In the field of coil coating production, micron-level three-dimensional defects such as pinholes and agglomerations directly affect the barrier, insulation, and appearance properties of products. Traditional 2D detection solutions struggle to meet the requirements of accurate detection under high-speed production. WeLinkirt's 3D fusion point cloud online detection solution provides an effective way to solve this problem.
Scenario: In the coil coating production of the chemical and material industries, product quality is of utmost importance. The quality of the coating is directly related to the barrier properties, insulation properties, and appearance of the product. For example, in the production of food packaging coils, a good coating can prevent food from being affected by the external environment and extend the shelf-life of food; in the production of coils for electronic devices, the insulation properties of the coating are the key to ensuring the normal operation of the equipment. However, during the coating production process, micron-level three-dimensional defects such as pinholes (local missing coating depressions) and agglomerations (paint lumps protrusions) occur from time to time. If these defects cannot be detected and handled in time, they will seriously affect the performance and quality of the product, and further affect the market competitiveness of the enterprise.
Pain Points: Why It's Difficult
From the dimension of detection difficulty, these defects are extremely small in size, usually on the micron level. For example, the minimum detectable defects of pinholes and agglomerations are greater than 50μm. Detecting such tiny defects requires high-precision equipment and technology. Moreover, in a 2D grayscale image, depressions, protrusions, and surface stains may show similar brightness and darkness, which makes it difficult for 2D imaging to distinguish the types of defects and even more difficult to determine the real severity of the defects. For example, a tiny pinhole may look very similar to a small stain on the surface in a 2D image, making it difficult for the detection system to accurately judge.
From the dimension of production speed, the production line usually runs at a high speed of 120-400 m/min. At such a high speed, any detection must complete a full-width scan without stopping the machine. This requires the detection system to have the ability to respond quickly and process data efficiently. If the detection speed cannot keep up with the production speed, it will lead to missed detections. Traditional 2D detection solutions have limited processing speed under high-speed production and are difficult to meet the needs of real-time detection.
From the dimension of false-positive rate analysis, the previous 2D solutions had a high false-positive rate for pinholes and agglomerations. Real defects smaller than the threshold were released as stains, while harmless stains frequently triggered machine shutdowns. For example, in a production process, about 10% of real pinhole defects were misjudged as stains and not detected, and at the same time, 15% of harmless stains caused unnecessary machine shutdowns. This not only reduces production capacity but also affects the quality stability of the product. The root cause is that the 2D solution lacks height information and cannot accurately distinguish the real types and severity of defects.
Technical Principle
The 3D fusion point cloud solution adopted by WeLinkirt focuses on reconstructing the three-dimensional morphology of the coating surface under high-speed coil online conditions. The solution obtains point cloud data of the coating surface through advanced hardware equipment, which contains the three-dimensional coordinate information of the coating surface. Then, specific algorithms are used to process and analyze these point cloud data, and the depressions of pinholes and the protrusions of agglomerations are directly expressed as positive and negative deviations in the height direction. For example, pinholes are represented as negative deviations in the height direction, while agglomerations are represented as positive deviations.
Compared with the traditional 2D method, the 3D fusion point cloud solution has obvious advantages. The 2D method can only obtain the planar information of the image and lacks height-dimension information, so it is difficult to distinguish the types of defects and determine the severity. The 3D solution can identify defects more accurately by introducing height information. Combined with AI-AOI judgment, the system can not only identify the presence or absence of defects but also distinguish the concave/convex types and quantify the depth or height. In the high-speed coil online detection at 120-400 m/min, defects larger than 50μm can be stably detected.
Typical Application Scenarios
- Food packaging coil coating detection: In the production of food packaging coils, pinhole defects may allow external microorganisms and oxygen to enter the food, affecting the shelf-life of the food. During detection, the difficulty lies in the small size of the pinholes, which may be confused with tiny impurities on the coating surface. The 3D fusion point cloud solution can accurately distinguish pinholes from impurities by reconstructing the three-dimensional morphology and quantify the depth of the pinholes to ensure the barrier properties of the product.
- Coil coating detection for electronic devices: Electronic devices have extremely high requirements for the insulation properties of coil coatings. Agglomeration defects may damage the insulation structure of the coating, leading to equipment short-circuits and other failures. During detection, due to the high production line speed and the hidden nature of agglomeration defects, the traditional method is prone to missed detections. The 3D fusion point cloud solution can accurately identify the protrusions of agglomerations at high speed and quantify their height to ensure the insulation properties of the product.
- Building material coil coating detection: The appearance quality of building material coils is crucial. Pinholes and agglomeration defects will affect the aesthetics and corrosion resistance of the coils. During detection, there may be some natural textures and slight stains on the coating surface, which will interfere with the 2D detection. The 3D fusion point cloud solution combines 2D information to distinguish defects from normal textures and stains and classify the defects.
- Automotive interior coil coating detection: Automotive interiors have certain requirements for the comfort and safety of coils. Pinhole defects may cause the leakage of harmful substances, while agglomeration defects may affect the flatness of the interior. During detection, due to the complex colors and textures of automotive interior coils, the traditional method is difficult to accurately detect. The 3D fusion point cloud solution can accurately identify defects in a complex background and classify them according to depth and height.
Implementation Case
There is a medium-sized coil coating factory with a production line speed between 120-400 m/min. Before introducing WeLinkirt's 3D fusion point cloud online detection solution, the factory used the traditional 2D detection solution. At that time, the false-positive rate of pinholes and agglomerations was as high as 30%, the missed-detection rate of real defects was 15%, and the number of false machine shutdowns caused by harmless stains reached about 20 times per month, seriously affecting production capacity and product quality.
During the implementation of WeLinkirt's solution, the technical team first conducted a detailed investigation and analysis of the factory's production line and customized the system configuration according to the characteristics and needs of the production line. During the installation and commissioning process, it was ensured that the system was smoothly integrated with the existing production line. After a period of trial operation and optimization, the system was officially put into use.
The 3D fusion point cloud technology makes coil coating defects nowhere to hide, improving both production capacity and yield.
WeLinkirt's Solution and Product
WeLinkirt's 3D fusion point cloud online detection solution is centered around advanced 3D point cloud reconstruction technology. The solution includes high-precision hardware equipment for obtaining point cloud data of the coating surface and a powerful algorithm and software system for processing and analyzing the data. Combined with AI-AOI judgment, the system can stably detect defects larger than 50μm under high-speed coil online conditions of 120-400 m/min. At the same time, the system can also quantify the depth/height of the defects and classify them according to the real severity, avoiding false machine shutdowns caused by harmless stains. In addition, the solution is also integrated with 2D information, taking into account both the morphology and appearance and reducing the confusion between concavities/convexities and stains.
Quantitative Results: After the implementation of the solution, the factory achieved online classification detection of pinholes and agglomerations at a full-speed production of 120-400 m/min. The detection rate of defects larger than 50μm increased from the original 85% to 98%, and the missed-detection rate decreased from 15% to 2%. The number of false machine shutdowns caused by harmless stains decreased from about 20 times per month to less than 3 times, a reduction of about 85%. Abnormalities in the coating process can be quickly traced according to the defect classification. The production capacity increased by 20%, and the yield increased from the original 80% to 95%. These data fully demonstrate the significant effectiveness of WeLinkirt's solution in improving production capacity and product quality.
FAQ
How does WeLinkirt solve the detection problems of pinholes and agglomerations in coil coatings?
WeLinkirt adopts a 3D fusion point cloud solution to reconstruct the three-dimensional morphology of the coating under high-speed coil online conditions. Combined with AI-AOI judgment, it can distinguish concave/convex types, quantify depth or height, reduce false-positives, and thus improve production capacity and yield, effectively solving the detection problems.
What are the advantages of WeLinkirt's 3D fusion point cloud detection solution?
This solution can stably detect defects larger than 50μm at a high speed of 120-400 m/min, quantify the depth/height and classify them, and avoid false machine shutdowns caused by harmless stains. At the same time, it integrates 2D information to reduce the confusion between concavities/convexities and stains.
What are the effects after the implementation of WeLinkirt's solution?
After the implementation of the solution, the factory achieved online classification detection of pinholes and agglomerations at full-speed production, reliably detected and classified defects larger than 50μm. It significantly reduced false machine shutdowns caused by stains and the missed-detection rate of real defects, could quickly trace abnormalities, and improved production capacity and yield.
This article was generated by AI. Customer cases are simulated scenarios based on real product capabilities and figures are illustrative; see product pages for official benchmarks.