Cotton defect detection method

Abstract

The invention relates to the technical field of detection methods, and provides a cotton defect detection method, which comprises the following steps: layering treatment: dividing a cotton sample into at least two sub-sample layers, and taking the close surfaces of any two adjacent sub-sample layers as to-be-detected surfaces; and detection processing: collecting image information of the to-be-detected surface of each sub-sample layer, and analyzing and processing the image information so as to count the type and the number of the defects on the to-be-detected surface of each sub-sample layer. According to the arrangement, the internal layers of the cotton sample can be exposed by layering the cotton sample, so that the internal layers of the cotton sample can be detected as a to-be-detected surface, the internal layers of the cotton sample are detected, and the detection of the cotton sample is more sufficient in combination with the detection of the surface layer of the cotton sample; and the accuracy of the detection result is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of detection methods, and in particular to a method for detecting cotton defects. Background Art

[0002] As a basic bulk commodity in the cotton textile industry, to maximize its value in the trade field, the key lies in the precise detection and strict control of the ginning quality in the production, processing, and inspection links. Ginning quality is one of the more critical indicators in cotton quality inspection. The quality of ginning has a great impact on the quality of the yarn in the production process. Ginning quality includes the smoothness of the appearance of lint after processing seed cotton and the types and quantities of defects (such as cotton knots, floss, impurities, abnormal short fiber rate, etc.). Among them, the detection of the types and quantities of defects constitutes the core of the detection work.

[0003] At present, the defect detection technologies in ginning quality are mainly divided into two categories. One is the human sensory observation method, and the other is the method using machine vision and optical detection technologies. The manual sensory observation method for defect detection has problems such as high labor intensity, strong subjectivity, and low detection efficiency. Compared with the traditional manual sensory detection method, mechanical vision and optical detection technologies have achieved significant improvements in consistency and accuracy. However, both of the above two defect detection methods mainly focus on the defect recognition on the surface of the cotton sample, rather than delving into the internal layers of the cotton sample. Given the diversity of cotton samples, relying solely on the detection of surface defects has limitations, lacks the necessary rigor, and the defect detection results of cotton samples are inaccurate.

[0004] Therefore, how to solve the problem of inaccurate defect detection results of cotton samples in the related art has become an important technical problem to be solved by those skilled in the art. Summary of the Invention

[0005] The present invention provides a method for detecting cotton defects to solve the defect of inaccurate defect detection results of cotton samples in the related art.

[0006] The present invention provides a method for detecting cotton defects, including: Stratification processing, dividing the cotton sample into at least two sub-sample layers, and the surfaces of any two adjacent sub-sample layers that are close to each other are the surfaces to be detected; Detection processing, collecting the image information of the surfaces to be detected of each sub-sample layer, and analyzing and processing the image information to count the types and quantities of defects on the surfaces to be detected of each sub-sample layer.

[0007] According to a cotton defect detection method provided by the present invention, the layering process is carried out by a rolling member on a support tabletop. The cotton sample is conveyed to the support tabletop by a conveyor belt mechanism. The rolling axis of the rolling member is parallel to the support tabletop, and the rolling axis of the rolling member is perpendicular to the conveying direction of the conveyor belt mechanism. Before the layering process, it further includes: Dispersion treatment, manually dispersing the cotton sample in a compact state into a loose state, and placing the cotton sample in the loose state on the conveyor belt mechanism.

[0008] According to a cotton defect detection method provided by the present invention, the rolling member is adapted to lift relative to the support tabletop and is adapted to reciprocate along the conveying direction of the conveyor belt mechanism above the conveyor belt mechanism and the support tabletop. The layering process includes: Providing a first adhesive layer on the rolling member and a second adhesive layer on the support tabletop; Moving the rolling member above the conveyor belt mechanism and pressing the rolling member against the cotton sample on the conveyor belt mechanism; Rolling the rolling member so that the cotton sample on the conveyor belt mechanism adheres to the first adhesive layer on the rolling member; Moving the rolling member above the support tabletop and pressing the rolling member against the second adhesive layer on the support tabletop; Rolling the rolling member and simultaneously moving the rolling member relative to the support tabletop to separate the cotton sample into a first sub-sample layer adhering to the first adhesive layer and a second sub-sample layer adhering to the second adhesive layer.

[0009] According to a cotton defect detection method provided by the present invention, pressing the rolling member against the cotton sample on the conveyor belt mechanism includes: Detecting the pressure of the rolling member on the conveyor belt mechanism; Controlling the lifting of the rolling member according to the pressure of the rolling member on the conveyor belt mechanism until the pressure of the rolling member on the conveyor belt mechanism is within a first preset pressure range; Pressing the rolling member against the second adhesive layer on the support tabletop includes: Detecting the pressure of the rolling member on the support tabletop; Controlling the lifting of the rolling member according to the pressure of the rolling member on the support tabletop until the pressure of the rolling member on the support tabletop is within a second preset pressure range.

[0010] According to a cotton defect detection method provided by the present invention, at least one of the rolling member and the support tabletop is provided with a heating mechanism. Before rolling the rolling member and simultaneously moving the rolling member relative to the support tabletop, it further includes: Detecting the humidity of the cotton sample; Controlling the start and stop of the heating mechanism according to the humidity of the cotton sample until the humidity of the cotton sample is less than a preset humidity.

[0011] According to a cotton defect detection method provided by the present invention, the layering process further includes: Controlling the vibration of the support tabletop; And / or, controlling the support tabletop to tilt at a preset angle.

[0012] According to a cotton defect detection method provided by the present invention, before the detection process, it further includes: Removing process, removing the first adhesive layer adhering to the first sub-sample layer from the rolling member, and removing the second adhesive layer adhering to the second sub-sample layer from the support tabletop; Flattening process, flattening the first sub-sample layer and the second sub-sample layer on the detection tabletop, and the opposite surface of the to-be-detected surface of each sub-sample layer is in contact with the detection tabletop.

[0013] According to a cotton defect detection method provided by the present invention, at least one of the surface of the rolling member and the support tabletop is provided with a texture structure.

[0014] According to a cotton defect detection method provided by the present invention, during the detection process, at least two camera mechanisms are used to collect the image information, and the shooting angles of each camera mechanism are different.

[0015] According to a cotton defect detection method provided by the present invention, the color of the detection tabletop is black, and the reflectivity of the detection tabletop is less than a preset value.

[0016] The cotton defect detection method provided by the present invention includes a layering process and a detection process. The layering process divides a cotton sample into at least two sub-sample layers, and the surfaces of any two adjacent sub-sample layers that are close to each other can be used as the surfaces to be detected. The detection process is to collect the image information of the surfaces to be detected of each sub-sample layer and analyze and process the image information to count the types and quantities of defects on the surfaces to be detected of each sub-sample layer. With such a setting, by performing a layering process on the cotton sample, the internal layers of the cotton sample can be exposed, and the internal defects of the cotton sample can be revealed. Thus, the internal layers of the cotton sample can be used as the surfaces to be detected for detection, realizing the detection of the internal layers of the cotton sample. Combining the detection of the surface layer of the cotton sample, the detection of the cotton sample is more sufficient, and the accuracy of the detection result is improved, solving the problem of inaccurate defect detection results for cotton samples in the related art. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in 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 drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0018] Figure 1 is a flowchart of the cotton defect detection method provided by the present invention.

[0019] Figure 2 is a schematic diagram of the relative positions of the support table and the rolling member during the cotton defect detection process provided by the present invention.

[0020] Figure 3 is a schematic diagram of the structure of the rolling member when it is arranged on the double-axis sliding table mechanism provided by the present invention.

[0021] Figure 4 is a schematic diagram of the adhesion effect of the first sub-sample layer and the second sub-sample layer adhered to the first adhesive layer and the second adhesive layer provided by the present invention.

[0022] REFERENCE NUMERALS: 1. Rolling member; 2. Support table; 3. Conveyor belt mechanism; 4. First adhesive layer; 5. Second adhesive layer; 6. Camera mechanism; 7. Double-axis sliding table mechanism. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without creative efforts shall fall within the protection scope of the present invention.

[0024] The following Figures 1 to 4 describes the cotton defect detection method of the present invention.

[0025] As Figure 1 and Figure 4 shown, the cotton defect detection method provided by the embodiment of the present invention mainly includes the following steps 110 to 120.

[0026] Step 110, hierarchical processing, dividing the cotton sample into at least two sub-sample layers, and the surfaces of any two adjacent sub-sample layers that are close to each other are the surfaces to be detected.

[0027] Step 120, detection processing, collecting the image information of the surface to be detected of each sub-sample layer, and analyzing and processing the image information to count the types and quantities of defects on the surfaces to be detected of each sub-sample layer.

[0028] Before the above hierarchical processing, the surface layer of the cotton sample can be detected first. Specifically, the image information of the surface layer of the cotton sample can be collected, and by analyzing and processing the image information of the surface layer of the cotton sample, the types and quantities of defects on the surface layer of the cotton sample can be counted.

[0029] After completing the detection of the surface layer of the cotton sample, the cotton sample is then subjected to hierarchical processing to expose the internal layers of the cotton sample and make the internal defects of the cotton sample appear, so that the internal layers of the cotton sample can be used as the surfaces to be detected for detection, realizing the detection of the internal layers of the cotton sample.

[0030] With such a setting, by combining the detection of the surface layer of the cotton sample and the detection of the internal layers of the cotton sample, the detection of the cotton sample is more sufficient, the accuracy of the detection result is improved, the limitation of the surface layer detection is overcome, and the problem of inaccurate defect detection results of cotton samples in the related art is solved.

[0031] In the embodiment of the present invention, the hierarchical processing process is carried out by using the rolling member 1 on the support table 2. The cotton sample can be conveyed to the support table 2 by the conveyor belt mechanism 3. The rolling axis of the rolling member 1 is parallel to the support table 2, and the rolling axis of the rolling member 1 is perpendicular to the conveying direction of the conveyor belt mechanism 3.

[0032] Before the layering process, a dispersion process is also required. The dispersion process can manually disperse the cotton samples in a compact state into a loose state and place the loose cotton samples on the conveyor belt mechanism 3.

[0033] The dispersion process can spread out the cotton samples, avoid the cotton samples from aggregating too tightly, and avoid affecting the subsequent layering process and detection process.

[0034] The conveying speed of the conveyor belt mechanism 3 can be adjusted according to the size and shape of the cotton samples to ensure that the cotton samples can approach the rolling member 1 at an appropriate speed.

[0035] In this embodiment, the rolling member 1 can move up and down relative to the support table 2, and the rolling member 1 can reciprocate in the conveying direction of the conveyor belt mechanism 3 above the conveyor belt mechanism 3 and the support table 2.

[0036] During the layering process, a first adhesive layer 4 needs to be set on the rolling member 1, and a second adhesive layer 5 needs to be set on the support table 2. Then, the rolling member 1 is raised to a certain height, and the rolling member 1 is moved above the conveyor belt mechanism 3. When the conveyor belt mechanism 3 conveys the cotton samples until the end of the cotton samples is exactly below the rolling member 1, the rolling member 1 is lowered until the rolling member 1 presses tightly on the cotton samples on the conveyor belt mechanism 3.

[0037] It should be noted that when the conveyor belt mechanism 3 conveys the cotton samples until the end of the cotton samples is exactly below the rolling member 1, the conveyor belt mechanism 3 can be first controlled to stop running. After the rolling member 1 is lowered until the rolling member 1 presses tightly on the cotton samples on the conveyor belt mechanism 3, the conveyor belt mechanism 3 is then controlled to run.

[0038] After the rolling member 1 presses tightly on the cotton samples on the conveyor belt mechanism 3, the rolling member 1 is made to roll, so that the cotton samples on the conveyor belt mechanism 3 can adhere to the first adhesive layer 4 on the rolling member 1.

[0039] After the cotton samples adhere to the first adhesive layer 4 on the rolling member 1, the rolling member 1 is then raised to a certain height, and the rolling member 1 is moved in the conveying direction of the conveyor belt mechanism 3 until the rolling member 1 is above the support table 2.

[0040] After the rolling member 1 is located above the support table 2, the rolling member 1 is lowered until it presses against the second adhesive layer 5 on the support table 2. Then the rolling member 1 is rolled while moving the rolling member 1 relative to the support table 2. When the cotton sample adhered to the rolling member 1 contacts the second adhesive layer 5, it will also adhere to the second adhesive layer 5 of the support table 2. As the rolling member 1 rolls and moves, a tearing force can be generated on the cotton sample, thereby tearing apart the part adhered to the first adhesive layer 4 and the part adhered to the second adhesive layer 5, completing the layering of the cotton sample. The part adhered to the first adhesive layer 4 and the part adhered to the second adhesive layer 5 are both sub-sample layers. Among them, the part adhered to the first adhesive layer 4 is the first sub-sample layer, and the part adhered to the second adhesive layer 5 is the second sub-sample layer.

[0041] When controlling the rolling and movement of the rolling member 1, it is necessary to make the rolling speed of the rolling member 1 match the movement speed to avoid stretching the cotton sample.

[0042] In some embodiments, the above-mentioned rolling member 1 can be set as a roller with a circumferential length greater than the length of the cotton sample, and the first adhesive layer 4 can be fixed on the circumferential cylindrical surface of the roller.

[0043] In other embodiments, the above-mentioned rolling member 1 can be set as a conveyor belt. Two conveyor belt pulleys are configured for the conveyor belt, and the two conveyor belt pulleys are spaced apart. The conveyor belt is wound around the outside of the two conveyor belt pulleys. When the conveyor belt pulleys rotate, the conveyor belt will roll around the conveyor belt pulleys. The first adhesive layer 4 can be fixed on the outer surface of the conveyor belt away from the conveyor belt pulleys. The circumferential length of the conveyor belt is relatively large, which can adapt to cotton samples with larger length dimensions.

[0044] The above-mentioned conveyor belt can be selected as a toothed belt. The toothed belt can effectively avoid the problem of slipping, and has a higher transmission efficiency, does not require lubrication, has a more compliant tooth engagement, lower running noise, and reduces the maintenance requirements.

[0045] The dimension of the rolling member 1 along its own rolling axis direction needs to be consistent with the dimension of the cotton sample along the rolling axis direction of the rolling member 1, so that when the rolling member 1 rolls once, all positions of the entire layer of the cotton sample can be revealed, which is beneficial to improving the layering efficiency.

[0046] When using the first adhesive layer 4 and the second adhesive layer 5 for layering, there is a tearing force on the cotton sample, which will cause the cotton sample to be arranged perpendicular to the adhesive layer. To avoid observing the defects, it is necessary to use a non-adhesive device to comb each sub-sample layer before the detection process.

[0047] The above-mentioned first adhesive layer 4 and second adhesive layer 5 can be tapes; they can also be adhesive material layers such as silicone and polyurethane with moderate viscosity and softness, which can adapt to cotton samples with different densities and forms. The first adhesive layer 4 and second adhesive layer 5 with corresponding viscosities can also be selected as required according to the density and form of the cotton sample.

[0048] The cotton sample adhered to the rolling member 1 rolls once on the support table 2, and a sub-sample layer can be formed on a second adhesive layer 5. After removing the second adhesive layer 5 from the support table 2, a second adhesive layer 5 can be set on the support table 2 again, and then the rolling member 1 rolls once on the second adhesive layer 5, and another sub-sample layer can be formed. By repeating this cycle, multiple sub-sample layers can be formed, so as to realize the detection of multiple internal positions of the cotton sample.

[0049] Using tapes as the first adhesive layer 4 and second adhesive layer 5 for a series of layering experiments, the adhesion effects on the first adhesive layer 4 and second adhesive layer 5 are as Figure 4 shown, and defects such as lint and dead cotton flakes contained in the cotton sample can be observed. The adhesive layering method provided by the embodiment of the present invention is feasible and effective.

[0050] In this embodiment, when pressing the rolling member 1 against the cotton sample on the conveyor belt mechanism 3, it is necessary to detect the pressure of the rolling member 1 on the conveyor belt mechanism 3, and control the lifting of the rolling member 1 according to the pressure of the rolling member 1 on the conveyor belt mechanism 3 until the pressure of the rolling member 1 on the conveyor belt mechanism 3 is within the first preset pressure range.

[0051] Similarly, when pressing the rolling member 1 against the second adhesive layer 5 on the support table 2, it is necessary to detect the pressure of the rolling member 1 on the support table 2, and control the lifting of the rolling member 1 according to the pressure of the rolling member 1 on the support table 2 until the pressure of the rolling member 1 on the support table 2 is within the second preset pressure range.

[0052] Such a setting can avoid excessive compression of the cotton sample and ensure that the cotton sample is not stretched or broken.

[0053] In a specific embodiment, a pressure sensor can be set on the rolling member 1, or a pressure sensor can be set on the conveyor belt mechanism 3 and the support table 2.

[0054] The rolling member 1 can move along the conveying direction of the conveyor belt mechanism 3 and can also move up and down. In a specific embodiment, the rolling member 1 can be set on a two-axis sliding table mechanism 7.

[0055] In the embodiment of the present invention, a heating mechanism is provided in at least one of the rolling member 1 and the support table 2, and the heating mechanism can heat the cotton sample.

[0056] During the layering process, before making the rolling member 1 roll and at the same time moving the rolling member 1 relative to the support tabletop 2, the humidity of the cotton sample can also be detected, and then according to the humidity of the cotton sample, the start and stop of the heating mechanism can be controlled until the humidity of the cotton sample is less than the preset humidity.

[0057] For cotton samples with humidity greater than the preset humidity, the heating mechanism can be used to heat the cotton samples to reduce the humidity of the cotton samples, increase the looseness of the cotton samples, reduce the layering difficulty of the cotton samples, and better display the internal layers of the cotton samples.

[0058] In this embodiment, during the layering process, the support tabletop 2 can also be controlled to vibrate or the support tabletop 2 can be controlled to tilt at a preset angle. With such a setting, the loosening and layering of the cotton samples can be further promoted, and the stability and uniform distribution of the cotton samples on the support tabletop 2 can be ensured.

[0059] In the embodiment of the present invention, before the flattening process, there are also a peeling process and a flattening process. The peeling process is to peel the first adhesive layer 4 adhering to the first sub-sample layer from the rolling member 1 and to peel the second adhesive layer 5 adhering to the second sub-sample layer from the support tabletop 2. The flattening process is to flatten the first sub-sample layer and the second sub-sample layer on the detection tabletop, and the opposite surface of the detection surface of each sub-sample layer is in contact with the detection tabletop.

[0060] After peeling off the first adhesive layer 4 and the second adhesive layer 5, then flatten the first adhesive layer 4 and the second adhesive layer 5 on the detection tabletop. After the first adhesive layer 4 and the second adhesive layer 5 are flattened, image information is collected, which can improve the clarity of the image information and the accuracy of the detection result.

[0061] Each time the layering process is carried out, the first adhesive layer 4 and the second adhesive layer 5 need to be reset on the rolling member 1 and the support tabletop 2.

[0062] In this embodiment, a texture structure is provided on at least one of the surface of the rolling member 1 and the support tabletop 2. The micro-texture structure can help the cotton sample to loosen and spread evenly during the process of the cotton sample adhering to the first adhesive layer 4, and avoid the cotton sample from being over-compressed.

[0063] The above texture structure can be a tiny concave-convex texture, a tiny grid-like or dot-like structure, and the size and spacing of the texture can be adjusted according to the lump characteristics of the cotton sample.

[0064] In this embodiment, during the detection process, at least two camera mechanisms 6 are used to collect image information, and the shooting angles of each camera mechanism 6 are different. Each camera mechanism 6 shoots the detection surface from different angles, which can ensure a comprehensive detection of the detection surface.

[0065] The above-mentioned camera mechanism 6 can, but is not limited to, an industrial camera. Electrically connecting the industrial camera to a computer can upload the image information collected by the industrial camera to the computer for analysis and processing by the computer. The quality of the rolling process can be evaluated in a non-site environment or laboratory conditions, significantly improving the flexibility and adaptability of the operation.

[0066] When collecting image information, a light source close to natural light can be selected. To avoid interference with the detection results caused by reflection due to the interaction between the light source and the first adhesive layer 4 and the second adhesive layer 5, the height of the light source can be appropriately adjusted or a polarizer can be installed on the camera mechanism 6 to reduce reflection.

[0067] A directional light source and a uniform light source can be used in combination to ensure uniform illumination and clear details of the cotton sample during shooting.

[0068] In this embodiment, the color of the detection tabletop can be set to black to increase the contrast between the sub-sample layer and the detection tabletop. And the reflectivity of the detection tabletop is made less than the preset value to ensure the clarity of the image information.

[0069] In some embodiments, the detection tabletop is set in a structural form with adjustable height. During the detection and processing, by adjusting the height of the detection tabletop, the distance between the camera mechanism 6 and the detection tabletop can be adjusted to adjust the clarity of the captured image information.

[0070] In still other embodiments, the detection tabletop is set in a structural form with adjustable tilt angle. During the detection and processing, by adjusting the tilt angle of the detection tabletop, the display angle of the cotton sample is adjusted to avoid the stacking of the sub-sample layer.

[0071] In summary, the cotton defect detection method provided by the embodiments of the present invention effectively ensures the efficient processing and accurate detection of cotton samples through manual preliminary dispersion, the dispersion effect of the rolling members, the stable stratification of the rolling member 1 and the support tabletop 2, and the image acquisition of the sub-sample layer by the camera mechanism 6, especially for the identification and classification of internal defects in lint cotton.

[0072] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting cotton defects, characterized in that, Including: Stratification processing, dividing the cotton sample into at least two sub-sample layers, and the surfaces of any two adjacent sub-sample layers that are close to each other are the surfaces to be detected; Detection processing, collecting the image information of the surface to be detected of each sub-sample layer, and analyzing and processing the image information to count the types and quantities of defects on the surface to be detected of each sub-sample layer.

2. The cotton defect detection method according to claim 1, characterized in that, The stratification processing is carried out by a rolling member (1) on a support table (2). The cotton sample is conveyed to the support table (2) by a conveyor belt mechanism (3). The rolling axis of the rolling member (1) is parallel to the support table (2), and the rolling axis of the rolling member (1) is perpendicular to the conveying direction of the conveyor belt mechanism (3). Before the stratification processing, it also includes: Dispersion processing, manually dispersing the cotton sample in a compact state into a loose state, and placing the cotton sample in the loose state on the conveyor belt mechanism (3).

3. The cotton defect detection method according to claim 2, wherein, The rolling member (1) is adapted to lift relative to the support table (2), and is adapted to reciprocate along the conveying direction of the conveyor belt mechanism (3) above the conveyor belt mechanism (3) and the support table (2). The stratification processing includes: Setting a first adhesive layer (4) on the rolling member (1) and a second adhesive layer (5) on the support table (2); Moving the rolling member (1) above the conveyor belt mechanism (3) and pressing the rolling member (1) against the cotton sample on the conveyor belt mechanism (3); Making the rolling member (1) roll so that the cotton sample on the conveyor belt mechanism (3) adheres to the first adhesive layer (4) on the rolling member (1); Moving the rolling member (1) above the support table (2) and pressing the rolling member (1) against the second adhesive layer (5) on the support table (2); Making the rolling member (1) roll and at the same time making the rolling member (1) move relative to the support table (2) to separate the cotton sample into a first sub-sample layer adhering to the first adhesive layer (4) and a second sub-sample layer adhering to the second adhesive layer (5).

4. The cotton defect detection method according to claim 3, wherein The pressing the rolling member (1) against the cotton sample on the conveyor belt mechanism (3) includes: Detecting the pressure of the rolling member (1) on the conveyor belt mechanism (3); Controlling the lifting of the rolling member (1) according to the pressure of the rolling member (1) on the conveyor belt mechanism (3) until the pressure of the rolling member (1) on the conveyor belt mechanism (3) is within a first preset pressure range; The pressing the rolling member (1) against the second adhesive layer (5) on the support table (2) includes: Detecting the pressure of the rolling member (1) on the support table (2); Controlling the lifting of the rolling member (1) according to the pressure of the rolling member (1) on the support table (2) until the pressure of the rolling member (1) on the support table (2) is within a second preset pressure range.

5. The cotton defect detection method according to claim 3, wherein, At least one of the rolling member (1) and the support tabletop (2) is provided with a heating mechanism. Before making the rolling member (1) roll and at the same time making the rolling member (1) move relative to the support tabletop (2), it further includes: Detecting the humidity of the cotton sample; Controlling the start and stop of the heating mechanism according to the humidity of the cotton sample until the humidity of the cotton sample is less than a preset humidity.

6. The cotton defect detection method according to claim 3, characterized in that, The layering process further includes: Controlling the vibration of the support tabletop (2); And / or controlling the support tabletop (2) to tilt at a preset angle.

7. The cotton defect detection method according to claim 3, characterized in that, Before the detection process, it further includes: Removing process, removing the first adhesive layer (4) adhering to the first sub-sample layer from the rolling member (1), and removing the second adhesive layer (5) adhering to the second sub-sample layer from the support tabletop (2); Laying process, laying the first sub-sample layer and the second sub-sample layer flat on the detection tabletop, and the opposite surface of the surface to be detected of each sub-sample layer is in contact with the detection tabletop.

8. The cotton defect detection method according to claim 2, characterized in that, At least one of the surface of the rolling member (1) and the support tabletop (2) is provided with a texture structure.

9. The cotton defect detection method according to claim 1, characterized in that During the detection process, at least two camera mechanisms (6) are used to collect the image information, and the shooting angles of each camera mechanism (6) are different.

10. The cotton defect detection method according to claim 7, characterized in that, The color of the detection tabletop is black, and the reflectivity of the detection tabletop is less than a preset value.

Images