In the electronics industry, the quality inspection of FPC flexible boards is of crucial importance. DaoAI provides an efficient solution for FPC flexible board defect inspection with its advanced AI vision technology.
User Scenario: On the FPC flexible board production line of a leading electronics manufacturer, defect inspection of FPC flexible boards is required after the assembly process. As an important connecting component in electronic products, the quality of FPC flexible boards directly affects the performance of the entire product. The inspection objects include various defects such as scratches, stains, and broken circuits on the surface of the flexible board.
Pain Points: Traditional inspection methods face many quantitative dilemmas. On the one hand, the miss - detection rate is relatively high, about 3%, which causes some defective products to enter the market and affects the product's reputation. On the other hand, the false - alarm rate reaches 20%, which leads to a large number of qualified products being misjudged as defective, increasing the re - inspection workload and production costs. In addition, the texture of FPC flexible boards is complex, and traditional methods are difficult to effectively detect defects that have never been seen before. Moreover, the model change time is long, taking 30 minutes each time, which seriously affects production efficiency.
Technical Principles
DaoAI uses unsupervised anomaly detection algorithms, combined with advanced imaging technology and hardware equipment to solve these problems. The unsupervised anomaly detection algorithm only needs to model good products. By learning the features of good products, it can identify abnormal areas different from the features of good products, thus detecting defects that have never been seen before. The principle is that normal FPC flexible boards have a certain texture and feature distribution. When a defect occurs, these textures and features will change. The algorithm establishes a normal feature model through learning a large number of good products. When the detected sample has a large difference from the model, it is judged as a defect.
- Advanced imaging technology can clearly capture the surface details of FPC flexible boards, providing accurate data for the algorithm.
- The hardware equipment has high - precision detection capabilities and can detect micron - level defects.
- The algorithm has self - learning ability and can continuously learn from the production line feedback to optimize the detection effect.
DaoAI Solutions and Product Introduction
DaoAI provides the DaoAI AI AOI software system and the DaoAI 2D / 3D AI AOI equipment. The DaoAI AI AOI software system uses the feature recognition of the visual basic model. With only one good product, it can complete 0 - code automatic programming in 5 minutes. It adopts APDT positive - sample/few - sample learning (only 1 - 20 good products are needed) and has the function of semantic false - alarm filtering, which can effectively reduce the false - alarm rate. The DaoAI 2D / 3D AI AOI equipment has a self - developed 3D camera and can perform three - dimensional shape reconstruction, which can detect hidden solder joints, coplanarity, and micron - level shapes, greatly improving the accuracy of detection. In the implementation process, the software system is combined with the hardware equipment to achieve efficient defect detection.
DaoAI's AI vision technology has brought new breakthroughs to FPC flexible board defect inspection.
Quantitative Results: By using DaoAI's solution, the defect detection rate of the manufacturer's FPC flexible boards has increased to 98%, and the miss - detection rate has decreased to <2%. The false - alarm rate has been reduced by -70%, greatly reducing the re - inspection workload. The model change time has been shortened to 5 minutes, and the production efficiency has been significantly improved.