Performance Improvement of 3D-CNN for Blink Types Classification by Data Augmentation
-
- Sato Hironobu
- College of Science and Engineering, Kanto Gakuin University
-
- Abe Kiyohiko
- School of System Design and Technology, Tokyo Denki University
-
- Matsuno Shogo
- Faculty of Informatics, Gunma University
-
- Ohyama Minoru
- School of System Design and Technology, Tokyo Denki University
Bibliographic Information
- Other Title
-
- 3D CNNをもちいた瞬目種類識別のデータ拡張による性能向上
Abstract
<p>When developing a blink input interface, conscious (voluntary) and natural (involuntary) blink types must be automatically classified. We previously proposed a method for blink type classification using a 3D convolutional neural network (3D CNN). This CNN model outputs a predicted probability that determines three classes: “voluntary blinking,” “involuntary blinking,” and “not blinking” from a periocular image sequence. Previously, we found that the bias of the eye position in the input image is a factor that reduces the classification accuracy. To address this problem, we employ data augmentation with a shifting 5 or 10 pixels in the horizontal and/or vertical directions.</p>
Journal
-
- IEEJ Transactions on Electronics, Information and Systems
-
IEEJ Transactions on Electronics, Information and Systems 144 (4), 328-329, 2024-04-01
The Institute of Electrical Engineers of Japan
- Tweet
Keywords
Details 詳細情報について
-
- CRID
- 1390862623771700608
-
- ISSN
- 13488155
- 03854221
-
- Text Lang
- ja
-
- Data Source
-
- JaLC
- Crossref
-
- Abstract License Flag
- Disallowed