranges from [10°, 75°].

[0056] Based on the aforementioned angular distance constraints, the relative pose calibration of the laser mechanism 1 and the industrial camera 4 is completed.

[0057] The relative positions of industrial camera 4 with wire 22 and wire 3 are the same as the relative positions of industrial camera 4 with wire 1 21.

[0058] In the actual testing process, the industrial camera 4, laser mechanism 1, and wire are set based on the above constraint equations. At the same time, the relative position of laser mechanism 1 and industrial camera 4 can be calibrated by limiting the relative position and optical axis angle between laser mechanism 1 and wire, as well as the relative position and imaging optical axis angle between industrial camera 4 and wire. The coordinate transformation relationship between laser optical axis, camera imaging plane and wire spatial position is established, and the establishment of spatial coordinate system is further refined.

[0059] At the same time, the relative positions of the industrial camera 4 and the different depth lines are kept consistent, so that the correspondence between the position, brightness value and spatial coordinate of the bright spot 3 of the different depth lines is unified. There is no need to calibrate separately for different depth lines, which greatly improves the calculation efficiency and accuracy of three-dimensional position reconstruction and reduces the possibility of coordinate calculation errors caused by inconsistent relative positions.

[0060] The host includes an image analysis module. The image analysis module divides the brightness feature area of ​​bright spot 3 in the image acquired by the camera array into detection block units according to a preset size. The image analysis module is used to extract the center point coordinates of bright spot 3 in the detection block unit, and reconstruct the three-dimensional position of the thread based on the center point coordinates and brightness value, thereby analyzing the operating status of the thread group 2 and whether there are broken threads.

[0061] The image analysis module divides the brightness feature region of bright spots 3 in the image acquired by the camera array into detection block units according to a preset size. The preset size is set based on the yarn spacing of the fabric and the pixel equivalent of the industrial camera 4, so that each detection block unit corresponds to the imaging area of ​​a single yarn, avoiding bright spots 3 of multiple yarns entering the same detection unit. The image analysis module is used to extract the center point pixel coordinates of bright spots 3 in the detection block unit, and reconstruct the three-dimensional position of the yarn based on the center point coordinates and brightness value, thereby analyzing the operating status of yarn group 2 and whether there are broken yarns.

[0062] It should be noted that, based on the calibrated pose parameters, the world coordinate system equation of the laser plane and the pinhole camera model of the industrial camera 4 are established. The pixel coordinates of the center point of the bright spot 3 extracted from the image are converted into coordinates in the camera coordinate system through the camera intrinsic and extrinsic parameter matrix. Then, combined with the laser plane equation, they are converted into three-dimensional spatial coordinates in the world coordinate system to complete the three-dimensional position reconstruction of the silk thread.

[0063] In the three-dimensional spatial coordinates, the center pixel coordinates of bright spot 3 are used as the X and Z coordinates, and the depth distance formed by fitting the brightness value is used as the Y coordinate.

[0064] The image analysis module is also used to re-examine and analyze N consecutive images of the same thread, where N is a positive integer that can be set by the user. The host outputs a thread breakage signal only when all N consecutive analysis results indicate that the thread is broken.

[0065] The image analysis module incorporates a continuous re-inspection mechanism. Instead of directly outputting a broken yarn signal for a single acquired image, it re-inspects and analyzes N consecutive acquired images of the same yarn. Here, N is a user-configurable positive integer ranging from 3 to 20, which can be adjusted according to the spindle speed of the textile equipment. The higher the speed, the larger the value of N can be. The host only outputs a broken yarn signal when all N consecutive acquired images determine that the yarn is broken.

[0066] This mechanism can effectively filter out single misjudgments caused by factors such as vibration of textile equipment, yarn jumping, dust interference, and laser spot flicker, greatly improving the accuracy of yarn breakage detection and reducing the loss of production efficiency caused by accidental shutdown.

[0067] The yarn group 2 is installed through the comb 7 of the textile equipment, and the breakage position of the yarn corresponds to the internal or external area of ​​the comb 7.

[0068] See Figure 4 The host's image analysis module can distinguish between broken wire 1 (5) with the break point located inside the comb 7 and broken wire 2 (6) with the break point located outside the comb 7 based on the three-dimensional position change of the bright spot 3 after the wire breaks.

[0069] The yarn group 2 is set through the guide bar 7 of the textile equipment. The guide bar 7 is the core component of the textile equipment that controls the yarn feeding and guiding. During the operation of the yarn, the breakage location may occur in the internal area of ​​the guide bar 7, that is, the yarn breakage in the yarn guide hole, or it may occur in the external area of ​​the guide bar 7, that is, the yarn breakage outside the guide bar 7. By limiting the range of the yarn breakage location, the detection boundary of this system is defined, which can detect not only the visible broken yarn outside the guide bar 7, but also the hidden broken yarn inside the guide bar 7.

[0070] In this embodiment, the center coordinates of the yarn guide holes of the comb 7 are pre-set as the reference coordinates to define the coordinate boundary between the inside and outside of the comb 7.

[0071] When the thread is running normally, its three-dimensional position is a fixed standard path; when the thread breaks, the broken thread will shift its position due to the loss of tension, and the three-dimensional position of bright spot 3 will change abruptly.

[0072] Among them, the broken filament 5, which is located inside the comb 7, is limited by the yarn guide hole after the filament breaks and only slightly shifts to the outside of the comb 7. The three-dimensional position change amplitude of the bright spot 3 is less than the preset first threshold and the shift direction is fixed. In this case of broken filament, the broken filament will accumulate in the yarn conveying direction of the comb 7.

[0073] The break point is located on the broken filament 6 outside the comb 7. After the break, the filament loses its tension constraint and completely deviates from the standard path. The abrupt change in the three-dimensional position of the bright spot 3 is greater than the preset second threshold. In some cases, the bright spot 3 disappears because the filament deviates from the laser irradiation area. In other words, under this broken filament condition, the broken filament will not enter the comb 7. The second threshold is greater than the first threshold.

[0074] The image analysis module can accurately distinguish the breakage location of the broken wire by identifying the abrupt change amplitude and offset direction of the three-dimensional position of the bright spot 3, providing accurate information for subsequent fault location and repair, and shortening downtime for maintenance.

[0075] The image acquisition frequency of the camera array is synchronized with the rotation speed of the main spindle motor of the textile equipment. For every rotation of the main spindle motor, the comb 7 completes one reciprocating motion. The encoder generates continuous pulse signals synchronously with the main spindle motor. The microcontroller sends acquisition commands to the camera array to trigger image acquisition based on the received pulse signals and the preset acquisition interval.

[0076] The image acquisition frequency of the camera array is synchronized with the rotation speed of the main shaft motor of the textile equipment. For each rotation of the main shaft motor, the guide bar 7 completes one reciprocating motion, and the yarn completes one weaving cycle. The encoder rotates synchronously with the main shaft motor, generating continuous pulse signals that correspond one-to-one with the rotation angle of the main shaft. The microcontroller generates acquisition commands according to the preset acquisition interval, triggering the camera array to acquire images at fixed positions during the reciprocating motion of the guide bar 7. This synchronization mechanism ensures that each acquired image is an image of the yarn in the same spatial position, avoiding yarn position fluctuations caused by the movement of the guide bar 7, eliminating the impact of motion blur on image quality, and improving the feature extraction accuracy and 3D position reconstruction accuracy of Highlight 3.

[0077] Example 2, see reference Figure 5 This embodiment provides a broken wire detection method based on three-dimensional measurement. This method is implemented based on the broken wire detection system based on three-dimensional measurement described in Embodiment 1, and specifically includes the following steps: S1. System calibration and parameter preset.

[0078] The pose calibration of the laser mechanism 1 and the camera array is completed in advance, and the transformation relationship between the laser plane equation, the camera pinhole imaging model and the world coordinate system is established. The fitting formula of the brightness value of the yarn bright spot 3 and the depth distance is established through calibration test. The size of the detection block unit, the position change threshold for yarn breakage judgment, the number of consecutive re-inspections N, and the acquisition interval parameters are preset, and the center coordinate of the yarn guide hole of the comb 7 is set as the reference coordinate for distinguishing the yarn breakage position.

[0079] S2, Device synchronization and data acquisition triggering.

[0080] The encoder is installed on the main shaft motor of the textile equipment. The encoder rotates synchronously with the main shaft motor, generating continuous pulse signals that correspond one-to-one with the rotation angle of the main shaft and sending them to the microcontroller. The microcontroller receives the pulse signals, generates synchronous acquisition commands according to the preset acquisition interval, and sends them to the host and camera array, so that the image acquisition frequency of the camera array is synchronized with the rotation speed of the main shaft motor of the textile equipment, and triggers image acquisition at a fixed phase of the reciprocating motion of the comb 7.

[0081] S3. Synchronous acquisition of silk thread images.

[0082] The laser mechanism 1 continuously emits a line laser beam to illuminate the running wire group 2, forming a bright spot 3 on each wire of the wire group 2; each industrial camera 4 in the camera array synchronously acquires the image of the wire group 2 containing the bright spot 3 according to the synchronous acquisition command, and transmits the acquired image information to the host in real time.

[0083] S4. Image preprocessing and feature extraction.

[0084] The host's image analysis module receives image information, performs filtering and noise reduction preprocessing on the image to eliminate image noise caused by ambient light and dust interference; for the brightness feature area of ​​bright spot 3 in the preprocessed image, it divides the detection block unit according to the preset size, and extracts the center pixel coordinates and brightness value of bright spot 3 in each detection block unit.

[0085] S5. Reconstruction of the three-dimensional position of the silk thread.

[0086] The extracted center pixel coordinates of bright spot 3 are converted into planar coordinates in the world coordinate system through a pre-established coordinate transformation relationship. The extracted brightness value of bright spot 3 is substituted into the pre-established brightness and depth distance fitting formula to obtain the depth direction coordinates corresponding to the silk line. The planar coordinates and depth coordinates are combined to complete the three-dimensional spatial position reconstruction of the silk line.

[0087] S6. Initial assessment of wire breakage status.

[0088] The reconstructed real-time three-dimensional position of the thread is compared with the standard three-dimensional path of the thread during normal operation. At the same time, it is detected whether there is a valid bright spot 3 in the corresponding detection block unit. If the bright spot 3 at the corresponding position disappears, or the abrupt change in the three-dimensional position of the thread exceeds the preset threshold, the thread is initially judged to be in a broken state.

[0089] S7. Continuous re-inspection and result confirmation.

[0090] For a thread initially identified as broken, the image analysis module performs a re-inspection and analysis on N consecutive images of the thread. Only when the analysis results of the N consecutive images all indicate that the thread is broken, the host finally confirms that the thread is broken; otherwise, it is considered a false judgment and no broken thread signal is output.

[0091] S8, Broken wire location differentiation and control output.

[0092] For the thread that is finally confirmed to have broken, the image analysis module distinguishes between broken threads whose break point is located inside the comb 7 and broken threads whose break point is located outside the comb 7 based on the abrupt change amplitude and offset direction of the three-dimensional position of the bright spot 3 of the thread and the preset reference coordinates of the comb 7. The host outputs the corresponding control signal according to the broken thread determination result, and controls the actuator to complete the corresponding actions such as audible and visual alarm and equipment shutdown.

[0093] Obviously, the embodiments described above are merely some, not all, embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort should fall within the scope of protection of the present invention.

Claims

1. A broken wire detection system based on three-dimensional measurement, characterized in that: include: Includes laser mechanism (1), camera array, encoder, microcontroller and host; The laser mechanism (1) is used to emit a line laser to irradiate the running wire group (2), forming a bright spot (3) on each wire of the wire group (2). The camera array includes at least one set of A cameras and / or at least one set of B cameras, the camera array being used to acquire images of the bright spots (3) of the silk thread group (2); The encoder is installed on the main shaft motor of the textile equipment, and the encoder is used to generate continuous pulse signals as the main shaft motor rotates. The microcontroller is communicatively connected to the encoder. The microcontroller is used to receive continuous pulse signals and generate image acquisition instructions according to a preset acquisition interval. The host is connected to the microcontroller and each camera in the camera array. The host is used to receive image information collected by each camera in the camera array. By analyzing the presence, brightness and position changes of bright spots (3) in the image, it can determine whether there are broken threads and whether the three-dimensional position of the threads has changed abruptly, and output the corresponding control signal to control the action of the actuator.

2. The three-dimensional measurement-based broken wire detection system according to claim 1, characterized in that: The wire group (2) includes wire one (21), wire two (22), and wire three (23) arranged sequentially along the optical axis of the laser mechanism (1). Wire one (21), wire two (22), and wire three (23) all form bright spots (3) under the illumination of the laser mechanism (1). The brightness of the bright spots (3) on wire one (21), wire two (22), and wire three (23) decreases sequentially along the depth direction as their distance from the laser mechanism (1) increases.

3. The three-dimensional measurement-based broken wire detection system according to claim 2, characterized in that: Each camera in the camera array is arranged at equal intervals along the fabric width direction to achieve three-dimensional position reconstruction of the yarn. Each camera in the camera array is an industrial camera (4).

4. The three-dimensional measurement-based broken wire detection system according to claim 3, characterized in that: The angle between the light output axis of the laser mechanism (1) and the horizontal reference line is F, and the value range of the angle F is [-75°, +75°]. The horizontal distance between the laser mechanism (1) and the first wire (21) is D. The horizontal distance between the optical center of the lens of the industrial camera (4) and the first filament (21) is C, the vertical distance from the optical center of the lens of the industrial camera (4) to the optical axis of the laser mechanism (1) is E, and the angle between the imaging optical axis of the industrial camera (4) and the vertical reference line is G. The value range of the angle G is [10°, 75°]. The relative positions of the industrial camera (4) with the second (22) and the third (23) of the wire are the same as the relative positions of the industrial camera (4) with the first (21) of the wire.

5. The three-dimensional measurement-based broken wire detection system according to claim 1, characterized in that: The host includes an image analysis module. The image analysis module divides the brightness feature area of ​​the bright spot (3) in the image acquired by the camera array into detection block units according to a preset size. The image analysis module is used to extract the center point coordinates of the bright spot (3) in the detection block unit, and reconstruct the three-dimensional position of the thread based on the center point coordinates and brightness value, and then analyze the running status of the thread group (2) and whether there is a broken thread.

6. The three-dimensional measurement-based broken wire detection system according to claim 1, characterized in that: The image analysis module is also used to re-examine and analyze N consecutive images of the same thread, where N is a positive integer that can be set by the user. The host outputs a thread breakage signal only when all N consecutive analysis results indicate that the thread is broken.

7. The three-dimensional measurement-based broken wire detection system according to claim 1, characterized in that: The yarn group (2) is set through the comb (7) of the textile equipment, and the breakage position of the yarn corresponds to the internal or external area of ​​the comb (7).

8. The three-dimensional measurement-based broken wire detection system according to claim 5, characterized in that: The image analysis module of the host can distinguish between broken wire 1 (5) with the break point located inside the comb (7) and broken wire 2 (6) with the break point located outside the comb (7) based on the three-dimensional position change of the bright spot (3) after the wire breaks.

9. The three-dimensional measurement-based broken wire detection system according to claim 1, characterized in that: The image acquisition frequency of the camera array is synchronized with the rotation speed of the spindle motor of the textile equipment. For each rotation of the spindle motor, the comb (7) completes one reciprocating motion. The encoder generates a continuous pulse signal synchronously with the spindle motor. The microcontroller sends an acquisition command to the camera array to trigger image acquisition based on the received pulse signal and the preset acquisition interval.

Images