In the electronics/PCBA industry, component offset and tombstoning problems seriously affect product quality. WeLinkirt provides an effective inspection solution for the industry with advanced AI vision technology.
User scenario: The SMT production line of a leading electronics/PCBA manufacturer, which mainly produces various circuit board products. The inspection objects are surface - mount components on the circuit boards, including resistors, capacitors, etc. The focus is on two common defects: component offset and tombstoning.
Pain points: Under the traditional inspection method, the miss - detection rate of component offset and tombstoning is as high as 2.5%. This means that for every 1000 circuit boards produced, there may be 25 boards with potential quality problems flowing into the market, bringing huge after - sales costs and brand reputation losses to the enterprise. At the same time, the false - alarm rate reaches 15%. A large number of false alarms not only increase the workload of manual re - inspection but also reduce the production efficiency of the production line. Moreover, when changing the production line model, the traditional programming method takes more than 30 minutes, seriously affecting the production flexibility.
Technical principle
WeLinkirt uses advanced AI algorithms and self - developed 3D camera technology. The AI algorithm can accurately identify the normal features and positions of components by learning from a large number of good samples. The self - developed 3D camera can achieve three - dimensional morphology reconstruction to obtain the accurate height and shape information of components. When a component is detected, the system compares the actual features of the component with the learned normal features. For component offset, the system calculates the deviation value between the actual position of the component and the standard position. When the deviation exceeds the set threshold, it is determined as an offset defect. For tombstoning problems, according to the height information of the component obtained by the 3D camera, if the height of the component is abnormally increased, it is determined as a tombstoning defect. This detection method based on feature comparison and 3D information greatly improves the detection accuracy.
- The AI algorithm has a strong feature - learning ability and can adapt to the detection needs of different types of components.
- The high - precision imaging of the 3D camera enables the detection of hidden defects clearly.
- Real - time comparison and analysis ensure the timeliness and accuracy of the detection results.
WeLinkirt's solution and products
WeLinkirt provides the DaoAI AI AOI software system and the DaoAI 2D / 3D AI AOI equipment. The DaoAI AI AOI software system has the feature - recognition ability of the visual basic model. With only 1 - 20 good samples, it can achieve 0 - code automatic programming in 5 minutes. It also uses APDT positive - sample/few - sample learning and semantic false - alarm filtering technology to effectively reduce false alarms. The DaoAI 2D / 3D AI AOI equipment is equipped with a self - developed 3D camera, which can perform three - dimensional morphology reconstruction and detect hidden solder joints, coplanarity, and micron - level morphology. It has extremely high accuracy in detecting component offset and tombstoning defects. In the implementation, the equipment is installed at a suitable position on the SMT production line, and the software system interacts with the equipment to process the detection data in real - time, achieving efficient and accurate detection.
WeLinkirt's AI vision technology provides reliable guarantee for the quality inspection of the electronics/PCBA industry.
Quantitative results: After adopting WeLinkirt's solution, the detection rate of component offset and tombstoning has increased to 97.5%, and the miss - detection rate has been reduced to <2.5%, effectively preventing a large number of products with potential quality problems from flowing into the market. The false - alarm rate has been reduced by - 60%, greatly reducing the workload of manual re - inspection and improving the production efficiency of the production line. The production line model - changing time has been shortened to 5min, improving the production flexibility and response speed.