Automatic identification device for surface defects of panel glass
By combining a multispectral light source component and a dynamic optical path adjustment mechanism, the bottleneck of panel glass inspection equipment under multiple defect types and environmental vibrations has been solved, realizing efficient and accurate fully automated inspection and improving inspection efficiency and compatibility.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- DONGGUAN MOORE INTELLIGENT TECHNOLOGY CO LTD
- Filing Date
- 2025-05-26
- Publication Date
- 2026-05-08
AI Technical Summary
Existing panel glass inspection equipment suffers from problems such as low inspection efficiency, high missed detection rate, high labor costs, poor inspection compatibility, insufficient positioning accuracy, and low degree of automation. In particular, its inspection performance is poor under multiple defect types and environmental vibration conditions.
It adopts a combination of multispectral light source components, optical path adjustment mechanism, image acquisition module and mechanical support, combined with three-dimensional adjustment seat, servo drive and air-floating shock-absorbing base to achieve multi-dimensional detection and environmental anti-interference. By intelligently selecting the best detection spectrum through multispectral light source and dynamically adjusting the incident angle, it ensures the accurate positioning and stable scanning of image acquisition module.
It significantly improves the efficiency of defect detection on the surface of panel glass, enhances detection efficiency and compatibility, reduces the false judgment rate and production cost, realizes a fully automated detection process, and adapts to high-precision detection under different material and environmental conditions.
Smart Images
Figure CN224216593U_ABST
Abstract
Description
Technical Field
[0001] This utility model relates to the field of testing equipment technology, specifically to an automatic identification device for defects on the surface of panel glass. Background Technology
[0002] In consumer electronics, flat panel displays, and architectural glass, surface defects in panel glass (such as cracks, bubbles, and scratches) directly affect product quality and reliability. Therefore, high-precision surface defect detection is a critical step in the production process. Traditional detection methods rely on manual visual inspection or single-light source imaging, resulting in low detection efficiency, high false negative rates, and high labor costs. Although existing automated inspection equipment incorporates machine vision technology, it generally suffers from the following shortcomings: First, the light source system often uses a single spectrum, making it difficult to simultaneously address the detection needs of both surface defects and internal defects. For example, the sensitivity of the ultraviolet band to microcracks and the penetrating advantage of the infrared band to interlayer impurities are not fully utilized. Second, the fixed optical path adjustment mechanism cannot dynamically adjust the incident angle according to the surface roughness of the glass, leading to poor compatibility in detecting different glass materials. Third, the positioning accuracy of the image acquisition module is insufficient, and there is a lack of effective vibration isolation measures, which can easily cause image blurring in high-speed production line environments. Fourth, workpiece positioning relies on manual placement, resulting in low automation and difficulty in adapting to the full-width inspection needs of large-size glass. Utility Model Content
[0003] In order to overcome the shortcomings of existing technical solutions, this utility model provides an automatic identification device for surface defects of panel glass, which can effectively solve the problems mentioned in the background art.
[0004] The technical solution adopted by this utility model to solve its technical problem is:
[0005] An automatic defect identification device for panel glass includes a detection platform, a light source system, an image acquisition module, an image processing unit, and an adjustable mechanical support. The detection platform is equipped with a three-dimensional adjustment base. The light source system includes a multispectral light source component and an optical path adjustment mechanism. The multispectral light source component integrates a visible light source, an infrared light source, and an ultraviolet light source through an optical fiber array. The optical path adjustment mechanism includes a rotating base installed on the edge of the detection platform. A liftable reflective prism group is installed on the rotating base. The reflective prism group is arranged opposite to the light-emitting end of the multispectral light source component.
[0006] The image acquisition module is mounted on the front end of the cantilever beam of the adjustable mechanical support via a three-dimensional adjustment base. The cantilever beam is slidably connected to a linear guide rail, which is controlled by a servo drive component. The image processing unit has a built-in edge computing chip and communicates with the image acquisition module via a gigabit Ethernet interface.
[0007] As a further description of the above technical solution, the three-dimensional adjustment seat includes an X-axis slide rail, a Y-axis slide rail, and a Z-axis lead screw arranged perpendicularly to each other. The image acquisition module adjusts the vertical height through the Z-axis lead screw and finely adjusts the horizontal position through the X-axis slide rail and the Y-axis slide rail.
[0008] As a further description of the above technical solution, the servo drive component includes a rack arranged parallel to the linear guide rail and a gear meshing with the rack, the gear being connected to the servo motor for transmission.
[0009] As a further description of the above technical solution, a vacuum adsorption platform is provided on the surface of the detection platform, and an infrared through-beam sensor is arranged around the edge of the vacuum adsorption platform. The infrared through-beam sensor is electrically connected to the servo drive component.
[0010] As a further description of the above technical solution, the adjustable mechanical support is provided with an air-floating shock-absorbing base at the bottom, and the air-floating shock-absorbing base includes a rubber vibration isolation layer and an air pressure balance valve.
[0011] As a further description of the above technical solution, the multispectral light source component is configured with a wavelength switching controller, which is connected to the visible light source, infrared light source or ultraviolet light source through a relay group.
[0012] As a further description of the above technical solution, the bottom of the testing platform is provided with a vibration isolation support assembly. The vibration isolation support assembly includes a support base fixedly connected to the testing platform, a rubber vibration isolation pad disposed below the support base, and a leveling bolt penetrating the support base and the rubber vibration isolation pad.
[0013] Compared with the prior art, the beneficial effects of this utility model are:
[0014] The automatic defect identification device for panel glass of this utility model has at least one of the following beneficial effects during use:
[0015] Multi-dimensional technological innovation significantly improves the efficiency of defect detection on panel glass surfaces. The multi-spectral light source assembly integrates visible light, infrared, and ultraviolet light sources, and, in conjunction with a wavelength switching controller, can intelligently select the optimal detection spectrum for different defect types (such as microcracks and internal impurities), improving the defect detection rate compared to traditional single-light source solutions. The rotating base and height-adjustable reflective prism assembly of the optical path adjustment mechanism allow for dynamic adjustment of the incident angle from 0-90°, adapting to the detection needs of glass surfaces with varying roughness and improving compatibility. The three-dimensional adjustment base and servo drive assembly ensure precise positioning and stable scanning of the image acquisition module, resulting in high pixel uniformity across the entire detection area. The air-bearing shock-absorbing base and vibration isolation support components work together to reduce the impact of environmental vibrations. The entire device achieves a fully automated detection process, improving single-batch detection efficiency compared to manual methods, reducing the false judgment rate, shortening maintenance time, and significantly reducing production costs. Attached Figure Description
[0016] Figure 1 This is a schematic diagram of the overall structure of an automatic identification device for surface defects of panel glass according to the present invention;
[0017] Figure 2 This is a schematic diagram of the first side view of the automatic identification device for surface defects of panel glass according to the present invention;
[0018] Figure 3 This is a schematic diagram of the second side structure of an automatic identification device for surface defects of panel glass according to the present invention;
[0019] Figure 4 This is a schematic diagram of the detection platform of an automatic identification device for surface defects of panel glass according to the present invention.
[0020] Numbering on the map:
[0021] 1. Testing platform; 101. Rotating base; 102. Light source system; 103. Adjustable mechanical support; 104. Vibration isolation support assembly; 105. Reflecting prism assembly; 106. Vacuum adsorption platform; 107. Multispectral light source assembly; 108. Air flotation shock absorption base; 2. Image acquisition module; 201. Image processing unit; 3. Three-dimensional adjustment seat; 301. Servo drive assembly; 302. Z-axis lead screw; 303. Cantilever beam; 304. X-axis slide rail; 305. Y-axis slide rail. Detailed Implementation
[0022] The technical solutions of the present utility model will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present utility model, and not all embodiments. Based on the embodiments of the present utility model, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of the present utility model.
[0023] like Figure 1-4 As shown, this utility model provides an automatic identification device for surface defects of panel glass, including a detection platform 1, a light source system 102, an image acquisition module 2, an image processing unit 201, and an adjustable mechanical support 103. The detection platform 1 is provided with a three-dimensional adjustment seat 3. The light source system 102 includes a multispectral light source component 107 and an optical path adjustment mechanism. The multispectral light source component 107 integrates a visible light source, an infrared light source, and an ultraviolet light source through an optical fiber array. The optical path adjustment mechanism includes a rotating base 101 installed on the edge of the detection platform 1. A liftable reflective prism group 105 is provided on the rotating base 101. The reflective prism group 105 is arranged opposite to the light-emitting end of the multispectral light source component 107.
[0024] In this embodiment, the panel glass to be inspected is placed on the vacuum adsorption platform 106 of the inspection platform 1. After the system is started, the infrared through-beam sensor surrounding the edge emits an infrared signal. When the glass completely covers the inspection area and triggers the sensor signal, it indicates that the workpiece positioning is complete. At this time, the vacuum adsorption system is activated, generating negative pressure through the uniformly distributed adsorption holes to tightly adhere the glass surface to the inspection platform 1, avoiding image acquisition deviations caused by workpiece displacement during the inspection process.
[0025] The multi-modal light source excites the multispectral light source assembly 107, which integrates three light sources—visible (400-700nm), infrared (800-1600nm), and ultraviolet (200-400nm)—through a fiber optic array. A wavelength switching controller receives instructions from the image processing unit 201 and selectively activates the target light source via a relay group (e.g., activating the ultraviolet light source when detecting cracks, and switching to the infrared light source when detecting internal impurities). The light emitted by the light source is converged by the fiber optic bundle and then transmitted to the optical path adjustment mechanism.
[0026] The image acquisition module is mounted on the front end of the cantilever beam 303 of the adjustable mechanical support 103 via a three-dimensional adjustment seat 3. The cantilever beam 303 is slidably connected to a linear guide rail, which is controlled by a servo drive component 301. The image processing unit 201 has a built-in edge computing chip and is connected to the image acquisition module via a gigabit Ethernet interface.
[0027] The Z-axis lead screw 302 drives the camera to move vertically, enabling precise adjustment of the object distance (e.g., step adjustment within a 10-50cm range). The X / Y-axis slide rail 305 uses a high-precision linear guide rail, in conjunction with a handwheel or servo motor (optional depending on the embodiment), to achieve horizontal position fine-tuning at the ±0.1mm level, ensuring that the camera's optical axis is perpendicular to the glass surface.
[0028] The reflective prism assembly 105, dynamically adjustable, is mounted on a rotating base 101 at the edge of the inspection platform 1. This base can rotate 360°, causing the reflective prism assembly 105 (composed of 2-3 independently adjustable prisms) above to adjust its angle. For example, when inspecting scratches on a glass surface, the prism assembly directs the light source at a 45° angle to enhance the diffuse reflection signal of surface defects; when inspecting internal bubbles, the prism assembly switches to vertical incidence, utilizing the difference in transmitted light for imaging. Dynamic matching of the optical path is achieved by controlling the stepper motor (accuracy ±0.1°) of the rotating base 101 and the prism lifting mechanism (accuracy ±0.5mm) via a PLC.
[0029] This embodiment integrates multispectral light sources, dynamic optical path adjustment, precision mechanical positioning, and edge computing technologies to construct a closed-loop "excitation-acquisition-processing" system, overcoming the bottlenecks of traditional detection devices in terms of adaptability to multiple defect types, dynamic detection accuracy, and environmental interference resistance. Actual measurements show that the system achieves an average detection rate of 99.7% for typical defects (cracks, bubbles, scratches), a false positive rate of <0.3%, and a detection speed of 0.8m. 2 With a speed of [speed per minute], all indicators are superior to similar equipment, demonstrating significant industrial application value.
[0030] Furthermore, the three-dimensional adjustment seat 3 includes an X-axis slide rail 304, a Y-axis slide rail 305, and a Z-axis lead screw 302 arranged perpendicularly to each other. The image acquisition module adjusts the vertical height through the Z-axis lead screw 302 and finely adjusts the horizontal position through the X-axis slide rail 304 and the Y-axis slide rail 305.
[0031] The servo drive and dynamic scanning image acquisition module 2 is mounted on the front end of the cantilever beam 303 of the adjustable mechanical support 103, and the cantilever beam 303 is slidably connected to the linear guide rail. The servo motor in the servo drive assembly 301 drives the gear to rotate, meshing with the rack parallel to the linear guide rail, realizing the linear motion of the cantilever beam 303 (speed adjustable from 0.1-10cm / s). When inspecting large-size glass, the system starts the scanning mode: the motor drives the camera to reciprocate along the X-axis at a constant speed, while the Z-axis adjustment seat dynamically compensates according to the preset height curve to ensure that the field of view (FOV) covers the entire glass surface (e.g., a detection range of 2m×3m).
[0032] Furthermore, the servo drive assembly 301 includes a rack arranged parallel to the linear guide rail and a gear meshing with the rack, the gear being connected to the servo motor for transmission.
[0033] The edge computing real-time image acquisition module (e.g., a 12-megapixel industrial camera at 30fps) transmits the raw images to the image processing unit 201 via gigabit Ethernet. The built-in edge computing chip (e.g., NVIDIA Jetson AGX) performs preprocessing (noise reduction, contrast enhancement), feature extraction (Canny edge detection, HOG features), and defect classification (based on deep learning models, such as an improved YOLOv8) in real time. Processing latency is <50ms, meeting the real-time inspection requirements of high-speed production lines (cycle time <2s / piece).
[0034] Furthermore, a vacuum adsorption platform 106 is provided on the surface of the detection platform 1, and an infrared beam sensor is arranged around the edge of the vacuum adsorption platform 106. The infrared beam sensor is electrically connected to the servo drive assembly 301.
[0035] Furthermore, the adjustable mechanical support 103 is provided with an air-floating shock-absorbing base 108 at its bottom, which includes a rubber vibration isolation layer and an air pressure balance valve.
[0036] Furthermore, the multispectral light source assembly 107 is equipped with a wavelength switching controller, which is connected to the visible light source, infrared light source, or ultraviolet light source via a relay group.
[0037] The multispectral image fusion system can generate single-spectral or fused images according to detection requirements: visible light images are used to identify surface scratches and stains; infrared images penetrate the glass surface to detect interlayer defects or temperature anomalies; and ultraviolet images highlight fluorescent markers or microcracks. The image processing unit 201 fuses multi-source data through a spectral registration algorithm (error < 0.5 pixels) to form a three-dimensional feature map containing surface and subsurface defect information, improving the defect detection rate (≥ 99.5%).
[0038] Furthermore, the bottom of the testing platform 1 is provided with a vibration isolation support assembly 104. The vibration isolation support assembly 104 includes a support base fixedly connected to the testing platform 1, a rubber vibration isolation pad disposed below the support base, and a leveling bolt penetrating the support base and the rubber vibration isolation pad.
[0039] The adjustable mechanical support 103 for vibration isolation and stable imaging has an air-floating shock-absorbing base 108 at its bottom that absorbs environmental vibrations through a rubber vibration isolation layer (natural frequency 5-10Hz). An air pressure balance valve adjusts the air pressure inside the airbag in real time to ensure the support's levelness deviation is <0.05°. The vibration isolation support assembly 104 at the bottom of the detection platform 1 further isolates low-frequency vibrations (such as 10-50Hz) transmitted from the ground through leveling bolts (accuracy ±0.01mm / m) and rubber vibration isolation pads (damping ratio 0.2-0.3), preventing image blurring caused by vibration.
[0040] It will be apparent to those skilled in the art that this invention is not limited to the details of the exemplary embodiments described above, and that it can be implemented in other specific forms without departing from the spirit or essential characteristics of this invention. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of this invention is defined by the appended claims rather than the foregoing description. Thus, it is intended that all variations falling within the meaning and scope of equivalents of the claims be included within this invention. No reference numerals in the claims should be construed as limiting the scope of the claims.
Claims
1. An automatic defect identification device for panel glass surface, comprising a detection platform, a light source system, an image acquisition module, an image processing unit, and an adjustable mechanical support, characterized in that: The detection platform is equipped with a three-dimensional adjustment base. The light source system includes a multispectral light source component and an optical path adjustment mechanism. The multispectral light source component integrates visible light, infrared light and ultraviolet light sources through an optical fiber array. The optical path adjustment mechanism includes a rotating base installed on the edge of the detection platform. A liftable reflective prism group is provided on the rotating base. The reflective prism group is arranged opposite to the light-emitting end of the multispectral light source component. The image acquisition module is mounted on the front end of the cantilever beam of the adjustable mechanical support via a three-dimensional adjustment base. The cantilever beam is slidably connected to a linear guide rail, which is controlled by a servo drive component. The image processing unit has a built-in edge computing chip and communicates with the image acquisition module via a gigabit Ethernet interface.
2. The automatic defect identification device for panel glass surface according to claim 1, characterized in that: The three-dimensional adjustment seat includes an X-axis slide rail, a Y-axis slide rail, and a Z-axis lead screw arranged perpendicularly to each other. The image acquisition module adjusts the vertical height through the Z-axis lead screw and finely adjusts the horizontal position through the X-axis and Y-axis slide rails.
3. The automatic defect identification device for panel glass surface according to claim 1, characterized in that: The servo drive assembly includes a rack arranged parallel to the linear guide rail and a gear meshing with the rack, the gear being connected to a servo motor for transmission.
4. The automatic identification device for surface defects of panel glass according to claim 1, characterized in that: The detection platform surface is provided with a vacuum adsorption platform, and an infrared through-beam sensor is arranged around the edge of the vacuum adsorption platform. The infrared through-beam sensor is electrically connected to the servo drive component.
5. The automatic identification device for surface defects of panel glass according to claim 1, characterized in that: The adjustable mechanical support is equipped with an air-floating shock-absorbing base at its bottom, which includes a rubber vibration isolation layer and an air pressure balance valve.
6. The automatic identification device for surface defects of panel glass according to claim 1, characterized in that: The multispectral light source assembly is equipped with a wavelength switching controller, which is connected to the visible light source, infrared light source, or ultraviolet light source via a relay group.
7. The automatic identification device for surface defects of panel glass according to claim 1, characterized in that: The bottom of the testing platform is provided with a vibration isolation support assembly, which includes a support base fixedly connected to the testing platform, a rubber vibration isolation pad disposed below the support base, and a leveling bolt passing through the support base and the rubber vibration isolation pad.