| Presentation | 2023-03-14 Feature Extraction of Grasped Objects for Precise Robot Hand Control using Depth Image Satoshi Katayama, Motoi Matsuda, Naohiro Fukumura, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In recent years, development of robots to support people has been promoted. To control a multi-fingered hand robot for executing grasping movements in a living space, it is necessary not only to assign several pre-defined patterns of hand posture but also to recognize the complex 3D shape of an object such as a cup by images and to obtain a more detailed hand posture . Therefore, in this study, we extract features from the images of the object that are required for precise robot hand control and verify the classification of size variation. In the experiment, a camera equipped with a sensor that measures distance information acquires depth images of cups of different sizes. Then, CNN or Variational Auto Encoder is trained with these images as input, and feature extraction is performed on the shapes to evaluate the accuracy of classification. In addition, generalizability evaluation will also be conducted using test data obtained in the bright or dark environment． |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Depth Image / VAE / Feature Extraction / Object recognition |
| Paper # | NC2022-105 |
| Date of Issue | 2023-03-06 (NC) |
| Conference Information | |
| Committee | NC / MBE |
|---|---|
| Conference Date | 2023/3/13(3days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | The Univ. of Electro-Communications |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | Brain architecture, General |
| Chair | Hiroshi Yamakawa(Univ of Tokyo) / Junichi Hori(Niigata Univ.) |
| Vice Chair | Hirokazu Tanaka(Tokyo City Univ.) / Hisashi Yoshida(Kinki Univ.) |
| Secretary | Hirokazu Tanaka(NTT) / Hisashi Yoshida(NICT) |
| Assistant | Yoshimasa Tawatsuji(Waseda Univ.) / Tomoki Kurikawa(KMU) / Emi Yuda(Tohoku Univ) / Miki Kaneko(Osaka Univ.) |
| Paper Information | |
| Registration To | Technical Committee on Neurocomputing / Technical Committee on ME and Bio Cybernetics |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Feature Extraction of Grasped Objects for Precise Robot Hand Control using Depth Image |
| Sub Title (in English) | |
| Keyword(1) | Depth Image |
| Keyword(2) | VAE |
| Keyword(3) | Feature Extraction |
| Keyword(4) | Object recognition |
| 1st Author's Name | Satoshi Katayama |
| 1st Author's Affiliation | Toyohashi University of Technology(Toyohashi Univ. of Tech) |
| 2nd Author's Name | Motoi Matsuda |
| 2nd Author's Affiliation | Toyohashi University of Technology(Toyohashi Univ. of Tech) |
| 3rd Author's Name | Naohiro Fukumura |
| 3rd Author's Affiliation | Toyohashi University of Technology(Toyohashi Univ. of Tech) |
| Date | 2023-03-14 |
| Paper # | NC2022-105 |
| Volume (vol) | vol.122 |
| Number (no) | NC-425 |
| Page | pp.pp.78-83(NC), |
| #Pages | 6 |
| Date of Issue | 2023-03-06 (NC) |