Online Handwriting Recognition Using an Accelerometer-Based Pen Device
Wang Jeen-Shing, Hsu Yu-Liang, Chu Cheng-Ling
Available Online July 2013.
- https://doi.org/10.2991/cse.2013.52How to use a DOI?
- Online handwriting recognition, accelerometer-based input device, dynamic time warping.
- This paper presents an accelerometer-based pen device for online handwriting recognition applications. The accelerometer-based pen device consists of a triaxial accelerometer, a microcontroller, and an RF wireless transmission module. Users can hold the pen device to write numerals in air without space limitations. The accelerations generated by hand motions are generated by the accelerometer embedded in the pen device, and are transmitted to a personal computer for further signal preprocess via the wireless module. Subsequently, a dynamic time warping (DTW) algorithm is applied to align the accelerations and search class templates for each digit in the training stage. Finally, the accelerations can be recognized via the alignment with the class templates in the testing stage. The average user-dependent and user-independent recognition rates were 90.6% and 84.8%, respectively. Our experimental results have successfully validated the effectiveness of the accelerometer-based pen device for 3-dimensional (3-D) online handwriting recognition.
- Open Access
- This is an open access article distributed under the CC BY-NC license.
Cite this article
TY - CONF AU - Wang Jeen-Shing AU - Hsu Yu-Liang AU - Chu Cheng-Ling PY - 2013/07 DA - 2013/07 TI - Online Handwriting Recognition Using an Accelerometer-Based Pen Device BT - 2nd International Conference on Advances in Computer Science and Engineering (CSE 2013) PB - Atlantis Press SP - 231 EP - 234 SN - 1951-6851 UR - https://doi.org/10.2991/cse.2013.52 DO - https://doi.org/10.2991/cse.2013.52 ID - Jeen-Shing2013/07 ER -