Flaw detection for disorderly feeding of hub
Publisher: Shiyan Time:12/11/2021 5:25:24 PM
During battery production, defects such as depression, scratch, exposed foil and fold will occur due to coater and light press. Defects with deep scratches may lead to safety hazards such as explosion. A visual scheme is needed to detect these defective products.
01. Traditional manual detection will have subjective will, emotion, fatigue and other factors, which can not ensure the stable detection of a variety of defects.
02. Some workpieces have strong reflection. When detecting wrinkles, bright spots, dark spots, edge banding and other defects, the local reflection effect is uneven, and will be affected by flatness and background cleanliness. The defect imaging effect is poor and it is difficult to identify.
03. It is necessary to detect scratches, pits and other surface defects with X /Y size > 0.5*0.5mm and depth ≥ 0.1mm.
- 01. 3D line laser + depth learning algorithm based on semi supervised learning is adopted.
- 02. Do not need lighting, compatible with all products and lighting conditions.
- 03. The scanning accuracy is high, and the detection accuracy of X, Y and Z axes can reach 20 μ m. Stable identification of surface defects;
- 04. Highly stable and reliable intelligent identification of defects and statistics of defect types to facilitate traceability of product quality.