Silk ingot defect image acquisition device

Abstract

The utility model discloses a filament ingot defect image acquisition device, which comprises a strip-shaped light source capable of forming a strip-shaped emergent light beam, the emergent light beam is perpendicular to the axis of a filament ingot and irradiates towards one end face of the filament ingot, and the end face of the filament ingot is positioned in the emergent light beam; the end face camera faces the end face of the silk spindle and is used for collecting images of the end face of the silk spindle. By adopting the technical scheme, the silk spindle defect image acquisition device disclosed by the utility model has the advantages that the defect visibility is improved, the defects formed on the end surface of the silk spindle can form obvious characteristics such as highlight or shadow under the assistance of illumination due to the design of the strip-shaped light source, and particularly, the identification capability on fine defects such as broken filaments and fine stumbling filaments is obviously enhanced.

Description

Technical Field

[0001] This utility model relates to a device for acquiring images of defects in silk spindles, belonging to the field of silk spindle appearance defect detection technology. Background Technology

[0002] The appearance quality defects of chemical fiber spindles or cakes include fuzz, strands, clumps, loops, oil stains, dents, dirt, poor forming, and paper tube damage. Visual inspection technology can effectively identify these defects. Modern visual inspection systems utilize high-resolution cameras, image processing algorithms, and intelligent analysis technology to achieve automated and accurate identification of fiber appearance. During visual inspection, an industrial camera first captures surface images of the spindle or cake within a defined inspection area. By controlling the light source angle and brightness, different defects are made clearer in the image. The acquired images are then preprocessed, including noise reduction, contrast enhancement, and edge enhancement, to ensure image clarity and effectiveness. Image processing algorithms then extract image features related to different defects, such as color, texture, shape, contour, and size. Different feature extraction methods are used based on the characteristics of different defects. Finally, a preset classification algorithm compares the extracted features with a standard template to determine the type and severity of the defect. The system can set different thresholds to automatically classify different defects, such as minor defects and severe defects. For example, the applicant's prior application CN117388269A discloses a device and method for detecting appearance defects in silk spindles, and another prior application CN118333955A discloses a model, method, and application for detecting and predicting the severity level of surface defects in industrial products.

[0003] However, some defects, such as fine fuzz or tiny strands, are so small that they may be difficult for a camera to detect. Fuzz typically refers to tiny, scattered, hair-like fibers on the surface of synthetic fibers. This is often caused by fiber bundle splitting during mechanical friction, traction, or winding, resulting in some monofilaments detaching from the main bundle and forming exposed fuzz. Strands refer to the phenomenon of fibers becoming entangled, crossed, or stuck during the winding of a yarn spindle or yarn cake, causing uneven yarn winding and resulting in localized fiber accumulation or irregular entanglement. Fuzz and strands are usually extremely fine fibers, with diameters that are significantly smaller than the entire fiber bundle or spindle. Even with a high-resolution camera, these small defects may not be clearly captured due to insufficient pixel resolution. While industrial cameras have high pixel density, they may still be insufficient to distinguish extremely fine fuzz, especially when the fuzz is only a few pixels in size in the image, affecting image clarity and detail. Fiber filaments and stray fibers share a high degree of similarity in color and material with the surrounding fibers, resulting in low contrast with the background in images and making them difficult to distinguish effectively. In visual inspection images, there may not be sufficient color contrast to accurately detect them. Often, fiber filament or stray fiber defects may have a certain three-dimensional structure; for example, fiber filaments may be suspended on the surface of the main fiber. Traditional two-dimensional imaging struggles to capture the depth information of these defects, making their outlines indistinct in planar images. Utility Model Content

[0004] Therefore, the purpose of this utility model is to overcome the above-mentioned defects in the prior art and provide a spindle defect image acquisition device. By optimizing the light source, the surface defects of the spindle are displayed more clearly in the image captured by the camera under the action of the light source, thereby helping to accurately identify the defects.

[0005] To achieve the above objectives, the present invention provides a wire spindle defect image acquisition device, comprising:

[0006] A strip light source is capable of forming a strip-shaped emitted beam. The emitted beam is perpendicular to the axis of the spindle and illuminates one end face of the spindle, which is located within the emitted beam.

[0007] An end-face camera, positioned facing the end face of the spindle, is used to acquire images of the spindle's end face.

[0008] There are multiple strip light sources, which are used to irradiate the upper and lower end faces of the wire spindle.

[0009] Multiple strip light sources are used to illuminate each end face of the wire spindle.

[0010] Multiple strip light sources are evenly distributed around the circumference of the spindle.

[0011] The image acquisition device for defects in the spindle also includes a spindle conveying mechanism, which is used to move the spindle to the shooting station for shooting.

[0012] Two end-face cameras are used to photograph the same end face of the spindle, and the two end-face cameras are symmetrically arranged on the left and right sides of the moving trajectory plane of the spindle's central axis.

[0013] There are two shooting stations; at each shooting station, each end-face camera captures a single image including a 1 / 4 image of the spindle end face.

[0014] The image acquisition device for defects in the spindle also includes a side-view camera for taking pictures toward the side wall of the spindle and an auxiliary light source.

[0015] The image acquisition device for defects in the silk spindle also includes an oblique camera for taking pictures of the paper tube facing the silk spindle.

[0016] The image acquisition device for defects in the spindle also includes a side-view camera for taking pictures toward the side wall of the spindle and an auxiliary light source.

[0017] The side-viewing cameras are multiple, and the multiple side-viewing cameras are evenly distributed in the circumferential direction on the outer side wall of the spindle.

[0018] By adopting the above technical solution, the wire spindle defect image acquisition device of this utility model has the following beneficial effects compared with the prior art:

[0019] 1. Improved defect visibility: The design of the strip light source allows defects formed on the end face of the spindle to form obvious features such as highlights or shadows with the aid of light, especially significantly enhancing the ability to identify small defects such as fuzz and fine strands.

[0020] 2. Using multiple strip light sources to illuminate from different directions avoids the paper tube of the silk spindle blocking the light. The uniform distribution of the strip light sources around the silk spindle helps to reduce shadows and light spots, thereby improving the overall image quality and contrast, making defects easier to identify.

[0021] 3. Two end-face cameras are symmetrically set on the left and right sides of the moving trajectory plane of the central axis of the spindle to take pictures of the same end face of the spindle. The range of each end-face camera in a single shot is 1 / 4 of the area of ​​the spindle end face, which avoids the obstruction of the paper tube and ensures that the key features and possible defects of the spindle can be captured in each acquisition.

[0022] 4. By setting up side-view and oblique-view cameras, the surface features of the spindle can be captured from multiple dimensions, further enhancing the comprehensiveness and accuracy of defect detection.

[0023] 5. The introduction of the spindle conveying mechanism makes the entire inspection process more automated, improves production efficiency, reduces the need for manual intervention, and also facilitates the automatic movement of the spindle from one shooting station to another for shooting. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the structure of the wire spindle defect image acquisition device of this utility model.

[0025] Figure 2 This is a schematic diagram showing the distribution of the bar light source, the end-face camera, the side-facing camera, and the oblique-facing camera.

[0026] Figure 3 for Figure 2 Top view

[0027] Figure 4 This is a schematic diagram showing the beam emitted by a bar light source illuminating the filaments on the end face of a spindle.

[0028] Figure 5 A schematic diagram illustrating the state of the end-face camera during the shooting process.

[0029] Figure 6 This is a schematic diagram of state two during the shooting process of the end-face camera.

[0030] Figure 7 This is a schematic diagram of state three during the shooting process of the end-face camera.

[0031] Figure 8 Image 1 of the end face of the spindle, obtained by the end face camera when the bar light source is not turned on.

[0032] Figure 9 Image 1 of the end face of the spindle obtained by the rear-facing camera after the bar light source is turned on.

[0033] Figure 10 for Figure 9 A magnified view of part A in the image.

[0034] Figure 11 Image 2 shows the end face of the spindle obtained by the end face camera when the bar light source is not turned on.

[0035] Figure 12 Image 2 of the end face of the spindle, obtained by the rear-facing camera after the bar light source is turned on. Detailed Implementation

[0036] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0037] like Figures 1-3As shown, this utility model provides a spindle defect image acquisition device, including a frame 1 and a spindle conveying mechanism 2, a strip light source 3, an end face camera 4, a side-facing camera 5, an auxiliary light source 6, and an oblique-facing camera 7, all mounted on the frame 1. The spindle conveying mechanism 2 can be a conveyor belt or a conveyor roller. The spindle 10 is placed on a spindle support 9, and the spindle 10 is conveyed to the shooting station for image acquisition via the spindle conveying mechanism 2.

[0038] In this embodiment, there are four side-shooting cameras 5, which are used to shoot towards the side wall of the spindle 10, and three oblique-shooting cameras 7, which are used to shoot towards the paper tube 13 of the spindle 10. During shooting, the multiple side-shooting cameras 5 are evenly distributed on the circumferential direction outside the side wall of the spindle 10, and the multiple oblique-shooting cameras 7 are evenly distributed above the spindle 10.

[0039] The strip light source 3 can form a strip-shaped emitted beam. When the end-face camera 4 takes a picture, the emitted beam is perpendicular to the axis of the spindle 10 and illuminates one end face of the spindle 10. The end face of the spindle 10 is located within the emitted beam, such as... Figure 4 As shown, the filaments 11 on the end face of the spindle 10 can be illuminated by the emitted light beam. An end-face camera 4, positioned facing the end face of the spindle 10, captures an image of the end face. In this image, the filaments 11 are highlighted compared to the end face of the spindle 10, thus creating a distinct feature for subsequent image analysis to identify defects. For example... Figure 8 As shown, when the strip light source 3 is not turned on, the hair defects in the image captured by the end-face camera 4 are not obvious, and it is basically impossible to extract the hair defects using existing image analysis and recognition technologies alone. However, as Figure 9 , 10 As shown, when the strip light source 3 is turned on, the fuzzy defects in the image captured by the end-face camera 4 are very obvious and can be clearly seen even with the naked eye, see areas D and E in the figure. In addition, as... Figure 11 As shown, when the bar light source 3 is not turned on, no obvious defects in the appearance of the spindle 10 are found in the image captured by the end face camera 4. However, when the bar light source 3 is turned on, as shown... Figure 12 As shown, because the end face of the spindle 10 has protruding ridges, when illuminated by the emitted light beam, it forms obvious bright and shadow areas (part B in the figure). This indicates that the spindle 10 has forming defects, and the identification features are very obvious. However, without the illumination of the strip light source 3, this forming defect area does not have sufficient color contrast, making it impossible to detect accurately. Traditional two-dimensional imaging is unable to capture the depth information of these defects.

[0040] In this embodiment, there are four strip light sources 3, arranged in pairs to illuminate the upper and lower end faces of the spindle 10. Two strip light sources 3 are used to illuminate each end face of the spindle 10, and they are symmetrically arranged on the left and right sides of the central axis movement trajectory plane L of the spindle 10.

[0041] There are two shooting stations for acquiring images of the end face of the spindle 10, and each single image captured by the end face camera 4 at each shooting station includes a 1 / 4 image of the end face of the spindle 10. When acquiring images of the end face of the spindle 10 using the defect image acquisition device for the spindle 10 as described above, a strip light source 3 illuminates the end face of the spindle 10, and the end face camera 4 acquires the image of the end face of the spindle 10. Specifically, as... Figure 5 As shown, the spindle 10 is conveyed forward by the spindle 10 conveying mechanism 2. The area within the box in section C of the figure represents the frame area of ​​the end-face camera 4. Figure 6 As shown, after the spindle 10 is conveyed to the first shooting station by the spindle 10 conveying mechanism 2, the strip light source 3 illuminates the end face of the spindle 10, and the end face camera 4 captures the front 1 / 4 area of ​​the end face of the spindle 10. Figure 7 As shown, the spindle 10 is then conveyed forward to the second shooting station, where the end-face camera 4 captures images of the rear 1 / 4 area of ​​the spindle 10's end face. By acquiring images of the spindle 10's end face using the above method, the paper tube 13 of the spindle 10 can be avoided from obstructing the view of the end-face camera 4, ensuring that key features and potential defects of the spindle 10 can be captured in each acquisition.

[0042] The present invention relates to a defect image acquisition device and method for a wire spindle 10. Through optimized design of the light source, defects on the surface of the wire spindle 10 become more apparent under the illumination of the strip light source 3, thus being clearly displayed in the image captured by the camera. This improvement not only helps to improve the accurate identification of these defects but also significantly enhances their visibility. The unique design of the strip light source 3 allows defects formed on the end face of the wire spindle 10 to exhibit obvious highlight or shadow features with the aid of illumination. This effect is particularly prominent, especially in identifying small defects, such as fuzzy threads, tiny stray threads, and some forming defects of the wire spindle 10, where the identification capability is significantly improved, greatly enhancing the accuracy and efficiency of detection. In this way, defect observation and analysis become more efficient, providing important support for subsequent quality control.

[0043] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the protection scope of this invention.

Claims

1. A device for acquiring images of defects in silk spindles, characterized in that, include: A strip light source is capable of forming a strip-shaped emitted beam. The emitted beam is perpendicular to the axis of the spindle and illuminates one end face of the spindle, which is located within the emitted beam. An end-face camera, positioned facing the end face of the spindle, is used to acquire images of the spindle's end face.

2. The image acquisition device for defects in silk spindles as described in claim 1, characterized in that: There are multiple strip light sources, which are used to irradiate the upper and lower end faces of the wire spindle.

3. The image acquisition device for defects in silk spindles as described in claim 1, characterized in that: Multiple strip light sources are used to illuminate each end face of the wire spindle.

4. The image acquisition device for defects in silk spindles as described in claim 3, characterized in that: Multiple strip light sources are evenly distributed around the circumference of the spindle.

5. The image acquisition device for defects in silk spindles as described in any one of claims 1-4, characterized in that: The image acquisition device for defects in the spindle also includes a spindle conveying mechanism, which is used to move the spindle to the shooting station for shooting.

6. The image acquisition device for defects in silk spindles as described in claim 5, characterized in that: Two end-face cameras are used to photograph the same end face of the spindle, and the two end-face cameras are symmetrically arranged on the left and right sides of the moving trajectory plane of the spindle's central axis.

7. The image acquisition device for defects in silk spindles as described in claim 6, characterized in that: There are two shooting stations; at each shooting station, each end-face camera captures a single image including a 1 / 4 image of the spindle end face.

8. The image acquisition device for defects in silk spindles as described in claim 5, characterized in that: The spindle conveying mechanism is a conveyor belt or a conveyor roller.

9. The image acquisition device for defects in silk spindles as described in any one of claims 1-4, characterized in that: The image acquisition device for defects in the spindle also includes a side-view camera for taking pictures toward the side wall of the spindle and an auxiliary light source.

10. The image acquisition device for defects in a wire spindle as described in claim 9, characterized in that: The side-viewing cameras are multiple, and the multiple side-viewing cameras are evenly distributed in the circumferential direction on the outer side wall of the spindle.

Images