Application of elastic principal component analysis to person recognition based on screen gestures
Authors:
- Mateusz Baran,
- Leszek Siwik,
- Krzysztof Rzecki
Abstract
Person identification based on touch screen gestures is a well-known method of authentication in mobile devices. Usually it is only checked if the user entered the correct pattern. Taking into account other biometric data based on the speed and shape of finger movements can provide higher security while the convenience of this authorisation method is not impacted. In this work the application of Sequential Joint Functional Principal Analysis (FPCA) as a dimensionality reduction method for gesture data is explored. Performance of the classifier is measured using 5-fold stratified cross-validation on a set of gestures collected from 12 people. The effects of sampling rate on classification performance is also measured. It is shown that the Support Vector Machine classifier reaches the accuracy of 79% using features obtained using the Sequential Joint FPCA, compared to 70% in the case of Euclidean PCA.
- Record ID
- CUT728f2ba5ebb14c47874a51556de1e3c2
- Publication categories
- ; ;
- Author
- Pages
- 553-560
- Other elements of collation
- il.; Bibliografia (na s.) - 558-560; Bibliografia (liczba pozycji) - 31; Oznaczenie streszczenia - Abstr.
- Book
- Rutkowski Leszek, Leszek Rutkowski Scherer Rafał, Rafał Scherer Korytkowski Marcin Marcin Korytkowski [et al.] (eds.): Artificial Intelligenceand Soft Computing : 18th International Conference, ICAISC 2019, Zakopane, Poland, June 16–20, 2019 : proceedings. Pt. 1, Lecture Notes in Artificial Intelligence, no. 11508, 2019, Cham, Springer, Springer, ISBN 978-3-030-20912-4 (eBook)
- Keywords in English
- touch screen gestures, biometrics, classification, elastic shape analysis, pattern recognition
- DOI
- DOI:10.1007/978-3-030-20912-4_50 Opening in a new tab
- URL
- https://link.springer.com/chapter/10.1007/978-3-030-20912-4_50 Opening in a new tab
- Language
- eng (en) English
- Score (nominal)
- 20
- Additional fields
- Indeksowana w: Web of Science, Scopus, CORE
- Uniform Resource Identifier
- https://cris.pk.edu.pl/info/article/CUT728f2ba5ebb14c47874a51556de1e3c2/
- URN
urn:pkr-prod:CUT728f2ba5ebb14c47874a51556de1e3c2
* presented citation count is obtained through Internet information analysis, and it is close to the number calculated by the Publish or PerishOpening in a new tab system.