Deep learning for part identification based on inherent features
| Autor: | Jan Lehr, Jörg Krüger, Marian Schlüter, Nils Bischoff |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
0209 industrial biotechnology
Training set Computer science business.industry Mechanical Engineering Deep learning 02 engineering and technology Machine learning computer.software_genre Automation Industrial and Manufacturing Engineering 020303 mechanical engineering & transports 020901 industrial engineering & automation 0203 mechanical engineering Robustness (computer science) Optical identification Artificial intelligence business computer |
| Zdroj: | CIRP Annals. 68:9-12 |
| ISSN: | 0007-8506 |
| DOI: | 10.1016/j.cirp.2019.04.095 |
| Popis: | The identification of parts is essential for the efficient automation of logistic processes such as part supply in assembly and disassembly. This paper describes a new method for the optical identification of parts without explicit codes but based on inherent geometrical features with Deep Learning. The paper focusses on the improvement of training of Deep Learning systems taking into account conflicting factors such as limited training data and high variety of parts. Based on a case study in turbine industry the effects of steadily growing training data on the robustness of part classification are evaluated. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect