A neural network approach to predict tactile comfort of applying cosmetic foundation
| Autor: | Toshiharu Soneda, Ken Nakano, Ryuta Tsuchiya, Akio Kashimoto, Keita Horiuchi, Masahito Yokoyama |
|---|---|
| Rok vydání: | 2010 |
| Předmět: |
Engineering
Artificial neural network business.industry Mechanical Engineering Supervised learning Foundation (engineering) Pattern recognition Tactile sensation Surfaces and Interfaces computer.software_genre Expert system Surfaces Coatings and Films Mechanics of Materials Feature (computer vision) Artificial intelligence business computer Simulation Friction test Tribometer |
| Zdroj: | Tribology International. 43:1978-1990 |
| ISSN: | 0301-679X |
| DOI: | 10.1016/j.triboint.2010.04.004 |
| Popis: | An expert system has been developed to predict the degree of tactile comfort during the application of cosmetic foundation. A tribometer using compliant silicone-rubber surfaces was developed to perform 4 sliding tests and an oscillating test. From the tests for 20 different samples, 11 feature quantities were identified. The values of the feature quantities were used as the inputs to an artificial neural network, and the scores of tactile sensations were obtained as the outputs. The neural network, after supervised learning, could predict 5 types of emotional tactile comfort with high accuracy. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect