Textile printing and dyeing defect detection equipment based on visual intelligence

Abstract

The application relates to the technical field of textile processing, and discloses a textile printing and dyeing defect detection equipment based on visual intelligence, which comprises a detection equipment shell body, and a supporting plate is arranged in the detection equipment shell body. The equipment is matched with core components such as a supporting plate, a disturbance wheel, a pressing wheel, a frame, an electric push rod and a belt, the disturbance wheel forms a uniform conical distribution of undulating support structures through the through slots on the supporting plate, stably drives the flexible belt to apply gentle and controllable wave-shaped deformation to the cloth in conveying, the convex structure of the disturbance wheel is used to effectively increase the local friction when the cloth passes, a conical arrangement of multi-point traction nets is formed, the stretching deformation of the cloth due to the traction force is accurately controlled, the full exposure of the hidden defects of the textile and the weakening of the natural warp and weft texture interference are realized, the false detection rate and the missed detection rate of defect detection are effectively reduced, and the overall detection precision is avoided from being affected by the large-area deformation of the cloth due to external force pulling from the source.

Description

Technical Field

[0001] This invention relates to the field of textile processing technology, and in particular to a textile printing and dyeing defect detection device based on visual intelligence. Background Technology

[0002] Textile printing and dyeing defect image visual inspection equipment is a core automated quality inspection equipment in modern intelligent manufacturing production lines for textile printing and dyeing. It is widely applicable to the online inspection process of various woven and knitted printed and dyed fabrics. It is specifically used for online identification of various printing and dyeing defects and fabric flaws such as missing prints, stains, broken yarns, hidden holes, color differences, skipped patterns, lint, dirt, and disordered weave patterns. It is a key piece of equipment for textile enterprises to control product quality, reduce defective product losses, and improve production efficiency. This type of equipment uses an image visual intelligent recognition device as the core detection unit. Under the operation state of continuous and stable fabric conveying, it acquires high-definition digital images of the fabric surface in real time. Then, through image recognition algorithms, it completes defect feature extraction, comparison, and intelligent judgment, realizing automatic identification, marking, and screening of defects. The entire process adopts a non-contact detection mode, replacing traditional manual visual inspection. It fundamentally solves the pain points of low efficiency, strong subjectivity, high rate of missed and false detections, and fatigue affecting quality in manual inspection, significantly improving the efficiency and accuracy of fabric quality inspection, and promoting the upgrading of textile printing and dyeing quality inspection to intelligence, standardization, and high speed.

[0003] However, the above-mentioned equipment still has the following shortcomings: First, hidden defects such as hidden holes in the fabric and weak points in the yarn cannot be fully exposed when the fabric is flat and taut. The equipment cannot collect effective image data that highlights the defects, resulting in serious missed detection problems. In addition, textile fabrics have natural warp and weft weave textures and fluff. When the equipment collects images when the fabric is flat and being transported, the image features of the natural textures will be highly confused with the features of printing and dyeing defects. The image visual intelligent recognition device cannot accurately distinguish them, which easily leads to false detection. Summary of the Invention

[0004] To address the shortcomings mentioned above, this invention proposes a textile printing and dyeing defect detection device based on visual intelligence.

[0005] To achieve the above objectives, this application adopts the following technical solution: a textile printing and dyeing defect detection device based on visual intelligence, including a detection device housing, a tray is provided inside the detection device housing, and a disturbance component is provided on the upper surface of the tray for disturbing and deforming the fabric during the conveying process.

[0006] The disturbance component includes multiple sets of through slots symmetrically opened on one side of the upper surface of the pallet, and a disturbance wheel is rotatably installed inside each set of through slots. The multiple sets of disturbance wheels are generally distributed in a conical shape. An adjustable frame is provided above the pallet, and multiple sets of pressure rollers are rotatably installed on the lower surface of the frame. The pressure rollers and disturbance wheels are staggered.

[0007] The detection equipment housing has a first pressure roller, a second pressure roller, and a third pressure roller arranged sequentially from one side to the other. The first pressure roller and the second pressure roller form a disturbance recognition zone, and the second pressure roller and the third pressure roller form a stable recognition zone.

[0008] The disturbance recognition area and the stable recognition area correspond to different positions on the fabric conveying path, and are used to collect image data of the same fabric area under different surface conditions.

[0009] Preferably, the outer wall of the frame is fixed with two sets of image vision intelligent recognition devices for collecting image information of the same textile fabric area under different surface conditions. One set of the image vision intelligent recognition devices is located above the disturbance recognition area, and the other set of the image vision intelligent recognition devices is located above the stable recognition area.

[0010] Preferably, two sets of rotating rollers are symmetrically arranged inside the outer shell of the detection equipment, and a belt is fitted on the outer wall of both sets of rotating rollers. The belt and the pallet are interlaced.

[0011] Preferably, an electric push rod is fixed to the upper surface of the housing of the testing equipment, and the output end of the electric push rod extends downward through and into the interior of the housing of the testing equipment, and the output end of the electric push rod is fixed to the frame.

[0012] Preferably, both ends of the first, second, and third pressure rollers are fixed with mounting shafts, and the mounting shafts extend to both sides and penetrate the outer wall of the outer shell of the testing equipment. Multiple sets of lifting adjustment devices are symmetrically fixedly installed on both sides of the outer wall of the outer shell of the testing equipment, and the output end of the lifting adjustment device is installed corresponding to the first, second, and third pressure rollers. The mounting shafts and the output end of the lifting adjustment device are in a rotating installation relationship.

[0013] Preferably, the inner wall of the belt contacts the upper surface of the support plate and multiple sets of agitators, and the pressure roller is located above the belt.

[0014] Preferably, a feed inlet is provided on one side of the outer wall of the outer casing of the testing equipment, and a discharge outlet is provided on the other side of the outer wall of the outer casing of the testing equipment. A guide plate is fixed on the inner wall of the outer casing of the testing equipment at the opening on one side of the feed inlet and the discharge outlet.

[0015] Preferably, a tensioning component and a front roller are sequentially arranged on one side of the outer wall of the outer shell of the testing equipment, and a guide roller is fixed on the outer wall of the outer shell of the testing equipment at the feed inlet opening.

[0016] Preferably, the guide roller, the front roller shaft, and the tensioning component are all located in the front feeding section of the feed inlet, and the third pressure roller, another set of guide plates, and the discharge port together constitute the end discharge section.

[0017] Preferably, the upper surface of the belt, deformed by the disturbance wheel, comes into contact with the fabric in the conveying state.

[0018] The technical effects and advantages of this invention are as follows:

[0019] In this invention, the device utilizes a combination of core components such as a pallet, a disturbance wheel, a pressure wheel, a frame, an electric push rod, and a belt. The disturbance wheel, relying on the grooves on the pallet, forms a uniformly conical undulating support structure, stably driving the flexible belt to apply gentle and controllable wave-like deformation to the conveyed fabric. Simultaneously, the raised structure of the disturbance wheel effectively increases the local friction force as the fabric passes through, forming a conical multi-point traction net. This precisely controls the tensile deformation of the fabric caused by traction, fully exposing hidden defects in textiles and weakening interference from natural warp and weft textures. This effectively reduces the false detection and false negative rates of defect detection, preventing large-area deformation of the fabric due to external force from affecting the overall detection accuracy. Furthermore, by constructing a surface state modulation mechanism for textile fabrics through a disturbance recognition area and a stable recognition area, the separability between defect features and background texture is enhanced from the image acquisition source, providing more discriminative input data for visual intelligent recognition.

[0020] In this invention, the equipment utilizes a first pressure roller, a second pressure roller, a third pressure roller, and two sets of image vision intelligent recognition devices to efficiently cooperate with each other. Relying on the three sets of pressure rollers, a disturbance recognition area and a stable recognition area are orderly divided. High-definition surface images of the fabric under deformation and flatness are collected simultaneously. The two sets of complementary image data are jointly transmitted to the image vision intelligent recognition system for intelligent comparison and analysis, realizing automated, interference-free, and accurate judgment of textile printing and dyeing defects, and significantly improving the overall accuracy and online detection efficiency of defect detection. Attached Figure Description

[0021] The disclosure of this invention is illustrated with reference to the accompanying drawings. It should be understood that the drawings are for illustrative purposes only and are not intended to limit the scope of protection of this invention. In the drawings, the same reference numerals are used to refer to the same parts:

[0022] Figure 1 This is a schematic diagram of the external structure of the present invention;

[0023] Figure 2 This is a schematic diagram of the external structure of the present invention;

[0024] Figure 3 This is a cross-sectional view of the internal structure of the present invention;

[0025] Figure 4 This is a schematic cross-sectional view of the structure of the present invention;

[0026] Figure 5 This is a schematic diagram showing the structural distribution of the pallet, disturbance wheel, and pressure wheel of the present invention;

[0027] Figure 6 This is a top view schematic diagram of the structure distribution of the disturbance wheel and pressure wheel of the present invention.

[0028] Legend: 1. Detection equipment housing; 11. Feed inlet; 12. Discharge outlet; 13. Guide plate; 14. Guide roller; 15. Front roller shaft; 16. Tensioning component; 2. Rotary roller; 21. Belt; 3. Support plate; 31. Disturbing wheel; 32. Through groove; 33. Frame; 34. Pressure roller; 35. Image visual intelligent recognition device; 36. Electric push rod; 4. Lifting and adjusting device; 41. First pressure roller; 42. Second pressure roller; 43. Third pressure roller. Detailed Implementation

[0029] It is readily understood that, based on the technical solution of this invention, those skilled in the art can propose various interchangeable structural methods and implementations without altering the essential spirit of the invention. Therefore, the following detailed embodiments and accompanying drawings are merely illustrative examples of the technical solution of this invention and should not be considered as the entirety of the invention or as limitations or restrictions on the technical solution of this invention.

[0030] Reference Figure 1-6 As shown, the present invention provides a technical solution: a textile printing and dyeing defect detection device based on visual intelligence, including a detection device housing 1, a tray 3 is provided inside the detection device housing 1, the tray 3 provides a stable support foundation for fabric conveying, and ensures the cooperative operation of the belt 21 and the disturbance component. The upper surface of the tray 3 is provided with a disturbance component for disturbing and deforming the fabric during the conveying process.

[0031] The disturbance component includes multiple sets of through slots 32 symmetrically opened on one side of the upper surface of the tray 3, and a disturbance wheel 31 is rotatably installed inside each set of through slots 32. The through slots 32 provide independent rotation space for the disturbance wheel 31 to avoid jamming or interference when the disturbance wheel 31 is running. The multiple sets of disturbance wheels 31 are generally distributed in a conical shape. A height-adjustable frame 33 is provided above the tray 3. Multiple sets of pressure wheels 34 are rotatably installed on the lower surface of the frame 33, and the pressure wheels 34 and the disturbance wheels 31 are staggered.

[0032] Inside the outer casing 1 of the detection equipment, a first pressure roller 41, a second pressure roller 42, and a third pressure roller 43 are arranged sequentially from one side to the other. The first pressure roller 41 and the second pressure roller 42 form a disturbance recognition zone, and the second pressure roller 42 and the third pressure roller 43 form a stable recognition zone.

[0033] The disturbance recognition area and the stable recognition area correspond to different positions on the fabric conveying path, and are used to collect image data of the same fabric area under different surface conditions.

[0034] Reference Figure 3-6 As shown in this embodiment: two sets of image vision intelligent recognition devices 35 are fixed on the outer wall of the frame 33, which are used to collect image information of the same textile fabric area under different surface conditions. One set of image vision intelligent recognition devices 35 is located above the disturbance recognition area, and the other set of image vision intelligent recognition devices 35 is located above the stable recognition area.

[0035] The outer casing 1 of the testing equipment has two sets of rotating rollers 2 symmetrically arranged inside. The outer walls of the two sets of rotating rollers 2 are fitted with belts 21, and the belts 21 and the support plate 3 are interlaced.

[0036] An electric push rod 36 is fixed on the upper surface of the outer shell 1 of the testing equipment, and the output end of the electric push rod 36 extends downward through and into the interior of the outer shell 1 of the testing equipment. The output end of the electric push rod 36 is fixed to the frame 33.

[0037] The first pressure roller 41, the second pressure roller 42, and the third pressure roller 43 are all fixed with mounting shafts at both ends. The mounting shafts extend to both sides and penetrate the outer wall of the outer shell 1 of the testing equipment. Multiple sets of lifting adjustment devices 4 are symmetrically fixedly installed on the outer walls of both sides of the outer shell 1 of the testing equipment. The output end of the lifting adjustment device 4 is installed corresponding to the first pressure roller 41, the second pressure roller 42, and the third pressure roller 43. The mounting shafts and the output end of the lifting adjustment device 4 form a rotating installation relationship.

[0038] The inner wall of the belt 21 is in contact with the upper surface of the support plate 3 and multiple sets of agitators 31, and the pressure roller 34 is located above the belt 21.

[0039] Reference Figure 1-6 As shown in this embodiment: a feed inlet 11 is provided on one side of the outer wall of the outer casing 1 of the testing equipment, and a discharge outlet 12 is provided on the other side of the outer wall of the outer casing 1 of the testing equipment. A guide plate 13 is fixed on the inner wall of the outer casing 1 of the testing equipment and at the opening on one side of the feed inlet 11 and the discharge outlet 12.

[0040] A tensioning component 16 and a front roller 15 are sequentially arranged on one side of the outer wall of the outer shell 1 of the testing equipment. A guide roller 14 is fixed on the outer wall of the outer shell 1 of the testing equipment and at the opening of the feed inlet 11.

[0041] The guide roller 14, the front roller shaft 15, and the tensioning component 16 are all located in the front feeding section of the feed inlet 11. The third pressure roller 43, another set of guide plates 13, and the discharge port 12 together constitute the end discharge section.

[0042] The upper surface of the belt 21, which is deformed by the disturbance wheel 31, comes into contact with the fabric in the conveying state.

[0043] Working Principle: First, the printed and dyed textile fabric to be inspected (hereinafter referred to as the fabric) passes sequentially through the tensioning component 16, guide roller 14, front roller 15, and guide plate 13. The components of the front feeding section work together to guide and pre-adjust the tension of the fabric, ensuring that the fabric enters the equipment in a flat, unbiased, and wrinkle-free state. It enters the outer shell 1 of the inspection equipment through the feed inlet 11. Because a traction device is set at the end of the inspection equipment, the printed and dyed textile fabric is in a constant conveying state. The traction device provides continuous and stable conveying power, keeping the fabric in a uniform conveying rhythm with the production line and avoiding fluctuations in start-stop operation that could affect the image acquisition accuracy. As the fabric continues to pass over the upper surface of the conveyor belt 21, because the belt 21 is always in close contact with the pallet 3 and the multiple sets of agitators 31, some areas of the belt 21 are uneven. The belt 21 forms a natural undulating surface following the contour of the agitators 31. This surface provides the basis for the controllable deformation of the fabric. The fabric that continues to pass over also undulates due to the bulge of the belt 21. At this time, the fabric is deformed locally due to the undulation. It should be noted that this deformation is a flexible passive deformation that can be achieved without rigid compression. It causes less scratching, pulling or fiber damage to the fabric and maintains the original quality of the fabric.

[0044] Secondly, the lifting and adjusting device 4 adjusts the height of the first pressure roller 41, the second pressure roller 42, and the third pressure roller 43 to adapt to textiles of different thicknesses. It can flexibly adjust the pressing height according to the thickness of different fabrics such as thin yarn, medium-thick cotton cloth, and thick canvas, ensuring that the pressure rollers are always in contact with the fabric surface and that the pressure is moderate. Simultaneously, the first pressure roller 41, the second pressure roller 42, and the third pressure roller 43 divide the fabric above the support plate 3 into a disturbance recognition zone and a stable recognition zone. These dual recognition zones enable step-by-step detection of different fabric states, forming a complete and accurate defect recognition structure. Meanwhile, the tensioning component 16, in conjunction with the guide roller 14 and the front roller shaft 15, maintains a stable textile conveying state. Through the control terminal, the fabric conveying tension can be dynamically adjusted in real time to prevent the fabric from becoming loose and slipping, excessively tensile and deformed, or shifting laterally, ensuring that the conveying path and deformation effect are always in a standard state.

[0045] Next, the textile first enters the disturbance identification zone between the first pressure roller 41 and the second pressure roller 42. Multiple sets of disturbance wheels 31, rotatably mounted inside the groove 32, are arranged in a conical shape. This conical layout allows the fabric to gradually undergo gradient deformation from the edge to the center, with the deformation force transitioning evenly from weak to strong. This, combined with a frame 33 that is adjusted for height via an electric push rod 36, and multiple sets of pressure rollers 34 rotatably mounted on the lower surface of the frame 33, press down on the deformed fabric. The alternating arrangement of the disturbance wheels 31 and pressure rollers 34 applies a gradient deformation force to the conveyed textile, causing the textile to... Controllable wave-like deformation can generate uniform local stress on the fabric surface. This avoids the problem of hidden defects not being visible when the fabric is transported flat, as well as the visual recognition problem of confusion between the image features of natural warp and weft texture and printing and dyeing defects. It fully exposes hidden defects and weakens the interference of natural texture. It makes defects such as hidden holes, minor yarn breaks, light color differences, and hidden lint that cannot be detected by conventional inspection fully visible under stress. At the same time, it effectively separates the image features of the natural warp and weft weave texture of the fabric from the defect features, reducing the interference of texture noise on image recognition from the source.

[0046] Furthermore, the raised deformation areas formed by multiple sets of disturbance wheels 31 can increase the local friction when the fabric passes through, forming a "multi-point traction net" with an overall cone-shaped distribution on the lower surface of the fabric. This structure supports the fabric with dispersed high-friction points, replacing the traditional integral support method. The relatively high friction at the undulations can divide and reduce the coverage area of ​​the fabric due to traction force stretching deformation, controlling the stretching deformation area between the staggered disturbance wheels 31. This prevents the fabric from being stretched and deformed over a large area, affecting the overall detection accuracy, and avoids problems such as large-area stretching distortion, pattern shift, and texture distortion caused by uneven overall traction force, ensuring the integrity and authenticity of the original features of the image acquisition.

[0047] At the same time, the electric push rod 36 outputs power to drive the frame 33 to move up and down, and precisely adjusts the distance between the pressure roller 34 and the disturbance roller 31. It can precisely adjust the deformation amplitude according to the material characteristics, thickness parameters and weaving density of the fabric, adapt to the deformation requirements of textiles of different sizes and materials, ensure that the deformation effect of the disturbance component is uniform and stable, and ensure that the degree of deformation can meet the defect exposure requirements without causing the fabric to wrinkle, curl or be damaged due to excessive deformation.

[0048] Next, the image vision intelligent recognition device 35 located above the disturbance recognition area collects surface images of the textile under deformation in real time, captures complete image information after the defect is highlighted in high-definition imaging mode, obtains clear image data with the defect highlighted and without texture interference, provides noise-free and highly recognizable core data for subsequent image comparison and analysis, and reduces the processing difficulty of image recognition algorithms.

[0049] Next, the textile, after being disturbed and deformed, is continuously conveyed to the stable identification zone formed by the second pressure roller 42 and the third pressure roller 43. It should be noted that the first pressure roller 41, the second pressure roller 42, and the third pressure roller 43 are all in contact with the upper surface of the fabric. Among them, the second pressure roller 42 and the third pressure roller 43 flatten and reset the deformed textile. Through the pressure of the two pressure rollers, the fabric is quickly restored to its initial flat and regular state, so that the textile is restored to a stable and regular conveying state.

[0050] Simultaneously, the image vision intelligent recognition device 35 located above the stable recognition area synchronously acquires surface images of the textile in a flat state. The two sets of image vision intelligent recognition devices 35 jointly transmit the image data of the deformed state (i.e., the fabric after disturbance) and the flat state to the image vision intelligent recognition system of the control terminal. The image vision intelligent recognition system of the control terminal performs regional correspondence matching on the two sets of images based on the rotation encoding signal or displacement synchronization information of the rotating roller 2, ensuring that the two sets of images originate from the same fabric area. The system performs grayscale difference, texture feature difference, or edge response difference processing on the two sets of images, and can combine feature extraction, image segmentation, or pattern recognition algorithms to enhance and distinguish the abnormal feature areas, extract abnormal feature areas with stable responses under different surface states, and perform defect matching and intelligent judgment based on the abnormal feature areas. Through image comparison analysis, the system completes the accurate identification and judgment of printing and dyeing defects. Relying on the complementary information of the dual-state images, the system reduces the false detection rate and false negative rate of single-state detection, and achieves accurate positioning, classification, and marking of defects.

[0051] Finally, the textiles that have completed defect detection are guided and conveyed by the guide plate 13. The guide plate 13 guides and limits the fabric smoothly to avoid deviation or jamming during discharge. The textiles are smoothly output from the discharge port 12 to the outer shell 1 of the detection equipment, realizing automated, interference-free, and accurate detection of textile printing and dyeing defects. The detection efficiency, accuracy, and stability of the entire set of equipment are far superior to traditional manual detection and single-state visual detection, improving the accuracy of the textile printing and dyeing quality inspection process.

[0052] The technical scope of this invention is not limited to the content described above. Those skilled in the art can make various modifications and variations to the above embodiments without departing from the technical concept of this invention, and all such modifications and variations should fall within the protection scope of this invention.

Claims

1. A textile printing and dyeing defect detection device based on visual intelligence, characterized in that, The device includes a housing for a testing device, and a tray is provided inside the housing. The upper surface of the tray is provided with a disturbance component for disturbing and deforming the fabric during the conveying process. The disturbance component includes multiple sets of through slots symmetrically opened on one side of the upper surface of the pallet, and a disturbance wheel is rotatably installed inside each set of through slots. The multiple sets of disturbance wheels are generally distributed in a conical shape. An adjustable frame is provided above the pallet, and multiple sets of pressure rollers are rotatably installed on the lower surface of the frame. The pressure rollers and disturbance wheels are staggered. The detection equipment housing has a first pressure roller, a second pressure roller, and a third pressure roller arranged sequentially from one side to the other. The first pressure roller and the second pressure roller form a disturbance recognition zone, and the second pressure roller and the third pressure roller form a stable recognition zone. The disturbance recognition area and the stable recognition area correspond to different positions on the fabric conveying path, and are used to collect image data of the same fabric area under different surface conditions.

2. The textile printing and dyeing defect detection device based on visual intelligence according to claim 1, characterized in that: The outer wall of the frame is fixed with two sets of image vision intelligent recognition devices, which are used to collect image information of the same textile fabric area under different surface conditions. One set of image vision intelligent recognition devices is located above the disturbance recognition area, and the other set of image vision intelligent recognition devices is located above the stable recognition area.

3. The textile printing and dyeing defect detection device based on visual intelligence according to claim 1, characterized in that: The detection equipment has two sets of rotating rollers symmetrically arranged inside its outer casing. The outer walls of the two sets of rotating rollers are fitted with belts, and the belts and the support plates are interlaced.

4. The textile printing and dyeing defect detection device based on visual intelligence according to claim 1, characterized in that: An electric push rod is fixed to the upper surface of the outer shell of the testing equipment, and the output end of the electric push rod extends downward through and into the interior of the outer shell of the testing equipment. The output end of the electric push rod is fixed to the frame.

5. The textile printing and dyeing defect detection device based on visual intelligence according to claim 1, characterized in that: The first, second, and third pressure rollers are all fixed with mounting shafts at both ends, and the mounting shafts extend to both sides and penetrate the outer wall of the outer shell of the testing equipment. Multiple sets of lifting adjustment devices are symmetrically fixedly installed on the outer walls of both sides of the outer shell of the testing equipment, and the output ends of the lifting adjustment devices are installed corresponding to the first, second, and third pressure rollers. The mounting shafts and the output ends of the lifting adjustment devices are in a rotating installation relationship.

6. The textile printing and dyeing defect detection device based on visual intelligence according to claim 3, characterized in that: The inner wall of the belt contacts the upper surface of the support plate and multiple sets of agitators, and the pressure roller is located above the belt.

7. The textile printing and dyeing defect detection device based on visual intelligence according to claim 1, characterized in that: The outer wall of the outer shell of the testing equipment has a feed inlet on one side and a discharge outlet on the outer wall of the other side. A guide plate is fixed on the inner wall of the outer shell of the testing equipment at the openings on one side of the feed inlet and the discharge outlet.

8. The textile printing and dyeing defect detection device based on visual intelligence according to claim 1, characterized in that: A tensioning component and a front roller are sequentially arranged on one side of the outer wall of the outer shell of the testing equipment. A guide roller is fixed on the outer wall of the outer shell of the testing equipment at the feed inlet opening.

9. A textile printing and dyeing defect detection device based on visual intelligence according to claim 8, characterized in that: The guide roller, the front roller shaft, and the tensioning component are all located in the front feeding section of the feed inlet, while the third pressure roller, another set of guide plates, and the discharge port together constitute the end discharge section.

10. A textile printing and dyeing defect detection device based on visual intelligence according to claim 1, characterized in that: The upper surface of the belt, deformed by the disturbance wheel, comes into contact with the fabric in the conveying state.

Images