Development and Application Study of a Calibration Certificate Anomaly Detection System |
|---|
| Dong-Hun Ryu, Chae-Wook Lim |
- Abstract:
- This study focuses on the development of a system to effectively extract and utilize data from approximately 3,000 types of unstructured calibration certificates (e.g., paper, PDFs) widely used in the calibration industry. To process document-based data that are difficult for machines to recognize, the Korea Testing Laboratory (KTL) has developed a calibration certificate anomaly detection system called K-Argos, which applies object detection AI technology based on Convolutional Neural Networks (CNNs). The core function of the system is to extract data from unstructured documents, convert them into structured data, and standardize and store the results. Using the converted structured data, K-Argos provides statistical data management features such as performance analysis of calibrated equipment and anomaly detection in calibration data. Additionally, the system can be used to correct inaccurate identification information (e.g., manufacturer, model, serial number) and to analyze errors within the calibration data. In the future, the K-Argos system will be expanded to include an online technical supervisor service that automatically determines anomalies in data during the certificate approval stage and provides relevant information to the approver (technical supervisor). This is expected to significantly improve the efficiency and reliability of calibration tasks and contribute to the digital transformation of the calibration industry through the implementation of Digital Calibration Certificates (DCCs).
- Download:
- IMEKO-TC6-2025-014.pdf
- DOI:
- 10.21014/tc6-2025.014
- Event details
- IMEKO TC:
- TC6
- Event name:
- TC6 M4Dconf2025
- Title:
2025 IMEKO TC-6 International Conference on Metrology and Digital Transformation
- Place:
- Benevento, ITALY
- Time:
- 03 September 2025 - 05 September 2025