Segment Matching Gesture Recognition Algorithm and its Application in Smartphone

Abstract:

Aim to the traditional acceleration gesture recognition system on PC platform had the problem of high power consumption, hard to carry and low recognition rate, the paper proposes a novel gesture recognition algorithm. The algorithm first sampled the gestures signal acceleration by acceleration sensor, and then segmented and smoothing filtered the collected original signal. After preprocessing, extracted the feature value and segmented the feature value according to segments signal energy. Finally for all the segments used the improved DTW(Dynamic Time Warping) algorithm[1] to match the extracted signal features with the predefined template feature respectively and integrated the matching results of them, then concluded the final recognition results. We apply the proposed algorithm to the smartphone and test the system. Testing result shows that: The novel algorithm can improve the recognition rate and enable the system to real-time and accuracy recognizes gestures.

• Periodical: Applied Mechanics and Materials (Volumes 511-512)
• Pages: 936-940
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.511-512.936
• Online since: February 2014
• Authors: Yi Zhang, Sheng Hui Li, Yuan Luo
• Keywords: Improved DTW Algorithm, Segment Matching, Smart Phone

References

[1] LUAN Fang-jun, LI Kai, MA Si-liang. On-line handwritten signature verification algorithm based on VDDTW[J]. Computer Engineering and Applications, 2009 , 45(13): 188-190. (In chinese).

[2] Shengli Zhou, QingShan. Gesture recognition for interactive controllers using MEMS motion sensors[C]. IEEE International Conference on Nano/Micro Engineered and Molecular Systems, 2009: 935-940.

DOI: 10.1109/nems.2009.5068728

[3] Jiayang Liu, Zhen Wang. uWave Accelerometer based Personalized Gesture Recognition and Its Applications[C]. IEEE International Conference on Pervasive Computing and Communications. 2009: 657-675.

DOI: 10.1109/percom.2009.4912759

[4] Juha Kela, Panu Korpipaa, Jani Mantyjarvi ，et a1. Accelerometer-based gesture control for a design environment[J]. Personal and Ubiquitous Computing, 2006, 10(5): 285-299.

DOI: 10.1007/s00779-005-0033-8

[5] SungJung Cho, Eunseok Choi. Two-stage Recognition of Raw Acceleration Signals for 3-D Gesture-Understanding Cell Phones[C]. Tenth International Workshop on Frontiers in Handwriting Recognition (2006), (2006).

[6] Ruize Xu, Shengli Zhou et a1. MEMS Accelerometer Based Nonspecific-User Hand Gesture Recognition[J]. IEEE SENSORS JOURNAL, 2012, 12(5): 1166-1173.

DOI: 10.1109/jsen.2011.2166953

[7] Junker H，Amft O，Lukowicz P，et a1．Gesture spotting with bodywom inertial sensors to detect user activities[J]. Pattern Recognition, 2008, 41(6): 2010-2024.

DOI: 10.1016/j.patcog.2007.11.016

[8] Wang Jeellshing，Hsu Yuliang，Liu jiunnan．An Inertial-Measurement-Unit-Based Pen with a Trajectory Reconstruction Algorithm and its Applications[J]. IEEE Transactions On Industrial Electronics, 2010, 57(10): 3508-3521.

DOI: 10.1109/tie.2009.2038339