| Autor: |
Jaekyung Song, Jeongsu Lee, Young Chul Lee |
| Jazyk: |
angličtina |
| Rok vydání: |
2020 |
| Předmět: |
|
| Zdroj: |
China Foundry, Vol 17, Iss 5, Pp 364-371 (2020) |
| Druh dokumentu: |
article |
| ISSN: |
1672-6421 |
| DOI: |
10.1007/s41230-020-0036-0 |
| Popis: |
A code-generation and recognition technology that uses a modified ejection system in the diecasting process is presented. To achieve the highest level of quality management, the first requirement in the manufacturing process is to establish a product management system according to the specific product unit. Thus, a method to individually identify each product, such as a barcode or QR code, is required during the production process. Products manufactured in the die-casting process always have ejector pin (EP) marks. Herein, an ejection system was modified to generate a unique code using EP marks. This ejection system has two features: an EP with a modified head to show the direction of rotation, and a function to dependently rotate EPs (five or six EPs) with a constant angle. The EPs are numbered according to the rotation angle. Thus, the EP marks can be viewed as a five- or six-digit code. A program was also developed to individually identify the products by automatically detecting and reading the EPs using deep learning-based object detection and classification technology. |
| Databáze: |
Directory of Open Access Journals |
| Externí odkaz: |
|
