Glass bottle body defect detection device
This glass bottle inspection device, which uses a turntable to rotate glass bottles and combines an image acquisition unit with a backlight, solves the problem of low efficiency in traditional manual inspection, achieving efficient and accurate defect identification. It is suitable for automated inspection in glass bottle production lines.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-15
- Publication Date
- 2026-03-17
AI Technical Summary
Traditional methods of manually inspecting glass bottles for defects are inefficient, have a high error rate, and are labor-intensive, making them unsuitable for the needs of modern large-scale production.
A turntable is used to rotate the glass bottle 360°. Combined with an industrial camera and lens image acquisition device and a backlight, the bottle is photographed every 60° rotation, for a total of 6 times, to obtain images of the bottle from various angles. This data is then used in conjunction with a deep learning model for defect identification and judgment.
It enables comprehensive inspection of a large number of glass bottles in a short time, improves inspection efficiency, reduces inspection discrepancies, and meets the needs of large-scale production.
Smart Images
Figure CN224004977U_ABST
Abstract
Description
Technical Field
[0001] This utility model relates to the field of glass bottle inspection technology, and in particular to a glass bottle body defect detection device. Background Technology
[0002] During the glass bottle production process, defects such as cracks, bubbles, scratches, deformation, and impurities may appear on the bottle body. Traditional manual inspection methods suffer from low efficiency, high error rate, and high labor intensity, making it difficult to meet the needs of modern large-scale production. Therefore, a glass bottle defect detection device is proposed to solve this problem. Utility Model Content
[0003] This utility model addresses the shortcomings of existing technologies by providing the following technical solution:
[0004] A glass bottle defect detection device includes a turntable for rotating the glass bottle 360°. An image acquisition device for capturing images of the glass bottle is provided on one side of the turntable, and a backlight is provided on the opposite side of the image acquisition device.
[0005] The image acquisition device includes an industrial camera and lens, primarily used for photographing the glass bottle. A backlight provides suitable illumination for the industrial camera, allowing defects on the bottle to be better captured in the lens. The bottle is placed on a turntable, which rotates it one full circle. The image acquisition device takes one picture every 60 degrees of rotation, for a total of six times. This allows the device to progressively capture images of the bottle from various angles, as if unfolding the bottle into a flat surface, thus comprehensively recording its appearance.
[0006] As an improvement to the above technical solution, the backlight source includes a flat panel light source and a striped light source.
[0007] As an improvement to the above technical solution, the flat panel light source, in conjunction with the image acquisition device, is used to detect cracks, bubbles, scratches, and dirt defects on the glass bottle body; the striped light source, in conjunction with the image acquisition device, is used to detect deformation and impurity defects on the bottle body.
[0008] As an improvement to the above technical solution, the image acquisition device is set vertically to the glass bottle, and the glass bottle is set parallel to the backlight.
[0009] The beneficial effects of this utility model are:
[0010] By placing glass bottles on a turntable, the turntable rotates the bottles one full circle. Every 60° of rotation, the image acquisition device takes one picture, for a total of six times. The image acquisition device can gradually capture images of the bottle from various angles, as if the bottle is unfolded into a flat plane, thus comprehensively recording the appearance of the bottle. Therefore, defects at any location on the bottle can be captured by the image acquisition device, allowing for the inspection of a large number of glass bottles in a short time. This greatly improves the inspection efficiency on the production line, meets the needs of large-scale production, and avoids the inspection differences caused by different personnel conditions in manual inspection. Attached Figure Description
[0011] Figure 1 This is a schematic diagram of the overall structure of this utility model.
[0012] Figure labels: 10, image acquisition device; 20, glass bottle; 30, backlight; 40, turntable. Detailed Implementation
[0013] To make the objectives, technical solutions, and advantages of this utility model clearer, the present utility model will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present utility model and are not intended to limit the present utility model.
[0014] A glass bottle defect detection device includes a turntable 40, which is used to drive the glass bottle 20 to rotate 360°. An image acquisition device 10 for capturing images of the glass bottle 20 is provided on one side of the turntable 40, and a backlight 30 is provided on the opposite side of the image acquisition device 10.
[0015] Image acquisition device 10 includes an industrial camera and lens, primarily used for photographing the body of glass bottle 20. Backlight 30 provides suitable illumination for the industrial camera, allowing defects on the glass bottle 20 to be better displayed in the lens. The glass bottle 20 is placed on a turntable 40, which rotates the bottle 20 one full circle. Image acquisition device 10 takes one picture every 60° of rotation, for a total of six pictures. This allows image acquisition device 10 to gradually capture images of the bottle from various angles, as if unfolding the bottle into a flat surface, thus comprehensively recording the bottle's appearance. The images acquired by image acquisition device 10... After the image of the glass bottle 20 is cropped to a suitable size, it is fed into a deep learning pixel segmentation model for recognition. Blob analysis is performed on the model prediction results to obtain information such as the length and contrast of various defects. Then, conditional filtering and judgment are performed on this information. NG products are sent to the waste recycling facility, while qualified products are transported to the product waiting area. With the cooperation of image acquisition device 10 and backlight 30, a large number of glass bottles 20 can be inspected in a short time, which greatly improves the inspection efficiency on the production line and meets the needs of large-scale production. The bottom of the turntable 40 can be equipped with a driving component, which can be a motor.
[0016] In one embodiment, reference Figure 1 The backlight 30 includes a flat panel light source and a striped light source. The flat panel light source works in conjunction with the image acquisition device 10 to detect defects such as cracks, bubbles, scratches, and dirt on the body of the glass bottle 20. The striped light source works in conjunction with the image acquisition device 10 to detect defects such as deformation and impurities on the bottle body. The distance between the image acquisition device 10 and the center line of the glass bottle 20 is A1, where A1 is 300mm. The distance between the backlight 30 and the center line of the glass bottle 20 is A2, where A2 is 160mm. The distance between the bottom of the glass bottle 20 and the bottom of the backlight 30 is A3, where A3 is 70mm. The distance between the position of the center line of the image acquisition device 10 at the body of the glass bottle 20 and the bottom of the glass bottle 20 is A4, where A4 is 80mm. This allows defects at any position on the body of the glass bottle 20 to be captured by the industrial camera.
[0017] In one embodiment, the image acquisition device 10 is set vertically to the glass bottle 20 to obtain the clearest and most complete image of the bottle. The glass bottle 20 is set parallel to the backlight 30 to provide stable and uniform background light to the glass bottle 20. When the backlight 30 is lit, the light can pass through the glass bottle 20 to provide good lighting conditions for the image acquisition device 10. This parallel setting allows the light to illuminate the glass bottle 20 as evenly as possible, reducing the impact of shadows and reflections on image quality.
[0018] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it.
Claims
1. A glass bottle body defect detection apparatus, characterized by, The device comprises a rotating disc (40) for rotating a glass bottle (20) by 360 degrees, one side of the rotating disc (40) is provided with an image collector (10) for shooting the bottle body of the glass bottle (20), and the opposite side of the image collector (10) is provided with a backlight source (30).
2. The glass bottle body defect detection apparatus according to claim 1, characterized in that: The backlight source (30) comprises a flat light source and a stripe light source.
3. The glass bottle body defect detection apparatus according to claim 2, characterized in that: The flat light source cooperates with the image collector (10) to detect cracks, bubbles, scratches and dirt defects of the bottle body of the glass bottle, and the stripe light source cooperates with the image collector (10) to detect deformation and impurity defects of the bottle body.
4. The glass bottle body defect detection apparatus according to claim 1, characterized by: The image collector (10) is arranged perpendicularly to the glass bottle (20), and the glass bottle (20) is arranged in parallel to the backlight source (30).