Fabric sticky hair identification method and identification system based on fluorescence enhancement and machine vision

Abstract

The invention relates to the technical field of textile detection, and provides a fabric sticky hair identification method and identification system based on fluorescence enhancement and machine vision, and the fabric sticky hair identification method comprises the following steps: carrying out fluorescent dyeing on a fleece when the fleece is dyed; rubbing the fuzzing cloth with a to-be-detected fabric so as to enable the to-be-detected fabric to be fuzzed; placing the to-be-detected fabric in a preset environment so as to enable the sticky hair on the to-be-detected fabric to emit fluorescence; acquiring a surface image of the to-be-detected fabric after hair sticking and fluorescent color development; analyzing the area proportion of a fluorescent region in the surface image based on image color identification; and according to the area proportion of the fluorescent region, determining the hair sticking grade of the fabric to be detected. According to the invention, the problem that the precision and reliability of sticky hair detection are seriously influenced by material and color differences of a machine vision detection system in the prior art can be solved.

Description

Technical Field

[0001] The present invention relates to the technical field of textile detection, and particularly to a method and a system for identifying fabric lint sticking based on fluorescence enhancement and machine vision. Background Art

[0002] The adhesion of fibers to the fabric, as the most important index for evaluating the anti-lint sticking effect of the fabric, currently lacks relevant testing instruments and characterization methods for detecting the lint sticking performance of the fabric. At present, the lint sticking performance of textiles is mostly detected manually, that is, by manually rubbing two fabrics against each other to simulate the occurrence process of fabric lint sticking, and then the number of lint sticking on the fabric is identified by the naked eye, and then the lint sticking performance of the fabric is evaluated.

[0003] However, the manual evaluation of lint sticking is limited by the color of the fabric itself. The lint sticking on dark fabrics is easy to appear and is easy to be identified by the naked eye; the lint sticking on light fabrics is not easy to appear and is prone to misjudgment. Moreover, the manual testing efficiency is low and quantitative judgment cannot be made. Due to the large number of lint sticking roots, manual counting takes too long and quantitative judgment cannot be achieved.

[0004] In addition, there is also a lint sticking recognition system based on machine vision in the prior art, that is, the number of lint sticking (fiber hairiness) on the fabric is accurately recognized by image recognition, and then the anti-lint sticking performance of the fabric is judged. However, for fabrics of different materials and different colors, the image representations obtained for machine vision recognition after lint sticking are uneven, which leads to deviations in the recognition of the number of lint sticking; at the same time, due to the different colors, lengths and bending shapes of the lint sticking (fiber hairiness) on the fabric, it is also difficult to accurately recognize the number of lint sticking by machine vision, and it is also easy to cause deviations in the recognition of the number of lint sticking.

[0005] When detecting the lint sticking of fabrics of different colors, the existing image recognition methods can only achieve relatively good recognition effects when the color difference between the lint sticking (fiber hairiness) and the fabric to be tested is relatively significant. For example, in the detection of lint sticking on black fabrics, a certain accuracy rate can be achieved. For fabrics of red, green and blue colors, the error of lint sticking recognition will increase. The main reason is that the differences in hue, contrast, etc. between these colors and the common lint sticking (fiber hairiness) colors are not as obvious as those of black fabrics, which makes it difficult for the detection model to accurately capture the lint sticking (fiber hairiness), resulting in frequent misjudgments and missed judgments, seriously affecting the accuracy and reliability of lint sticking detection and unable to meet the actual application requirements in the lint sticking detection scenarios of fabrics of various colors.

[0006] Most of the existing machine vision models for detecting hairiness directly identify yarns or fabrics. It is difficult to identify fabrics with similar colors, hairiness, and fine fibers on the fabrics, which greatly affects the detection accuracy. Determining the hairiness detection result based on the preset threshold and mean value also cannot accurately quantify and evaluate the result. Summary of the Invention

[0007] The purpose of the present invention is to provide a method and a system for identifying adhered hair on fabrics based on fluorescence enhancement and machine vision, so as to solve the problem that the machine vision detection system in the prior art seriously affects the accuracy and reliability of adhered hair detection due to material and color differences.

[0008] The above technical purpose of the present invention is mainly achieved through the following technical solutions:

[0009] On the one hand, the present invention provides a method for identifying adhered hair on fabrics based on fluorescence enhancement and machine vision, which includes the following steps:

[0010] Perform fluorescence dyeing on the raising cloth when dyeing the raising cloth;

[0011] Friction the raising cloth and the fabric to be tested to make the fabric to be tested adhere to hair;

[0012] Place the fabric to be tested in a preset environment to make the adhered hair on the fabric to be tested emit fluorescence;

[0013] Obtain the surface image of the fabric to be tested with adhered hair and fluorescence color development;

[0014] Based on image color recognition, analyze the area ratio of the fluorescence region in the surface image;

[0015] Determine the adhered hair grade of the fabric to be tested according to the area ratio of the fluorescence region.

[0016] In a preferred embodiment of the present invention, performing fluorescence dyeing on the raising cloth when dyeing the raising cloth includes adding a fluorescent brightener when dyeing the raising cloth.

[0017] In a preferred embodiment of the present invention, friction the raising cloth and the fabric to be tested includes:

[0018] Install both the raising cloth and the fabric to be tested on the hair adhesion generating device;

[0019] Start the hair adhesion generating device to drive the raising cloth to move relative to the fabric to be tested, so that the raising cloth and the fabric to be tested come into contact and friction to adhere to hair.

[0020] In a preferred embodiment of the present invention, placing the fabric to be tested in a preset environment includes:

[0021] Place the fabric to be tested after lint sticking into a completely dark light box;

[0022] Turn on the ultraviolet lamp in the completely dark light box to irradiate the fabric to be tested so that the lint sticking fluoresces and shows color.

[0023] In a preferred embodiment of the present invention, obtaining the surface image of the fabric to be tested after lint sticking and fluorescence color development includes:

[0024] Turn on the imaging device in the completely dark light box to collect an image of the fabric to be tested after fluorescence color development to obtain the surface image.

[0025] In a preferred embodiment of the present invention, the resolution of the captured image of the imaging device is 1 million pixels to 100 million pixels; and / or, the distance between the imaging device and the fabric to be tested is 10 cm to 50 cm.

[0026] In a preferred embodiment of the present invention, determining the lint sticking grade of the fabric to be tested according to the area ratio of the fluorescent region includes:

[0027] The lower the area ratio of the fluorescent region, the higher the anti-lint sticking grade of the fabric to be tested and the better the anti-lint sticking performance; the higher the area ratio of the fluorescent region, the lower the anti-lint sticking grade of the fabric to be tested and the worse the anti-lint sticking performance.

[0028] On the other hand, the present invention also provides a fabric lint sticking recognition system based on fluorescence enhancement and machine vision, which is used to implement the fabric lint sticking recognition method as described above. The fabric lint sticking recognition system has a lint sticking generating device and a lint sticking detecting device;

[0029] The lint sticking generating device includes:

[0030] An experimental area for installing the fabric to be tested;

[0031] Two groups of parallel slide rails, and the two groups of slide rails are respectively arranged on both sides of the experimental area;

[0032] A roller for installing the lint raising cloth, and both ends of the roller are slidably connected to the two groups of slide rails through slide tables respectively to drive the lint raising cloth to move relative to the fabric to be tested so as to generate contact friction between the lint raising cloth and the fabric to be tested;

[0033] The lint sticking detecting device includes:

[0034] A completely dark light box, a detection area for placing the fabric to be tested is arranged in the completely dark light box, and an ultraviolet lamp and an imaging device are arranged in the completely dark light box;

[0035] A computer processing unit, which is electrically connected to the photographing device, and a machine vision detection model is built in the computer processing unit.

[0036] In a preferred embodiment of the present invention, a lifting mechanism is provided on the sliding table, both ends of the roller are connected to the lifting mechanism, and the lifting mechanism can adjust the distance between the raising cloth on the roller and the fabric to be tested on the experimental area.

[0037] In a preferred embodiment of the present invention, the roller includes a central shaft and a drum rotatably sleeved on the central shaft, and the raising cloth is mounted on the drum.

[0038] In a preferred embodiment of the present invention, the lifting mechanism is an adjusting bolt, both ends of the central shaft are threadedly connected to the adjusting bolt, and the rotation of the adjusting bolt can drive the lifting of the central shaft.

[0039] In a preferred embodiment of the present invention, the hair sticking generating device further includes:

[0040] A synchronous belt arranged parallel to the slide rail, the synchronous belt is rotatably sleeved at both ends of the slide rail, and the sliding table is fixedly connected to the synchronous belt;

[0041] A driving motor is arranged at one end of the slide rail, and the driving motor is drivingly connected to the synchronous belt to drive the synchronous belt to rotate.

[0042] In a preferred embodiment of the present invention, the hair sticking generating device further includes:

[0043] A lead screw arranged parallel to the slide rail, both ends of the lead screw are respectively rotatably connected to both ends of the slide rail, and the sliding table is threadedly sleeved on the lead screw;

[0044] A driving motor, the driving motor is arranged at one end of the slide rail, and the driving motor is drivingly connected to the lead screw.

[0045] Compared with the prior art, the technical solution of the present invention has the following characteristics and advantages:

[0046] 1. Utilize fluorescence to enhance contrast: Use a raising cloth with fluorescence characteristics. Through the excitation of a UV lamp, the fluorescent substances on the fiber hairs are excited to emit fluorescence with a specific wavelength, significantly enhancing the visual contrast between the fiber hairs and the fabric background, and effectively solving the problem that it is difficult to distinguish fiber hairs from the background for fabrics of different colors, especially fabrics with low color contrast with fiber hairs (such as red, green, blue, etc.).

[0047] 2. Wide range of applications: Precisely control the rolling speed and weight of the roller in the fluff generating device, and flexibly adjust according to the material and characteristics of the fabric to be tested, so that the fiber fluff on the raising cloth adheres to the fabric to be tested evenly and controllably. It can be used for fabrics of different materials and thicknesses, and the test range can be extended to fields such as clothing and home furnishings.

[0048] 3. Precise quantitative evaluation of results: According to the proportion of the fluorescent area obtained by analyzing the machine vision detection model in the image of the fabric to be tested, combined with a large amount of experimental data and industry standards, set the ratio value of the fluorescent area with clear boundaries and grade divisions to achieve an intuitive and accurate determination of the fluff adhesion condition of the fabric, providing a scientific, reliable and highly guiding basis for multiple links in the textile and clothing industries.

[0049] Through the combination of fluff generation, fluorescence excitation, image acquisition and model analysis, the present invention effectively solves the problem of fluff detection for fabrics of different colors, and provides an efficient and accurate technical means for fabric quality detection in the textile and clothing industries. Brief Description of the Drawings

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following described drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts. In the drawings:

[0051] The drawings described herein are only for illustrative purposes and are not intended to limit the scope of the present disclosure in any way. In addition, the shapes and proportional dimensions of the components in the drawings are only schematic and are used to assist in understanding the present invention, and do not specifically limit the shapes and proportional dimensions of the components of the present invention. Those skilled in the art can select various possible shapes and proportional dimensions according to the specific situation to implement the present invention under the teaching of the present invention.

[0052] Figure 1 It is a flowchart of the method for identifying fabric fluff adhesion based on fluorescence enhancement and machine vision according to the present invention;

[0053] Figure 2 It is a schematic structural diagram of the fluff generating device according to the present invention;

[0054] Figure 3 It is a schematic structural diagram of another embodiment of the fluff generating device according to the present invention;

[0055] Figure 4 It is a schematic structural diagram of the fluff detection device according to the present invention;

[0056] Figure 5Schematic diagram of fluorescence lint sticking for the first fabric to be tested;

[0057] Figure 6 For Figure 5 Schematic diagram of the recognition result of fluorescence lint sticking;

[0058] Figure 7 Schematic diagram of fluorescence lint sticking for the second fabric to be tested;

[0059] Figure 8 For Figure 7 Schematic diagram of the recognition result of fluorescence lint sticking;

[0060] Figure 9 Schematic diagram of fluorescence lint sticking for the third fabric to be tested;

[0061] Figure 10 For Figure 9 Schematic diagram of the recognition result of fluorescence lint sticking;

[0062] Figure 11 Schematic diagram of non-fluorescent lint sticking for the fabric to be tested;

[0063] Figure 12 For Figure 11 Schematic diagram of the recognition result of non-fluorescent lint sticking in

[0064] Figure 13 Schematic diagram of non-fluorescent lint sticking for another fabric to be tested;

[0065] Figure 14 For Figure 13 Schematic diagram of the recognition result of non-fluorescent lint sticking in

[0066] Figure 15 Schematic diagram of non-fluorescent lint sticking for yet another fabric to be tested;

[0067] Figure 16 For Figure 15 Schematic diagram of the recognition result of non-fluorescent lint sticking in

[0068] Explanation of reference numerals:

[0069] 10. Workbench; 11. Experimental area; 12. Fixing mechanism;

[0070] 20. Slide rail; 21. Floor feet; 22. Slide table; 221. Adjusting bolt; 23. Tailstock; 24. Driving seat; 241. Connecting shaft; 25. Driving motor; 26. Synchronous belt; 27. Lead screw; 28. Limit switch; 29. Inductive switch;

[0071] 30. Roller; 31. Central shaft;

[0072] 40. Full black light box; 41. Ultraviolet lamp; 42. Photographing device; 43. Detection area; 44. Computer processing unit. Detailed implementation manners

[0073] In order to enable those skilled in the art of this technology to better understand the technical solutions in the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present invention.

[0074] It should be noted that when an element is referred to as being "disposed on" another element, it can be directly on the other element or there may also be an intermediate element. When an element is considered to be "connected" to another element, it can be directly connected to the other element or there may be an intermediate element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are only for the purpose of illustration and do not represent the only embodiments.

[0075] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which this invention belongs. The terms used herein in the description of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0076] Embodiment 1:

[0077] As Figure 1 shown, the present invention provides a method for identifying lint sticking on fabrics based on fluorescence enhancement and machine vision, which includes the following steps:

[0078] Step S1: Fluorescently dye the raised fabric when dyeing the raised fabric;

[0079] Step S2: Rub the raised fabric with the fabric to be tested to make the fabric to be tested stick lint;

[0080] Step S3: Place the fabric to be tested in a preset environment so that the lint sticking on the fabric to be tested emits fluorescence;

[0081] Step S4: Obtain the surface image of the fabric to be tested after sticking lint and fluorescing;

[0082] Step S5: Analyze the area ratio of the fluorescent region in the surface image based on image color recognition;

[0083] Step S6: Determine the lint sticking level of the fabric to be tested according to the area ratio of the fluorescent region.

[0084] The fabric fluff adhesion recognition method of the present invention uses fluorescence technology to enhance the contrast between the fabric to be tested and the fluff (fiber hairiness) on it. A napping cloth with fluorescence characteristics is used to perform napping on the fabric to be tested, and the fluorescent substances on its fiber hairiness are excited by an ultraviolet lamp 41, enhancing the visual contrast between the fiber hairiness and the fabric background, which can effectively solve the problem that it is difficult to distinguish fiber hairiness from the background for fabrics of different colors, especially fabrics with low color contrast with fiber hairiness (such as red, green, blue, etc.).

[0085] At the same time, combined with the proportion of the obtained fluorescent area in the image of the fabric to be tested analyzed by the machine vision detection model, and combined with a large amount of experimental data and industry standards, a clear boundary and a graded fluorescent area occupancy ratio are set to achieve an intuitive and accurate determination of the fabric fluff adhesion status, providing a scientific, reliable and highly guiding basis for multiple links in the textile and clothing industries.

[0086] The following will detail the specific operation methods of each step in the fabric fluff adhesion recognition method based on fluorescence enhancement and machine vision of the present invention.

[0087] In step S1, fluorescent dyeing of the napping cloth is carried out. The napping cloth refers to the cloth used to rub against the fabric to be tested to simulate the fluff adhesion phenomenon of the fabric to be tested during actual use. This cloth can make the fiber hairiness (or called fluff) on it adhere to the fabric to be tested through friction. The operation of fluorescent dyeing can be achieved by adding a fluorescent brightener when dyeing the napping cloth, so that the napping cloth has stable and high-intensity fluorescence characteristics, and the fiber hairiness on the napping cloth has good adhesion and fluorescence effect.

[0088] In step S2, fluff adhesion operation is carried out on the fabric to be tested. The fabric to be tested refers to the fabric that needs to evaluate the anti-fluff adhesion performance, that is, to obtain the adhesion of fiber hairiness to the fabric to be tested. The fluff adhesion operation is mainly achieved through the friction operation between the above-mentioned napping cloth and the fabric to be tested; during the friction process, the fiber hairiness with fluorescent substances in the napping cloth can partially adhere to the fabric to be tested. The fluff adhesion operation can directly use manual hands to rub the above two fabrics; in order to achieve a precise evaluation of the test results, the friction fluff adhesion operation of the above two fabrics can also be realized through a separate fluff adhesion generating device.

[0089] If the lint sticking operation is performed using a lint sticking device, first, both the napping cloth and the fabric to be tested are installed on the lint sticking device. Then, the lint sticking device is started to drive the napping cloth to move relative to the fabric to be tested, so that contact friction occurs between the napping cloth and the fabric to be tested to stick lint. Since the lint sticking device is usually a mechanical structure, it can accurately control the friction force and the number of friction times between the above two fabrics, and thus the same lint sticking operation (lint sticking scenario) can be generated when testing different fabrics. Since the lint sticking performance of different fabrics is different, different lint sticking effects (the number of adsorbed fiber fluff is different) can be presented under the same lint sticking scenario, which is convenient for comparing and evaluating the anti-lint sticking performance of different fabrics. Among them, the structure of the lint sticking device will be described in detail below.

[0090] In step S3, the fiber fluff on the fabric to be tested after lint sticking is made to fluoresce. After the lint sticking operation of the fabric to be tested is completed, the fabric to be tested is transferred into the fully black light box 40. The light in the fully black light box 40 is relatively dim, which is convenient for the fluorescence of the fiber fluff. An ultraviolet lamp 41 is provided in the fully black light box 40. After the fabric to be tested is placed, the ultraviolet lamp 41 is turned on to emit ultraviolet rays towards the fabric to be tested. The wavelength and intensity of the ultraviolet rays emitted by the ultraviolet lamp 41 are strictly and precisely adjusted. The ultraviolet rays it emits can accurately excite the fluorescent substances in the fiber fluff, making it emit fluorescence of a specific wavelength (such as 365nm, 385nm, 395nm, 405nm), significantly enhancing the visual contrast between the fiber fluff and the background of the fabric to be tested.

[0091] In step S4, the surface image after fluorescence is obtained. A photographing device 42 is also provided in the fully black light box 40, usually a high-resolution camera. After the fluorescence of the fabric to be tested is stable, the photographing device 42 is turned on to collect an image of the fabric to be tested, and the surface image after fluorescence is obtained. The range photographed by the photographing device 42 can be the entire fabric to be tested or a partial area of the fabric to be tested, but it is necessary to ensure that this partial area can reflect the general characteristics of the fiber fluff adsorbed on the fabric to be tested, and the test results should not be accidental due to a small photographing area.

[0092] Preferably, parameters such as the resolution and focal length of the photographing device 42 in step S4 are set in advance and calibrated regularly. Generally, the resolution of the image photographed by the photographing device 42 is 1 million pixels - 100 million pixels; the distance between the photographing device 42 and the fabric to be tested is 10 cm - 50 cm. If the resolution of the image photographed by the photographing device 42 is too low, it will lead to a large error in the subsequent recognition process. If the resolution is too high, it will reduce the efficiency in the subsequent color recognition process. The limitation of the distance between the photographing device 42 and the fabric to be tested can ensure the clarity of the image.

[0093] In step S5, the proportion of the area of the fluorescent region in the surface image is detected. The captured surface image is accurately input into a machine vision detection model that has been trained and optimized with a large number of known linty fabric image samples in advance. This model can analyze the proportion value of the fluorescent region in the fabric image (area of the fluorescent region in the surface image / area of the fabric to be measured in the surface image) based on the obvious color difference between the background color and the fluorescent color development.

[0094] In a specific embodiment, the acquired surface image can be recognized and analyzed based on the HALCON machine vision software:

[0095] In a feasible embodiment, first, the emphasize operator in halcon is used to enhance the captured surface image, and the parameters maskwidth and maskheight are set to 7, and factor is set to 2. Then, the enhanced surface image is recognized using the trained yolo object detection model, and the recognized result image and the hair proportion are saved in a folder: the threshold operator in halcon is used to perform threshold segmentation on the enhanced surface image, the parameters mingray is set to 120, maxgray is set to 255, the hair region is segmented out, then the reduce_domain operator is used to extract the hair region, and then the area_center operator is used to obtain the area of the original image and the area of the hair region respectively. Finally, the percent operator is used to obtain the ratio of the hair region to the original image, that is, the hair proportion, and visualization processing is performed to display the data in the upper left corner of the surface image.

[0096] In another feasible embodiment, first, the emphasize operator in Halcon is used to enhance the captured surface image, and the parameters maskwidth and maskheight are set to 7, and factor is set to 2. The threshold operator in Halcon is used to perform threshold segmentation on the result image, and the parameters mingray is set to 120 and maxgray is set to 255 to segment the hair area; the obtained hair area is separately operated using the connection operator; the select_shape operator is used. First, the features are set to area, and the areas with sizes of 10 - 500 are selected. Then the features are set to contlength, and the areas with sizes of 0 - 300 are selected. Finally, the features are set to rectangularity, and the areas with values of 0 - 0.8 are selected. The count_obj operator is used to calculate the number of adhered hairs obtained; the obtained areas are merged using the union operator; the reduce_domain operator is used to cut out the hair area, the area_center operator is used to separately obtain the area of the original image and the area of the hair area, and the percent operator is used to obtain the ratio of the hair area to the original image, that is, the hair proportion. In this way, the number of hairs and the hair proportion are obtained, and visualization processing is performed to display the data in the upper left corner of the surface image.

[0097] Step S6: Determine the anti-adhesion grade of the fabric to be tested according to the area ratio of the fluorescent area. The lower the area ratio of the fluorescent area, the higher the anti-adhesion grade of the fabric to be tested, and the better the anti-adhesion performance; the higher the area ratio of the fluorescent area, the lower the anti-adhesion grade of the fabric to be tested, and the worse the anti-adhesion performance. The corresponding table of the anti-adhesion grade and the fluorescent area ratio is as follows:

[0098] Anti-fuzzing level Proportion of fluorescent area 5 (Very good anti-fuzzing performance) <0.3％ 4 (Good anti-fuzzing performance) 0.3～0.7％ 3 (Average anti-fuzzing performance) 0.7～1.1％ 2 (Poor anti-fuzzing performance) 1.1％～1.5％ 1 (No anti-fuzzing performance) >1.5％

[0099] Embodiment 2:

[0100] Such as Figure 2 and Figure 3As shown in the figure, the present invention also provides a fabric lint recognition system based on fluorescence enhancement and machine vision, which is used to implement the fabric lint recognition method described in Embodiment 1. The fabric lint recognition system has a lint generation device and a lint detection device; the lint generation device includes a workbench 10, two groups of parallel slide rails 20, and rollers 30 for installing the lint-raising cloth. There is an experimental area 11 for installing the fabric to be tested on the workbench 10. The two groups of slide rails 20 are respectively arranged on both sides of the experimental area 11. The two ends of the roller 30 are respectively slidably connected to the two groups of slide rails 20 through slide blocks 22; the lint detection device includes a completely black light box 40 and a computer processing unit 44. There is a detection area 43 for placing the fabric to be tested in the completely black light box 40. An ultraviolet lamp 41 and a photographing device 42 are also arranged in the completely black light box 40; the computer processing unit 44 is electrically connected to the photographing device 42, and the computer processing unit 44 is built-in with a machine vision detection model.

[0101] The fabric lint recognition system described in the present invention can be used to implement the fabric lint recognition method described in Embodiment 1, and accurately determine the lint condition on the fabric based on fluorescence recognition. The corresponding technical effects have been described above and will not be elaborated here.

[0102] In addition, the fabric lint recognition system described in the present invention can accurately control the rolling speed and weight of the roller 30 in the lint generation device, and flexibly adjust according to the material and characteristics of the fabric to be tested, so that the fiber fluff on the lint-raising cloth adheres to the fabric to be tested evenly and controllably. It can be used for fabrics of different materials and different thicknesses, and the test range can be extended to fields such as clothing and home furnishings.

[0103] The specific structures of each part of the fabric lint recognition system described in the present invention, as well as the positional and cooperative relationships between each part, will be described in detail below.

[0104] The fabric lint recognition system described in the present invention includes two major parts: a lint generation device and a lint detection device; among them, the lint generation device is used to perform linting operations on the fabric to be tested, and the lint detection device is used to detect the lint condition on the fabric to be tested after linting, so as to obtain the lint performance of the fabric to be tested.

[0105] As Figure 2 and Figure 3 shown, the lint generation device includes a workbench 10. The workbench 10 is a horizontal working surface, and an experimental area 11 is arranged on the workbench 10. The experimental area 11 is used to install the fabric to be tested. In order to fix the fabric to be tested, a fixing mechanism 12 is arranged at the edge position of the experimental area 11. The fixing mechanism 12 can be an openable and closable elastic pressing strip.

[0106] On both sides of the experimental area 11, a set of slide rails 20 are respectively arranged. The two sets of slide rails 20 are arranged in parallel. The bottom of the slide rails 20 is installed and fixed on the tabletop of the workbench 10 through the floor feet 21. A slide table 22 is slidably arranged on each slide rail 20, and the slide table 22 can slide along the slide rail 20. Slide rails 20 with different lengths and sizes can be selected according to requirements. Preferably, the length of the slide rail 20 is less than 2m, and the distance between the two sets of slide rails 20 is less than 100cm; different specifications of slide tables 22 can also be selected according to different experimental conditions. Preferably, the lengths of the slide tables 22 are 104mm, 130mm, and 200mm, and the widths are 90mm, 110mm, and 180mm.

[0107] A roller 30 is connected between the two slide tables 22. The extending direction of the roller 30 is perpendicular to the extending direction of the slide rail 20, and the roller 30 is located above the experimental area 11. The roller 30 is used to install the lint-raising cloth. The lint-raising cloth has a large amount of sticky lint. The lint-raising operation on the fabric to be tested can be realized through the friction between the lint-raising cloth and the fabric to be tested. In order to fix the lint-raising cloth, a tearable sticky lint paper is arranged on the outer circumference of the roller 30, and the lint-raising cloth can be fixed on the roller 30 by bonding.

[0108] When the lint-raising device is working, the roller 30 slides on the slide rail 20 along with the slide table 22, so that friction is generated between the lint-raising cloth on the roller 30 and the fabric to be tested on the experimental area 11, and then the lint-raising operation on the fabric to be tested is realized.

[0109] As Figure 4 shown, the lint-raising detection device mainly includes a completely black light box 40 for image acquisition and a computer processing unit 44 for image processing. A detection area 43 is arranged in the completely black light box 40, and the fabric to be tested after the lint-raising operation can be placed in the detection area 43. An ultraviolet lamp 41 is arranged in the completely black light box 40. The ultraviolet lamp 41 is usually installed directly below the fabric to be tested, and it is necessary to ensure that there is no other light interference in the completely black light box 40 except the ultraviolet lamp 41. The ultraviolet lamp 41 can emit ultraviolet rays towards the fabric to be tested. The ultraviolet rays can accurately excite the fluorescent substances in the fiber hairs on the fabric to be tested, making them emit fluorescence with a specific wavelength, thereby enhancing the visual contrast between the fiber hairs and the background of the fabric to be tested. A photographing device 42 is also arranged in the completely black light box 40. The photographing device 42 is usually a high-resolution camera, and the photographing device 42 is used to collect images of the fabric to be tested after fluorescence color development.

[0110] As Figure 4As shown, the computer processing unit 44 is electrically connected to the imaging device 42. The surface image obtained by the imaging device 42 can be transmitted through a wire into the computer processing unit 44. The computer processing unit 44 has a machine vision detection model built in. This model can analyze the ratio of the fluorescent area in the fabric image (area of the fluorescent area in the surface image / area of the fabric to be measured in the surface image) based on the obvious color difference between the background color and the fluorescent color development, and then determine the lint adhesion level of the fabric to be measured according to the area ratio of the fluorescent area.

[0111] The following will further describe the structure and technical effects of the preferred embodiment of the fabric lint adhesion recognition system of the present invention.

[0112] According to an embodiment of the present invention, as Figure 2 and Figure 3 shown, a lifting mechanism is provided on the sliding table 22. Both ends of the roller 30 are connected to the lifting mechanism, and the lifting mechanism can adjust the distance between the lint-raising cloth on the roller 30 and the fabric to be measured in the experimental area 11.

[0113] The lifting mechanism can adjust the height of the roller 30 relative to the workbench 10, and then adjust the pressing degree between the lint-raising cloth and the fabric to be measured, changing the frictional force between the two, that is, the lint adhesion force can be adjusted according to actual test requirements.

[0114] Specifically, in this embodiment, the lifting mechanism is an adjusting bolt 221. Both ends of the roller 30 are threadedly connected to the adjusting bolt 221. The rotation of the adjusting bolt 221 can drive the roller 30 to lift and lower; of course, the lifting mechanism can also adopt other methods, such as using an electric push rod to control the lifting and lowering of the roller 30. The specific structure of the lifting mechanism is not limited here.

[0115] According to an embodiment of the present invention, as Figure 3 shown, the roller 30 includes a central shaft 31 and a drum rotatably sleeved on the central shaft 31. The lint-raising cloth is installed on the drum.

[0116] The roller 30 adopts a rotatable structure. During the friction lint adhesion process, the lint-raising cloth can rotate with the drum, that is, the lint-raising cloth rolls on the upper surface of the fabric to be measured along with the drum, thereby improving the lint adhesion effect on the fabric to be measured.

[0117] Specifically, in this embodiment, both ends of the central shaft 31 are respectively connected to the two sliding tables 22. The drum is rotatably sleeved on the central shaft 31 through bearings. The size and weight of the roller 30 (drum) can be replaced according to different experimental conditions. Preferably, the diameter of the roller 30 is 2.5 cm - 20 cm, the length is 10 cm - 100 cm, and the weight is 0.1 kg - 10 kg. The material of the roller 30 (drum) can be selected from stainless steel, carbon steel, nylon or polyurethane.

[0118] According to an embodiment of the present invention, as Figure 2 shown, the slide table 22 slides on the slide rail 20 and is driven by a synchronous belt 26. The hair sticking device further includes a synchronous belt 26 arranged parallel to the slide rail 20 and a driving motor 25. The synchronous belt 26 is rotatably sleeved at both ends of the slide rail 20, and the slide table 22 is fixedly connected to the synchronous belt 26; the driving motor 25 is arranged at one end of the slide rail 20, and the driving motor 25 is drivingly connected to the synchronous belt 26 to drive the synchronous belt 26 to rotate.

[0119] Specifically, as Figure 2 shown, a tailstock 23 is arranged at one end of the slide rail 20, and a driving seat 24 is arranged at the other end of the slide rail 20. Rotors are arranged in both the tailstock 23 and the driving seat 24, and the synchronous belt 26 is connected between the two rotors. The rotation of the rotors can drive the synchronous belt 26 to rotate. The slide table 22 is fixed at a fixed position on the synchronous belt 26 through screws. The reciprocating rotation of the synchronous belt 26 can drive the slide table 22 to reciprocate on the slide rail 20. A driving motor 25 is arranged on one side of the driving seat 24, and the output shaft of the driving motor 25 is drivingly connected to the rotor in the driving seat 24. The rotation of the driving motor 25 can drive the rotor in the driving seat 24 to rotate. The rotor in the driving seat 24 serves as the driving wheel, and the rotor in the tailstock 23 serves as the driven wheel. The synchronous belt 26 rotates under the action of the driving wheel and the driven wheel.

[0120] Preferably, as Figure 2 shown, the synchronous belts 26 on the two groups of slide rails 20 share a driving motor 25. The driving seats 24 on the two groups of slide rails 20 are arranged on the same side, and the rotors in the two driving seats 24 are connected by a connecting shaft 241. The driving motor 25 is arranged on one of the driving seats 24 and is connected to the rotor in the driving seat 24. Under the action of the connecting shaft 241, the driving motor 25 can drive the rotors in the two driving seats 24 to rotate synchronously, thereby driving the two synchronous belts 26 to rotate synchronously, so as to ensure the stable movement of the roller 30. The driving motor 25 can be a stepping motor, and the stepping motor can accurately position the position of the slide table 22 and improve the system control accuracy.

[0121] According to an embodiment of the present invention, as Figure 3 shown, the slide table 22 slides on the slide rail 20 and is driven by a ball screw. The hair sticking device further includes a lead screw 27 arranged parallel to the slide rail 20 and a driving motor 25; both ends of the lead screw 27 are rotatably connected to both ends of the slide rail 20, and the slide table 22 is threadedly sleeved on the lead screw 27; the driving motor 25 is arranged at one end of the slide rail 20, and the driving motor 25 is drivingly connected to the lead screw 27 to drive the two lead screws 27 to rotate.

[0122] Specifically, as Figure 3As shown, the lead screw 27 is arranged above the slide rail 20. A tailstock 23 is provided at one end of the slide rail 20, and a drive seat 24 is provided at the other end of the slide rail 20. Both ends of the lead screw 27 are rotatably connected to the tailstock 23 and the drive seat 24 through bearings. A slide table 22 slidably arranged on the slide rail 20 is threadedly sleeved on the lead screw 27. The rotation of the lead screw 27 can drive the slide table 22 to move along the slide rail 20. A drive motor 25 is provided on one side of the drive seat 24. The output shaft of the drive motor 25 is connected to the lead screw 27. The rotation of the drive motor 25 can drive the lead screw 27 to rotate, thereby driving the slide table 22 to slide. The drive motor 25 can be selected as a stepper motor, and the stepper motor can accurately position the position of the slide table 22 to improve the system control accuracy.

[0123] According to an embodiment of the present invention, as Figure 2 and Figure 3 shown, a limit switch 28 is installed on the side of the slide rail 20, and an induction switch 29 for sensing the position of the limit switch 28 is provided on the side of the slide table 22. The cooperation between the induction switch 29 and the limit switch 28 is used to limit the stroke of the slide table 22. When the slide table 22 moves to the limit switch 28, it will immediately move in the opposite direction, thereby realizing the reciprocating movement of the slide table 22.

[0124] The technical effects of the present invention will be further described below in combination with the comparative analysis between specific test results.

[0125] In a specific test example, a piece of black pure cotton fabric is selected as the first fabric to be tested. The test results are as Figure 5 and Figure 6 shown:

[0126] First, the flocked cloth with fluorescent substances is fixed on the roller 30 of the hair sticking device. The device is started, and the roller 30 is rolled on the surface of the black fabric at a speed of 5 cm / s for 30 times back and forth, and then the device is paused. Then, the black fabric with adhered fiber fluff is placed in a completely black light box. The ultraviolet lamp in the completely black light box can emit ultraviolet light with a wavelength of 365 nm. The ultraviolet lamp is turned on to irradiate the fabric to be tested so that the fiber fluff on it fluoresces. A photographing device with a resolution of 20 million pixels and a photographing distance of 30 cm is used to take pictures in the light box to obtain the surface image of the fluorescent color development. Then, the taken surface image is transmitted into the visual detection model for analysis, and the hair sticking situation is judged according to the proportion of the fluorescent area. The results are as Figure 5 and Figure 6 shown, Figure 5 is the image before fluorescent color development, Figure 6 is the image after fluorescent color development. The proportion of the fiber fluff area after fluorescent treatment is 1.275%, and the anti-hair sticking performance of the fabric is evaluated as level 2, and the anti-hair sticking performance is poor.

[0127] In another specific test example, a piece of red polyester fiber fabric was selected as the second fabric to be tested. The test results are as Figure 7 and Figure 8 shown:

[0128] First, the flocked cloth with fluorescent substances was fixed on the roller 30 of the hair sticking device. The device was started, and the roller 30 was rolled on the surface of the black fabric at a speed of 5 cm / s for rolling friction. After rolling back and forth 30 times, the device was paused. Then, the black fabric with adhered fiber fluff was placed in a completely dark light box. The ultraviolet lamp in the completely dark light box can emit ultraviolet light with a wavelength of 365 nm. The ultraviolet lamp was turned on to irradiate the fabric to be tested so that the fiber fluff on it fluoresced. A photographing device with a resolution of 20 million pixels and a photographing distance of 30 cm was used in the light box to take pictures to obtain the surface image of the fluorescent color development. After that, the taken surface image was transmitted into the visual detection model for analysis, and the hair sticking situation was judged according to the proportion of the fluorescent area. The results are as Figure 7 and Figure 8 shown, Figure 7 is the image before fluorescent color development, Figure 8 is the image after fluorescent color development. The proportion of the fiber fluff area after fluorescence treatment is 0.657%, and the anti-hair sticking performance of this fabric is evaluated as level 4, with good anti-hair sticking performance.

[0129] In yet another specific test example, a piece of blue wool fabric was selected as the third fabric to be tested. The test results are as Figure 9 and Figure 10 shown:

[0130] First, the flocked cloth with fluorescent substances was fixed on the roller 30 of the hair sticking device. The device was started, and the roller 30 was rolled on the surface of the black fabric at a speed of 5 cm / s for rolling friction. After rolling back and forth 30 times, the device was paused. Then, the black fabric with adhered fiber fluff was placed in a completely dark light box. The ultraviolet lamp in the completely dark light box can emit ultraviolet light with a wavelength of 365 nm. The ultraviolet lamp was turned on to irradiate the fabric to be tested so that the fiber fluff on it fluoresced. A photographing device with a resolution of 20 million pixels and a photographing distance of 30 cm was used in the light box to take pictures to obtain the surface image of the fluorescent color development. After that, the taken surface image was transmitted into the visual detection model for analysis, and the hair sticking situation was judged according to the proportion of the fluorescent area. The results are as Figure 9 and Figure 10 shown, Figure 9 is the image before fluorescent color development, Figure 10 is the image after fluorescent color development. The proportion of the fiber fluff area after fluorescence treatment is 1.648%, and the anti-hair sticking performance of this fabric is evaluated as level 1, without anti-hair sticking performance.

[0131] To more clearly show the problem that the recognition effect of non-fluorescent dyeing is inaccurate, as Figures 11 to 16As shown, it shows the recognition effect after using non-fluorescent staining.

[0132] like Figure 11 and Figure 12 As shown, it is a first embodiment. Figure 11 This is the surface image taken after the hair is glued. Figure 12 The red area in the figure is the image after contrast enhancement processing and analysis by the visual detection model, rather than a real image. The fiber hairiness area that has not been subjected to fluorescence treatment accounts for 12.5%. The anti-hair performance of the fabric is evaluated as level 1, which means it has no anti-hair performance.

[0133] like Figure 13 and Figure 14 As shown, it is a second embodiment, Figure 13 This is the surface image taken after the hair is glued. Figure 14 The red area in the figure is the image after contrast enhancement processing and analysis by the visual detection model, rather than a real image. The fiber hairiness area that has not been subjected to fluorescence treatment accounts for 8.63%. The anti-hair performance of the fabric is evaluated as level 1, which means it has no anti-hair performance.

[0134] like Figure 15 and Figure 16 As shown, it is a third embodiment, Figure 15 This is the surface image taken after the hair is glued. Figure 16 The red area in the figure is the image after contrast enhancement processing and analysis by the visual detection model, rather than a real image. The fiber hairiness area that has not been subjected to fluorescence treatment accounts for 19.452%. The anti-hair performance of the fabric is evaluated as level 1, which means it has no anti-hair performance.

[0135] pass Figures 11 to 16 It can be seen from the test results shown that due to the low contrast of fiber hairiness on the fabric surface, if the fluorescent color development technology is not used, there will be a lot of misjudgment in the recognition result, and the recognition area will be much larger than the actual hairy area; while in the present application, the fiber hairiness identified after fluorescent color development is basically the fiber hairiness on the fabric surface, with almost no misjudgment and more accurate data.

[0136] Through the above examples of fabrics of different materials and colors, it is verified that the high-precision fabric hair recognition method based on fluorescence enhancement and machine vision of the present invention can better recognize the hair on fabrics of any color, effectively solving the problem of inaccuracy of traditional methods in recognizing the hair on fabrics of certain colors.

[0137] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for identifying lint on fabrics based on fluorescence enhancement and machine vision, characterized in that, It includes the following steps: When dyeing the raised fabric, perform fluorescent dyeing on the raised fabric; Friction the raised fabric with the fabric to be tested so that the fabric to be tested gets hairy; Place the fabric to be tested in a preset environment so that the hairs adhered to the fabric to be tested emit fluorescence; Obtain the surface image of the fabric to be tested after getting hairy and fluorescing; Based on image color recognition, analyze the area ratio of the fluorescent area in the surface image; Determine the hair adhesion level of the fabric to be tested according to the area ratio of the fluorescent area.

2. The method for identifying lint on fabric based on fluorescence enhancement and machine vision according to claim 1, wherein Performing fluorescent dyeing on the raised fabric when dyeing the raised fabric includes adding a fluorescent brightener when dyeing the raised fabric.

3. The fabric lint recognition method based on fluorescence enhancement and machine vision according to claim 1, characterized in that, Friction the raised fabric with the fabric to be tested includes: Mount both the raised fabric and the fabric to be tested on the hair adhesion generating device; Start the hair adhesion generating device to drive the raised fabric to move relative to the fabric to be tested, so that the raised fabric and the fabric to be tested come into contact and friction to get hairy.

4. The method for identifying fabric lint sticking based on fluorescence enhancement and machine vision according to claim 1, characterized in that Placing the fabric to be tested in a preset environment includes: Put the fabric to be tested after getting hairy into a completely black light box (40); Start the ultraviolet lamp (41) in the completely black light box (40) to irradiate the fabric to be tested so that the adhered hairs fluoresce.

5. The fabric lint recognition method based on fluorescence enhancement and machine vision according to claim 4, characterized in that Obtaining the surface image of the fabric to be tested after getting hairy and fluorescing includes: Start the photographing device (42) in the completely black light box (40) to collect an image of the fabric to be tested after fluorescing to obtain the surface image.

6. The method for identifying fabric lint based on fluorescence enhancement and machine vision according to claim 5, wherein, The resolution of the photographed image of the photographing device (42) is 1 million pixels - 100 million pixels; and / or, the distance between the photographing device (42) and the fabric to be tested is 10 cm - 50 cm.

7. The fabric lint recognition method based on fluorescence enhancement and machine vision according to claim 1, characterized in that, Determining the hair adhesion level of the fabric to be tested according to the area ratio of the fluorescent area includes: The lower the area ratio of the fluorescent area, the higher the anti-hair adhesion level of the fabric to be tested and the better the anti-hair adhesion performance; the higher the area ratio of the fluorescent area, the lower the anti-hair adhesion level of the fabric to be tested and the worse the anti-hair adhesion performance.

8. A fabric lint recognition system based on fluorescence enhancement and machine vision, which is used to implement the fabric lint recognition method described in any one of claims 1-7, and is characterized in that, The fabric hair adhesion recognition system has a hair adhesion generating device and a hair adhesion detection device; The hair adhesion generating device includes: An experimental area (11) for mounting the fabric to be tested; Two groups of parallel slide rails (20), and the two groups of slide rails (20) are respectively arranged on both sides of the experimental area (11); A roller (30) for mounting the raised fabric, and both ends of the roller (30) are slidably connected to the two groups of slide rails (20) through a slide table (22) respectively to drive the raised fabric to move relative to the fabric to be tested so as to generate contact friction between the raised fabric and the fabric to be tested; The hair adhesion detection device includes: A completely black light box (40), a detection area (43) for placing the fabric to be tested is arranged in the completely black light box (40), and an ultraviolet lamp (41) and a photographing device (42) are arranged in the completely black light box (40); A computer processing unit (44), the computer processing unit (44) is electrically connected to the photographing device (42), and a machine vision detection model is built in the computer processing unit (44).

9. The fabric lint recognition system based on fluorescence enhancement and machine vision according to claim 8, characterized in that, A lifting mechanism is provided on the sliding table (22), and both ends of the roller (30) are connected to the lifting mechanism, and the lifting mechanism can adjust the distance between the raising cloth on the roller (30) and the fabric to be tested on the experimental area (11).

10. The fabric lint recognition system based on fluorescence enhancement and machine vision according to claim 9, wherein The roller (30) includes a central shaft (31) and a drum rotatably sleeved on the central shaft (31), and the raising cloth is mounted on the drum.

11. The fabric lint recognition system based on fluorescence enhancement and machine vision according to claim 10, characterized in that, The lifting mechanism is an adjusting bolt (221), and both ends of the central shaft (31) are threadedly connected to the adjusting bolt (221), and the rotation of the adjusting bolt (221) can drive the lifting of the central shaft (31).

12. The fabric lint recognition system based on fluorescence enhancement and machine vision according to claim 8, characterized in that, The lint generating device further includes: A synchronous belt (26) arranged parallel to the slide rail (20), the synchronous belt (26) is rotatably sleeved at both ends of the slide rail (20), and the sliding table (22) is fixedly connected to the synchronous belt (26); A driving motor (25) is provided at one end of the slide rail (20), and the driving motor (25) is drivingly connected to the synchronous belt (26) to drive the synchronous belt (26) to rotate.

13. The fabric lint recognition system based on fluorescence enhancement and machine vision according to claim 8, wherein The lint generating device further includes: A lead screw (27) arranged parallel to the slide rail (20), both ends of the lead screw (27) are respectively rotatably connected to both ends of the slide rail (20), and the sliding table (22) is threadedly sleeved on the lead screw (27); A driving motor (25), the driving motor (25) is provided at one end of the slide rail (20), and the driving motor (25) is drivingly connected to the lead screw (27).

Images