A Hilbert warping method for handwriting gesture recognition |
| |
Authors: | Hiroyuki Ishida [Author Vitae] Tomokazu Takahashi [Author Vitae] [Author Vitae] Hiroshi Murase [Author Vitae] |
| |
Affiliation: | a Graduate School of Information Science, Nagoya University, Furo-cho, Chikusa-ku, Nagoya, Aichi 464-8601, Japan b Department of Economics and Information, Gifu Shotoku Gakuen University, Nakauzura 1-38, Gifu-shi, Gifu 501-6194, Japan |
| |
Abstract: | We propose a novel sequence alignment algorithm for recognizing handwriting gestures by a camera. In the proposed method, an input image sequence is aligned to the reference sequences by phase-synchronization of analytic signals which are transformed from original feature values. A cumulative distance is calculated simultaneously with the alignment process, and then used for the classification. A major benefit of this method is that over-fitting to sequences of incorrect categories is restricted. The proposed method exhibited higher recognition accuracy in handwriting gesture recognition, compared with the conventional dynamic time warping method which explores optimal alignment results for all categories. |
| |
Keywords: | Sequence alignment Classification method Analytic signal Gesture recognition |
本文献已被 ScienceDirect 等数据库收录! |
|