Abstract
Turning-milling machine contains a large number of tools, where wrong disposition of tools possesses collision risks. Operator mistakes and lack of proper communication feedback of the machine technology may lead to error that harms the machines. The aim of this study is developing fast, precise, and reliable automatic machining tool recognition system in turning-milling machines. The approach consists of two steps, predicting the class of the tool using convolutional neural network, and identifying the exact type of the tools by performing feature detection, description, and matching.
Original language | English |
---|---|
Pages | 477-482 |
Number of pages | 6 |
Publication status | Published - 2017 |
Event | 20th International Symposium on Advances in Abrasive Technology, ISAAT 2017 - Okinawa, Japan Duration: 2017 Dec 3 → 2017 Dec 6 |
Conference
Conference | 20th International Symposium on Advances in Abrasive Technology, ISAAT 2017 |
---|---|
Country/Territory | Japan |
City | Okinawa |
Period | 17/12/3 → 17/12/6 |
Keywords
- Collision avoidance
- Convolutional neural network
- Description
- Feature detection
- Matching
- Tools identification
- Turning-milling machine
ASJC Scopus subject areas
- Mechanics of Materials
- Industrial and Manufacturing Engineering
- Mechanical Engineering
- Materials Science(all)