Verification of turning insert specifications through three-dimensional vision system
| Autor: | Geng Hao Ping, Pei Ju Chiang, Jui Hung Chien |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
0209 industrial biotechnology
Insert (composites) business.product_category Machine vision business.industry Computer science Mechanical Engineering 010401 analytical chemistry Three dimensional vision 02 engineering and technology Edge (geometry) 01 natural sciences Industrial and Manufacturing Engineering 0104 chemical sciences Computer Science Applications Machine tool Identification (information) 020901 industrial engineering & automation Control and Systems Engineering Computer vision Artificial intelligence business Software |
| Zdroj: | The International Journal of Advanced Manufacturing Technology. 96:3391-3401 |
| ISSN: | 1433-3015 0268-3768 |
| Popis: | Most computer numerically controlled (CNC) machine tools with turning capability use turning tools with clamped inserts due to their high precision and ease of replacement. However, although the handles of the turning tools are marked with their specification details, such labels do not appear on the inserts themselves and, thus, often lead to misplacement and installation of incorrect inserts. Accordingly, many researchers have proposed imaging systems based on scanners, single cameras, or microscopes for the automatic measurement and identification of inserts. However, such systems require that the inserts be unloaded from the turning tool and positioned precisely in front of the imaging system. Consequently, online measurement is impossible. This study thus proposes a three-dimensional (3D) vision system capable of identifying inserts in situ based on 3D measurements. Specifications such as insert angles, edge lengths, and nose radii of each insert were identified. The feasibility of the proposed system is demonstrated by identifying the specifications of nine types of insert. The experimental results show that the system achieves an average recognition rate of 98.89% for insert angles, 95.56% for cutting edge lengths, and 92.22% for nose radii. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|