In the high-SKU label printing scenario of the consumer goods/general industry, traditional inspection methods face many challenges. WeLinkirt's AI AOI software system provides an efficient inspection solution for this scenario with advanced technology.
User Scenario: The label printing production line of a leading consumer goods manufacturer is responsible for producing various product labels with different specifications and styles. The inspection objects are the text, patterns, colors, and barcodes on the labels. Due to the large number of product SKUs and frequent changes in label styles, high requirements are put forward for the flexibility and accuracy of the inspection system.
Pain Points: Currently, the production line uses a combination of manual visual inspection and traditional machine vision inspection, which has many quantitative difficulties. Manual visual inspection is inefficient, with an average of only about 500 labels inspected per hour. Moreover, workers are prone to fatigue after long-term work, resulting in a missed detection rate of up to 2%. Traditional machine vision inspection has poor adaptability to new label styles, and the changeover time is as long as 30 minutes, affecting the production rhythm. Meanwhile, with the continuous in - depth application of AI computing power in the industrial field, the existing inspection methods are difficult to meet the enterprise's requirements in terms of efficiency and accuracy, and cannot fully leverage the advantages of the integration of AI computing power and industrial vision quality inspection.
Technical Principle
WeLinkirt's AI AOI software system is based on the feature recognition algorithm of the visual basic model, which can quickly and accurately identify various features on the labels. Through a large number of sample trainings, the algorithm learns the normal feature patterns of labels and establishes a feature database. During the inspection process, the system compares the collected label images with the features in the database to determine whether the labels have defects.
- Zero - code automatic programming: Only by providing one good-quality label, the system can complete automatic programming within 5 minutes without the need for professional programmers, greatly shortening the changeover time.
- APDT positive sample/few-shot learning: The system can learn the normal features of labels with only 1-20 good-quality labels as positive samples, reducing the workload of sample collection.
- Semantic false-alarm filtering: The system can conduct semantic analysis on false-alarm information and filter out false alarms caused by environmental factors or image noise, improving the accuracy of detection.
WeLinkirt Solution and Product
Centered on the AI AOI software system, WeLinkirt provides a complete label printing inspection solution for the manufacturer. The system supports 100% local private deployment of SDK/API/Docker, ensuring the security of enterprise data. At the same time, it can be used in conjunction with other product lines of WeLinkirt, such as DaoAI 2D/3D AI AOI equipment and SkyVision, to further improve the inspection effect. During the implementation process, the system can quickly adapt to label changes of different SKUs and achieve efficient and accurate inspection.
The AI AOI software system brings an efficient and accurate solution for high - SKU label printing inspection.
Quantitative Results: After using WeLinkirt's AI AOI software system, the label inspection detection rate of the manufacturer has increased to 99.2%, and the missed detection rate has decreased to <0.8%. The false-alarm rate has been reduced by -65%, reducing a large amount of re-inspection work. The changeover time has been shortened from 30 minutes to 5 minutes, improving production efficiency and meeting the rapid changeover requirements of high - SKU label printing.
FAQ
What are the requirements for the number of samples of the AI AOI software system?
The system uses APDT positive sample/few-shot learning. It only needs 1-20 good-quality labels as positive samples to learn the normal features of labels, reducing the workload of sample collection.
Why can the changeover time of the system be so short?
The system supports zero-code automatic programming within 5 minutes with one good-quality label. Without the need for professional programmers, it can quickly adapt to label changes of different SKUs, so the changeover time can be shortened from 30 minutes to 5 minutes.
How to ensure the security of enterprise data?
The system supports 100% local private deployment of SDK/API/Docker, and the data does not leave the factory, effectively ensuring the security of enterprise data.