In the quality inspection process of the electronics/PCBA industry, false alarms caused by component polarity reversal have always been a difficult problem for enterprises. WeLinkirt's AI AOI software system provides an effective solution to this problem with its advanced technology.
User scenario: A leading electronics manufacturer's PCBA production line, which mainly produces printed circuit boards for various electronic products. In the assembly process, it is necessary to detect the polarity of electronic components on the circuit board to ensure that the components are installed in the correct direction. The detection objects include various surface-mounted capacitors, diodes and other components with polarity.
Pain points: Traditional industrial vision quality inspection methods have a relatively high false alarm rate of about 20% when detecting component polarity. This not only increases the workload of manual re-inspection, reduces production efficiency, but also may lead to missed detections. At the same time, with the development of Industry 4.0, the application of industrial AI controllers in industrial vision quality inspection has gradually become a hot topic. However, traditional methods are difficult to effectively combine with industrial AI controllers and cannot fully发挥 their advantages. In addition, when changing the production line model, it is necessary to reprogram, which takes a long time, about 30 minutes, affecting the flexibility of production.
Technical principle
WeLinkirt's AI AOI software system uses a feature recognition algorithm based on a visual foundation model. This algorithm can accurately identify the features of components, including polarity features, through learning from a large number of good products. During the detection process, the system compares the actually detected component features with the learned features to determine whether the component polarity is correct.
- The visual foundation model has strong generalization ability, which can adapt to different types of components and circuit boards, improving the detection coverage rate.
- The APDT positive sample/few-shot learning technology only requires 1-20 good product samples to quickly complete the model training, greatly shortening the training time.
- The semantic false alarm filtering function can intelligently analyze and filter false alarm information, reducing unnecessary false alarms.
WeLinkirt's solution and product
Centered on the AI AOI software system, this system has a 0-code automatic programming function. It only takes 5 minutes to complete the programming for one good product, greatly shortening the production line change time. At the same time, the system supports 100% local private deployment of SDK/API/Docker, and the data does not leave the factory, ensuring data security. In practical applications, when combined with the DaoAI 2D/3D AI AOI equipment, it can detect the component polarity more accurately.
The AI AOI software system provides an efficient and accurate solution for quality inspection in the electronics/PCBA industry.
Quantitative results: After using the AI AOI software system, the detection rate of component polarity reversal reached 98%, and the missed detection rate was reduced to <2%. The false alarm rate was reduced by -75%, greatly reducing the workload of manual re-inspection. The production line change time was shortened from the original 30 minutes to 5 minutes, improving the flexibility and efficiency of production.