Polyester filament broken filament detection system and method based on image processing

Abstract

The invention discloses a polyester filament broken filament detection system and method based on image processing, and relates to the technical field of polyester detection.The polyester filament broken filament detection system comprises a machine shell, an image processing device is arranged at the top of an inner cavity of the machine shell, and a tension buffering mechanism is arranged on one side of the machine shell and used for automatically conducting micro-amplitude displacement along with instantaneous tension changes of tows; high-frequency vibration is absorbed; a broken filament carding mechanism is arranged on the bottom wall of the machine shell and located on one side of the tension buffering mechanism. Tows enter the broken filament carding mechanism after being subjected to tension buffering, a suction fan forms local negative pressure airflow through an air suction groove and an air suction hole in an independent channel limited by adjacent partition pieces, broken filaments floating on the surfaces of the tows are actively adsorbed to brush pieces, the brush pieces are gradually arranged from high to low in height in the running direction, and the broken filaments are separated from the broken filaments. A flexible multi-stage contact structure is formed, on the premise that main tows are not damaged, impurity filaments, floating hair and fibers prone to breakage are removed step by step, and the phenomenon that the impurity filaments, the floating hair and the fibers are wound or disturbed by other tows in the follow-up process is avoided.

Description

Technical Field

[0001] This invention relates to the technical field of polyester detection, and in particular to a polyester filament fiber detection system and method based on image processing. Background Technology

[0002] For example, patent CN104685347A, entitled "A Method for Inspecting Threads," describes a method for inspecting threads during the manufacturing process of a thread that travels continuously along its length. This method involves photographing the thread as it travels and rapidly inspecting its entire length based on the obtained image data. The method obtains information about thread defects without misidentifying instances of thread swaying or oblique movement as defects. The method is characterized by using an imaging unit to photograph the thread as it travels and using a data processing unit to process the image data obtained by the imaging unit through the following processes: (a) calculating multiple thread widths within a predetermined interval along the direction of travel based on the image data of the thread as it travels; (b) calculating thread width deviations based on the multiple thread widths; and (c) comparing the thread width deviations with a first threshold.

[0003] For example, the patent with publication number CN116380943B, entitled "A Non-destructive Testing Method and Application for Off-axis Angle of Composite Material", is the first invention in the prior art to observe the distribution position of fiber bundles inside composite material through micro-nano CT. Through geometric magnification technology and sub-voxel measurement technology, high-resolution (micrometer level) three-dimensional imaging of fiber bundles in composite material is performed. High-precision detection of the distribution of fiber bundles inside composite material can be achieved without damaging the composite material. Furthermore, the off-axis angle of fiber bundles is calculated through image comparison processing.

[0004] However, during the high-speed production of polyester filament, the slight vibration or acceleration changes of the filament bundle cause the ends of the filaments to appear as diffuse light spots in the image, which are easily misjudged as reflections. At the same time, the polyester main filament has strong specular reflection characteristics, forming bright areas, while the fine filaments, due to their rough surface, exhibit weak reflection or shadows, resulting in severely uneven grayscale distribution in the image. Furthermore, in multiple parallel filament bundles, the filaments produced by a single filament are easily affected by airflow disturbances and overlap with adjacent filament bundles, forming cross-filament connection structures, making it difficult to accurately determine the defect attribution. Therefore, this application provides a polyester filament filament detection system and method based on image processing to meet the requirements. Summary of the Invention

[0005] The purpose of this application is to provide a polyester filament fiber detection system and method based on image processing, which can effectively solve the problems mentioned in the background art.

[0006] To achieve the above objectives, this application provides the following technical solution: a polyester filament fuzz detection system based on image processing, comprising a housing, an image processing device disposed at the top of the inner cavity of the housing, and a tension buffer mechanism disposed on one side of the housing, for automatically performing micro-displacement according to the instantaneous tension change of the filament bundle, so as to absorb high-frequency vibration; A hair combing mechanism is provided on the bottom wall of the housing and on one side of the tension buffer mechanism for combing the hair bundles and threads. A flow stabilizing mechanism is provided inside the housing and on one side of the hair combing mechanism to force the hair bundle to run along a fixed trajectory and suppress lateral swaying. The bottom of the housing, directly below the image processing device, is provided with a multi-filament adsorption mechanism to suppress the diffusion of filaments and the mutual interference effect between multiple filaments. A leveling mechanism is provided inside the housing and on one side of the multi-filament adsorption mechanism to prevent the filaments on the surface of the filaments from scattering and affecting the image acquisition and processing of the image processing device.

[0007] The tension buffer mechanism includes a first support roller, which is rotatably installed inside the housing. The outer surface of the first support roller is fitted with a plurality of inner cylinders that are evenly spaced. The inner walls of the plurality of inner cylinders are provided with a plurality of disc springs that are arranged in a ring array, and the disc springs are fixedly installed on the outer surface of the first support roller.

[0008] The outer surface of the inner cylinder is provided with a drive cylinder, and the outer surface of the drive cylinder is covered with a rubber sleeve.

[0009] The wool combing mechanism includes a bottom shell, which is fixedly installed on the bottom of the inner wall of the machine housing. A suction fan is provided on one side of the bottom of the bottom shell, and an air suction groove is provided at the upper end of the bottom shell.

[0010] The bottom of the inner wall of the bottom shell is provided with several partitions that are evenly distributed. A bottom plate is provided at the bottom of the inner wall of the bottom shell and between two partitions. Several air suction holes are opened at the upper end of the bottom plate. Several brush plates with heights arranged in descending order are provided at the upper end of the bottom plate.

[0011] The flow stabilizing mechanism includes a second support roller, which is rotatably installed inside the housing. The outer surface of the second support roller is provided with a plurality of guide plates that are evenly distributed.

[0012] The multi-filament adsorption mechanism includes a mounting frame, which is fixedly installed on the bottom of the inner wall of the housing. A supplementary light is provided on the bottom of the inner wall of the mounting frame, and a filament adsorption plate is provided on the upper end of the mounting frame.

[0013] The surface of the hair adsorption plate is arc-shaped, and the hair adsorption plate is located directly below the image processing device.

[0014] The leveling mechanism includes a third support roller, which is rotatably installed inside the housing. A roller is fitted on the outer surface of the third support roller. A blower is provided on both sides of the roller. Several air suction holes are opened on the outer surface of the roller. A breathable film is fitted on the outer surface of the roller.

[0015] This invention also provides a method for detecting polyester filament fibers based on image processing: Step 1: The polyester filament bundle is wound around the surface of the tension buffer mechanism and conveyed along it. During the conveying process, the filament bundle vibrates due to its operation. The tension buffer mechanism automatically makes a slight displacement according to the instantaneous tension change of the filament bundle, thereby absorbing the vibration force and realizing dynamic tension buffering. Step 2: After being adjusted by the tension buffer mechanism, the filament bundle enters the filament combing mechanism. The filament combing mechanism captures the filaments floating on the surface of the filament bundle through negative pressure adsorption and combs the filaments in a directional manner. At the same time, it removes easily broken fibers and impurities, thus achieving preliminary purification of the filament bundle surface. Subsequently, the filament bundle enters the flow stabilization mechanism, which forces the filament bundle to run along a fixed trajectory, effectively suppressing its lateral sway and improving operational stability. The tension buffer mechanism 2 dynamically absorbs high-frequency vibrations to prevent single filament breakage due to sudden tension changes. Then, the filament combing mechanism 4 uses negative pressure and directional brushes to remove floating debris and easily broken fibers in advance. Finally, the flow stabilizing mechanism 6 forces the filament bundle to run along a fixed trajectory, effectively suppressing lateral sway. Together, the three constitute a pretreatment chain of vibration reduction, debris removal and flow stabilization. Step 3: After being regulated by the flow stabilizing mechanism, the filament bundle passes over the surface of the multi-filament bundle adsorption mechanism. The multi-filament bundle adsorption mechanism uses local negative pressure to adsorb the filaments remaining on the surface of the filament bundle to its surface, which on the one hand inhibits the diffusion of filaments, and on the other hand prevents interference between adjacent filament bundles. At this time, the image processing device located at the top of the inner cavity of the housing performs imaging detection on the filaments passing through the surface of the multi-filament adsorption mechanism, while the multi-filament adsorption mechanism serves as a high-contrast background to assist the image processing device in accurately identifying filament defects. Finally, the filament bundle passes through a leveling mechanism, which adsorbs any remaining filaments in the tested area, suppressing filament vibration while controlling filament dispersion.

[0016] In summary, the technical effects and advantages of this invention are as follows: 1. In the process of high-speed conveying of filament bundle, when the traction force or surface tension fluctuates instantaneously, the drive cylinder moves axially under force, compressing the disc spring and generating elastic deformation. The disc spring then dynamically adjusts the position of the drive cylinder through its reversible compression and rebound, thereby achieving real-time compensation for the tension of the filament bundle, effectively absorbing high-frequency vibration, and avoiding single filament breakage or fuzz formation due to sudden tension changes.

[0017] 2. In this invention, the filament bundle enters the filament combing mechanism after being buffered by tension. In the independent channel defined by adjacent partitions, the suction fan forms a local negative pressure airflow through the suction groove and suction hole, which actively adsorbs the filaments floating on the surface of the filament bundle onto the brush plate. The brush plate is arranged gradually from high to low along the running direction to form a flexible multi-level contact structure. Without damaging the main filament bundle, it removes impurities, floating hairs and easily broken fibers step by step, so as to prevent them from entangled or interfering with other filament bundles in subsequent processes.

[0018] 3. In this invention, the filament bundle, after being stabilized, runs closely against the filament adsorption plate. The supplementary light in the mounting frame evenly illuminates the filament adsorption plate, creating a high-contrast, low-reflection optical background. This significantly improves the imaging clarity and recognition accuracy of the image processing device for micron-sized filaments and prevents filaments from crossing the filament path due to airflow disturbance. At the same time, the static electricity generated by the friction between the filament bundle and the filament adsorption plate further enhances the adsorption force on the filaments, effectively suppressing them from drifting away or interfering with adjacent filament paths during high-speed detection. After detection, the filament bundle enters the leveling mechanism and is wound around the surface of the breathable membrane. The third support roller drives the roller to move the membrane accordingly, ensuring smooth traction. The exhaust fan establishes negative pressure inside the roller and generates continuous adsorption force on the surface of the breathable membrane through the air intake holes, firmly fixing any residual filaments that may be stirred up again after detection. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A first-view stereoscopic structural diagram of an image-processing-based polyester filament fiber detection system. Figure 2 A second-view stereoscopic structural diagram of an image-processing-based polyester filament fiber detection system. Figure 3 A third-view stereoscopic structural diagram of an image-processing-based polyester filament fiber detection system. Figure 4 This is a schematic diagram of the three-dimensional connection structure of the tension buffer mechanism; Figure 5 This is a schematic diagram of a partial three-dimensional connection structure of the tension buffer mechanism; Figure 6 This is a three-dimensional sectional view of the connection structure of the tension buffer mechanism; Figure 7 A first-person perspective three-dimensional connection structure diagram of the wool combing mechanism; Figure 8 A second-view, three-dimensional connection structure diagram of the wool combing mechanism; Figure 9 This is a schematic diagram of a partial three-dimensional connection structure of a wool combing mechanism; Figure 10 This is a schematic diagram of the three-dimensional connection structure between the brush plate and the substrate; Figure 11 A schematic diagram of the three-dimensional connection structure of the flow stabilization mechanism; Figure 12 A three-dimensional sectional view of the connection structure of the flow stabilizing mechanism; Figure 13 This is a schematic diagram of the three-dimensional connection structure of the multi-filament adsorption mechanism; Figure 14 A three-dimensional cross-sectional view of the connection structure of the multi-filament adsorption mechanism; Figure 15 A schematic diagram of the three-dimensional connection structure of the leveling mechanism; Figure 16 This is a sectional view of the three-dimensional connection structure of the leveling mechanism.

[0021] In the diagram: 1. Housing; 2. Tension buffer mechanism; 21. First support roller; 22. Drive cylinder; 23. Rubber sleeve; 24. Inner cylinder; 25. Disc spring; 4. Wool combing mechanism; 41. Bottom shell; 42. Fan; 43. Partition plate; 44. Brush plate; 45. Suction groove; 46. Base plate; 47. Suction hole; 5. Image processing device; 6. Flow stabilizing mechanism; 61. Second support roller; 62. Guide plate; 7. Multi-filament adsorption mechanism; 71. Mounting frame; 72. Wool adsorption plate; 73. Supplemental light; 8. Leveling mechanism; 81. Third support roller; 82. Breathable membrane; 83. Fan; 84. Roller; 85. Suction hole. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] Example 1, Reference Figures 1 to 16The image processing-based polyester filament fuzz detection system shown includes a housing 1, an image processing device 5 is provided at the top of the inner cavity of the housing 1, and a tension buffer mechanism 2 is provided on one side of the housing 1, which is used to automatically make micro-displacements according to the instantaneous tension changes of the filament bundle in order to absorb high-frequency vibrations. A hair combing mechanism 4 is provided on the bottom wall of the housing 1 and on one side of the tension buffer mechanism 2 for combing the hair bundles and threads. Inside the housing 1 and on one side of the wool combing mechanism 4, there is a flow stabilizing mechanism 6, which is used to force the wool bundle to run along a fixed trajectory and suppress lateral swaying. A multi-filament adsorption mechanism 7 is provided at the bottom of the housing 1 and directly below the image processing device 5 to suppress the diffusion of filaments and the mutual interference effect between multiple filaments. A leveling mechanism 8 is provided inside the housing 1 and on one side of the multi-filament adsorption mechanism 7 to prevent the filaments on the surface of the filaments from scattering and affecting the image acquisition and processing of the image processing device 5.

[0024] It is worth noting that the polyester filament bundle is wound around the surface of the tension buffer mechanism 2 and transported along it. During the transport process, the filament bundle vibrates due to its operation. The tension buffer mechanism 2 automatically makes a slight displacement according to the instantaneous tension change of the filament bundle, thereby absorbing the vibration force and realizing dynamic tension buffering.

[0025] After being adjusted by the tension buffer mechanism 2, the filament bundle enters the filament combing mechanism 4. The filament combing mechanism 4 captures the filaments floating on the surface of the filament bundle through negative pressure adsorption and combs the filaments in a directional manner. At the same time, it removes easily broken fibers and impurities, thus achieving preliminary purification of the filament bundle surface. Subsequently, the filament bundle enters the flow stabilizing mechanism 6, which forces the filament bundle to run along a fixed trajectory, effectively suppressing its lateral oscillation and improving operational stability.

[0026] Among them, the tension buffer mechanism 2 dynamically absorbs high-frequency vibrations to avoid single filament breakage due to sudden tension changes. Then, the filament combing mechanism 4 uses negative pressure and directional brushes to remove floating impurities and easily broken fibers in advance. Finally, the flow stabilization mechanism 6 forces the filament bundle to run along a fixed trajectory, effectively suppressing lateral sway. The three together constitute a pretreatment chain for vibration reduction, impurity removal and flow stabilization.

[0027] After being regulated by the flow stabilizing mechanism 6, the filament bundle passes over the surface of the multi-filament bundle adsorption mechanism 7. The multi-filament bundle adsorption mechanism 7 uses local negative pressure to adsorb the residual filaments on the surface of the filament bundle to its surface, which on the one hand inhibits the diffusion of filaments, and on the other hand prevents interference between adjacent filament bundles. At this time, the image processing device 5 located at the top of the inner cavity of the housing performs imaging detection on the filaments passing through the surface of the multi-filament adsorption mechanism 7, while the multi-filament adsorption mechanism 7 serves as a high-contrast background to assist the image processing device 5 in accurately identifying filament defects. Finally, the filament bundle passes through the leveling mechanism 8, which adsorbs the residual filaments in the detected area, suppressing the vibration of the filament bundle while controlling the filaments from scattering.

[0028] Example 2: This example provides a further technical solution for the tension buffer mechanism 2.

[0029] The tension buffer mechanism 2 includes a first support roller 21, which is rotatably installed inside the housing 1. The outer surface of the first support roller 21 is fitted with a plurality of inner cylinders 24 that are evenly distributed. The inner walls of the plurality of inner cylinders 24 are provided with a plurality of disc springs 25 that are arranged in a ring array, and the disc springs 25 are fixedly installed on the outer surface of the first support roller 21.

[0030] The outer surface of the inner cylinder 24 is provided with a drive cylinder 22, and the outer surface of the drive cylinder 22 is covered with a rubber sleeve 23.

[0031] It is worth noting that the housing 1 is equipped with a drive device, which drives the first support roller 21 to rotate. The first support roller 21 is connected to the inner cylinder 24 through the disc spring 25 and drives the inner cylinder 24 to rotate synchronously. The inner cylinder 24 further drives the drive cylinder 22 to rotate. The polyester filaments are wound around the outer surface of the rubber sleeve 23 and are transported to the inside of the wool combing mechanism 4 as the drive cylinder 22 rotates. During the transport process, due to the fluctuation of the filament traction force and surface tension, the drive cylinder 22 will be subjected to the filament traction force and thus squeeze the inner cylinder 24, causing the inner cylinder 24 to compress the disc spring 25 and produce elastic deformation. The disc spring 25 dynamically adjusts the axial position of the drive cylinder 22 through compression and rebound, thereby compensating for the filament tension change in real time and suppressing the high-frequency vibration generated during the traction process. The rubber sleeve 23 is made of rubber material with a high coefficient of friction to provide stable traction friction for the filaments and prevent slippage.

[0032] During the high-speed transport of the filament bundle, when the traction force or surface tension fluctuates instantaneously, the drive cylinder 22 moves axially under force, compressing the disc spring 25 and causing elastic deformation. The disc spring 25 then dynamically adjusts the position of the drive cylinder 22 through its reversible compression and rebound, thereby achieving real-time compensation for the tension of the filament bundle, effectively absorbing high-frequency vibration, and avoiding single filament breakage or fuzz formation due to sudden tension changes.

[0033] Example 3: This example provides a further technical solution for the hair combing mechanism 4 and the flow stabilizing mechanism 6.

[0034] The wool combing mechanism 4 includes a bottom shell 41, which is fixedly installed on the bottom of the inner wall of the housing 1. A suction fan 42 is provided on one side of the bottom of the bottom shell 41, and an air suction groove 45 is provided at the upper end of the bottom shell 41.

[0035] The bottom of the inner wall of the bottom shell 41 is provided with several partitions 43 that are evenly distributed. The bottom of the inner wall of the bottom shell 41 and located between two partitions 43 is provided with a bottom plate 46. Several air suction holes 47 are opened at the upper end of the bottom plate 46. Several brush plates 44 arranged in descending order of height are provided at the upper end of the bottom plate 46.

[0036] It is worth noting that the filament bundles pulled by the tension buffer mechanism 2 are guided into the filament combing mechanism 4, specifically into the channel between every two adjacent partitions 43. The suction fan 42 connects to each suction hole 47 through the suction groove 45, and forms a negative pressure airflow in the area between every two partitions 43. The negative pressure airflow adsorbs the filaments floating on the surface of the filament bundles onto the surface of the brush plate 44. The brush plate 44 is arranged in order of height from high to low along the direction of filament bundle movement, forming a progressive contact structure. During the continuous traction and movement of the filament bundle, the brush plate 44 combs the surface of the filament bundle in multiple stages, removing impurities and easily broken fibers.

[0037] The flow stabilizing mechanism 6 includes a second support roller 61, which is rotatably installed inside the housing 1. The outer surface of the second support roller 61 is provided with a plurality of guide plates 62 that are evenly distributed.

[0038] The filaments combed by the brush 44 are wound around the surface of the second support roller 61. The guide plate 62, which is located near the second support roller 61, is used to separate the filaments and force them to run along a preset fixed trajectory, thereby suppressing the lateral sway of the filaments during the conveying process.

[0039] In this process, the filament bundle enters the filament combing mechanism 4 after being buffered by tension. In the independent channel defined by the adjacent partitions 43, the suction fan 42 forms a local negative pressure airflow through the suction groove 45 and the suction hole 47, which actively adsorbs the filaments floating on the surface of the filament bundle onto the brush plate 44. The brush plate 44 is arranged gradually from high to low along the running direction to form a flexible multi-level contact structure. Without damaging the main filament bundle, it removes the impurities, floating hair and easily broken fibers step by step, so as to prevent them from entangled or interfering with other filament bundles in the subsequent process.

[0040] Subsequently, the filament bundle is wound around the surface of the second support roller 61 and physically separated and constrained by the adjacent guide plate 62, forcing it to run stably along the preset path, effectively suppressing lateral sway and filament deviation. This not only reduces the source of filament interference from the source, but also improves the trajectory accuracy and spatial stability of the filament bundle, creating clean and orderly physical conditions for subsequent high-precision image detection.

[0041] Example 4: This example provides further technical solutions for the multi-filament adsorption mechanism 7 and the leveling mechanism 8.

[0042] The multi-filament adsorption mechanism 7 includes a mounting frame 71, which is fixedly installed on the bottom of the inner wall of the housing 1. A supplementary light 73 is provided on the bottom of the inner wall of the mounting frame 71, and a filament adsorption plate 72 is provided on the upper end of the mounting frame 71.

[0043] The surface of the hair adsorption plate 72 is arc-shaped, and the hair adsorption plate 72 is located directly below the image processing device 5.

[0044] It is worth noting that the filaments guided by the guide plate 62 adhere closely to the surface of the filament adsorption plate 72 and move along its surface under the traction. During this process, the filaments remaining on the surface of the filaments are also pressed onto the filament adsorption plate 72 and move synchronously with it. The mounting frame 71 is equipped with a supplementary light 73, which emits light that evenly illuminates the filament adsorption plate 72, making it a high-contrast background light source to assist the image processing device 5 located on the top of the housing in clearly imaging and accurately identifying the filaments and filaments. In addition, as the filament bundle moves continuously under traction on the surface of the filament adsorption plate 72, static electricity accumulates on the surface of the filament adsorption plate 72 due to friction. This static electricity effect further enhances the adsorption capacity of the filaments and effectively prevents the filaments from detaching from the surface of the filament bundle and drifting away.

[0045] The leveling mechanism 8 includes a third support roller 81, which is rotatably installed inside the housing 1. A roller 84 is fitted on the outer surface of the third support roller 81. A blower 83 is provided on both sides of the roller 84. Several air suction holes 85 are opened on the outer surface of the roller 84. A breathable film 82 is fitted on the outer surface of the roller 84.

[0046] The filament bundle detected by the image processing device 5 is wound around the surface of the breathable membrane 82. When the third support roller 81 rotates, it drives the roller 84 to rotate synchronously, thereby driving the breathable membrane 82 to move accordingly and assisting the filament bundle to be pulled smoothly. The exhaust fan 83 draws air from the inside of the roller 84 to form a negative pressure in its inner cavity. This negative pressure is conducted to the outside through the air suction hole 85 on the peripheral wall of the roller 84, causing the surface of the breathable membrane 82 to produce an adsorption effect. During operation, the residual filaments on the surface of the filament bundle are adsorbed and fixed by the negative pressure through the breathable membrane 82, thereby preventing the filaments from scattering.

[0047] The filament bundle, after being stabilized, runs closely against the filament adsorption plate 72. The supplementary light 73 in the mounting frame 71 illuminates the filament adsorption plate 72 evenly, making it form a high-contrast, low-reflection optical background, which significantly improves the imaging clarity and recognition accuracy of the image processing device 5 for micron-level filaments, and also prevents the filaments from crossing the filament path due to airflow disturbance. Meanwhile, the static electricity generated by the friction between the filament bundle and the filament adsorption plate 72 further enhances the adsorption force on the filaments, effectively suppressing them from drifting away or interfering with adjacent filament paths during high-speed detection. After the detection is completed, the filament bundle enters the leveling mechanism 8 and is wound around the surface of the breathable membrane 82. The third support roller 81 drives the roller 84 to move the membrane accordingly, ensuring smooth traction. The exhaust fan 83 establishes negative pressure inside the roller 84 and generates a continuous adsorption force on the surface of the breathable membrane 82 through the air suction hole 85, firmly fixing any residual filaments that may be lifted again after the detection.

[0048] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A polyester filament fuzz detection system based on image processing, comprising a housing (1), wherein an image processing device (5) is disposed at the top of the inner cavity of the housing (1), characterized in that: A tension buffer mechanism (2) is provided on one side of the housing (1) to automatically make a small displacement according to the instantaneous tension change of the filament bundle in order to absorb high frequency vibration; A hair combing mechanism (4) is provided on the bottom wall of the housing (1) and on one side of the tension buffer mechanism (2) for combing the hair bundles and threads; A flow stabilizing mechanism (6) is provided inside the housing (1) and on one side of the hair combing mechanism (4) to force the hair bundle to run along a fixed trajectory and suppress lateral swaying. A multi-filament adsorption mechanism (7) is provided at the bottom of the housing (1) and directly below the image processing device (5) to suppress the diffusion of filaments and the mutual interference effect between multiple filaments. A leveling mechanism (8) is provided inside the housing (1) and on one side of the multi-filament adsorption mechanism (7) to prevent the filaments on the surface of the filaments from drifting and affecting the image acquisition and processing of the image processing device (5).

2. The polyester filament fiber detection system based on image processing according to claim 1, characterized in that: The tension buffer mechanism (2) includes a first support roller (21), which is rotatably installed inside the housing (1). The outer surface of the first support roller (21) is fitted with a plurality of inner cylinders (24) that are evenly distributed. The inner walls of the plurality of inner cylinders (24) are provided with a plurality of disc springs (25) that are arranged in a ring array, and the disc springs (25) are fixedly installed on the outer surface of the first support roller (21).

3. The polyester filament fiber detection system based on image processing according to claim 2, characterized in that: The outer surface of the inner cylinder (24) is provided with a drive cylinder (22), and the outer surface of the drive cylinder (22) is covered with a rubber sleeve (23).

4. The polyester filament fiber detection system based on image processing according to claim 1, characterized in that: The hair combing mechanism (4) includes a bottom shell (41), which is fixedly installed on the bottom of the inner wall of the housing (1). A suction fan (42) is provided on one side of the bottom of the bottom shell (41), and an air suction groove (45) is provided at the upper end of the bottom shell (41).

5. The polyester filament fiber detection system based on image processing according to claim 4, characterized in that: The bottom of the inner wall of the bottom shell (41) is provided with a number of partitions (43) that are evenly distributed. The bottom of the inner wall of the bottom shell (41) and between two partitions (43) is provided with a bottom plate (46). The upper end of the bottom plate (46) is provided with a number of air suction holes (47). The upper end of the bottom plate (46) is provided with a number of brush plates (44) arranged in descending order of height.

6. The polyester filament fiber detection system based on image processing according to claim 1, characterized in that: The flow stabilizing mechanism (6) includes a second support roller (61), which is rotatably installed inside the housing (1). The outer surface of the second support roller (61) is provided with a plurality of guide plates (62) that are evenly distributed.

7. The polyester filament fiber detection system based on image processing according to claim 1, characterized in that: The multi-filament adsorption mechanism (7) includes a mounting frame (71), which is fixedly installed on the bottom of the inner wall of the housing (1). A supplementary light (73) is provided on the bottom of the inner wall of the mounting frame (71), and a filament adsorption plate (72) is provided on the upper end of the mounting frame (71).

8. The polyester filament fiber detection system based on image processing according to claim 7, characterized in that: The surface of the hair adsorption plate (72) is arc-shaped, and the hair adsorption plate (72) is located directly below the image processing device (5).

9. The polyester filament fiber detection system based on image processing according to claim 1, characterized in that: The leveling mechanism (8) includes a third support roller (81), which is rotatably installed inside the housing (1). A roller (84) is fitted on the outer surface of the third support roller (81). A blower (83) is provided on both sides of the roller (84). Several air suction holes (85) are opened on the outer surface of the roller (84). A breathable film (82) is fitted on the outer surface of the roller (84).

10. A method for detecting polyester filament fuzz based on image processing, employing the polyester filament fuzz detection system based on image processing as described in any one of claims 1-9, characterized in that: Step 1: The polyester filament bundle is wound around the surface of the tension buffer mechanism (2) and transported along it. During the transport process, the filament bundle vibrates due to its operation. The tension buffer mechanism (2) automatically makes a slight displacement according to the instantaneous tension change of the filament bundle, thereby absorbing the vibration force and realizing dynamic tension buffering. Step 2: After being adjusted by the tension buffer mechanism (2), the filament bundle enters the filament combing mechanism (4). The filament combing mechanism (4) captures the filaments floating on the surface of the filament bundle through negative pressure adsorption and combs the filaments in a directional manner. At the same time, it removes easily broken fibers and impurities, thus achieving preliminary purification of the filament bundle surface. Subsequently, the filament bundle enters the flow stabilizing mechanism (6), which forces the filament bundle to run along a fixed trajectory, effectively suppressing its lateral sway and improving operational stability; Step 3: After being regulated by the flow stabilizing mechanism (6), the filament bundle passes over the surface of the multi-filament bundle adsorption mechanism (7). The multi-filament bundle adsorption mechanism (7) adsorbs the residual filaments on the surface of the filament bundle to its surface through local negative pressure, which on the one hand inhibits the diffusion of filaments, and on the other hand prevents interference between adjacent filament bundles. At this time, the image processing device (5) located at the top of the inner cavity of the housing performs imaging detection on the filaments passing through the surface of the multi-filament adsorption mechanism (7), while the multi-filament adsorption mechanism (7) serves as a high-contrast background to assist the image processing device (5) in accurately identifying filament defects. Finally, the filament bundle passes through the leveling mechanism (8), which adsorbs the residual filaments in the detected area, controlling the filaments from drifting while suppressing the vibration of the filament bundle.

Images