본문 바로가기
KERD
논문KCI 등재

Combining Dynamic Time Warping and Single Hidden Layer Feedforward Neural Networks for Temporal Sign Language Recognition

김선희|양형정|김수형|Ngoc Anh Nguyen Thi
한국콘텐츠학회 (The Korea Contents Association)|2011.03|International Journal of Contents|제7권 제1호|pp. 14-22

0

조회

0

다운로드

0

인용

0

참고문헌

주제분류컴퓨터학

초록

Temporal Sign Language Recognition (TSLR) from hand motion is an active area of gesture recognition research in facilitating efficient communication with deaf people. TSLR systems consist of two stages: a motion sensing step which extracts useful features from signers’ motion and a classification process which classifies these features as a performed sign. This work focuses on two of the research problems, namely unknown time varying signal of sign languages in feature extraction stage and computing complexity and time consumption in classification stage due to a very large sign sequences database. In this paper, we propose a combination of Dynamic Time Warping (DTW) and application of the Single hidden Layer Feedforward Neural networks (SLFNs) trained by Extreme Learning Machine (ELM) to cope the limitations. DTW has several advantages over other approaches in that it can align the length of the time series data to a same prior size, while ELM is a useful technique for classifying these warped features. Our experiment demonstrates the efficiency of the proposed method with the recognition accuracy up to 98.67%. The proposed approach can be generalized to more detailed measurements so as to recognize hand gestures, body motion and facial expression.

Abstract

Temporal Sign Language Recognition (TSLR) from hand motion is an active area of gesture recognition research in facilitating efficient communication with deaf people. TSLR systems consist of two stages: a motion sensing step which extracts useful features from signers’ motion and a classification process which classifies these features as a performed sign. This work focuses on two of the research problems, namely unknown time varying signal of sign languages in feature extraction stage and computing complexity and time consumption in classification stage due to a very large sign sequences database. In this paper, we propose a combination of Dynamic Time Warping (DTW) and application of the Single hidden Layer Feedforward Neural networks (SLFNs) trained by Extreme Learning Machine (ELM) to cope the limitations. DTW has several advantages over other approaches in that it can align the length of the time series data to a same prior size, while ELM is a useful technique for classifying these warped features. Our experiment demonstrates the efficiency of the proposed method with the recognition accuracy up to 98.67%. The proposed approach can be generalized to more detailed measurements so as to recognize hand gestures, body motion and facial expression.

참고문헌 (25)

  1. G. B. Huang, "Extreme learning machine: Theory and application", Nero computing, 70, pp.489~501 (2006)
  2. H.T. Huynh, "AN IMPROVEMENT OF EXTREME LEARNING MACHINE FOR COMPACT SINGLE-HIDDEN-LAYER FEEDFORWARD NEURAL NETWORKS", INTERNATIONAL JOURNAL OF NEURAL SYSTEMS, 18(5), pp.433~441, WORLD SCIENTIFIC PUBL CO PTE LTD
  3. 원용관, "Small number of hidden units for ELM with two-stage liner model", IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, E91D(4), pp.1042~1049, IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG
  4. X.Weng, "Classification of multivariate time series using two dimensional singular value decomposition", Knowledge-Based Systems, 21(7), pp.535~539 (2008)
  5. G. B. Huang, "Learning capability and storage capacity of two hidden layer feedforward networks", IEEE Transactions on Neural Networks, 14(2), pp.274~281 (2003)
  6. J. F. Lichtenauer, "Sign Language Recognition by combining Statiscal DTW and Independent Classification", IEEE transactions on pattern analysis and machine intelligence, 30(11), pp.2040~2046 (2008)
  7. P. Vamplew, "Recognition of sign language gestures using neural network", Proc. 1st Euro. Conf. Disability Virtual Reality & Assoc. Tech, pp.27~33 (1996)
  8. E. Jung Holden, "Hand movement Classification an adaptive fuzzy expert system", Intl. J. Expert Systems, pp.465~480 (1996)
  9. E. Wilson, "Neural networks for sign language translation", SPIE: Applications of Artificial Neural networks, pp.589~599
  10. D. Gavrila, "Towards 3D Model Based Tracking and Recognition of Human Movement: A Multi View Approach", Proc. IEEE Int’l Workshop Face and Gesture Recognition, pp.272~277 (1995)
  11. A. Corradini, "Dynamic Time Warping for OffLine Recognition of a Small Gesture Vocabulary", Proc. IEEE ICCV Workshop Recognition, Analysis, and Tracking of Faces and Gestures in RealTime Systems (RATFGRTS ’01), pp.82~89 (2001)
  12. S. Yang, "Gesture Recognition Using Hidden Markov Models from Fragmented Observations", Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR ’06), pp.766~773 (2006)
  13. N. Morgan, "Continuous Speech Recognition Using Multilayer Perceptrons with Hidden Markov Models", Proc. Int’l Conf. Acoustics, Speech and Signal Processing (ICASSP ’90), pp.413~416 (1990)
  14. Y. Matsuura, "Word Recognition Using a Neural Network and a Phonetically Based DTW", Proc. IEEE Int’l Workshop Neural Networks for Signal Processing (NNSP ’94), pp.329~334 (1994)
  15. A. Corradini, "Camera Based Gesture Recognition for Robot Control", Proc. Int’l Joint Conf. Neural Networks (IJCNN ’00), pp.133~138 (2000)
  16. J. Ye, "Based on HMM and SVM Multilayer Architecture Classifier for Chinese Sign Language Recognition with Large Vocabulary", Proc. Third Int’l Conf. Image and Graphics (ICIG ’04), pp.377~380 (2004)
  17. O. Aran, "Recognizing Two Handed Gestures with Generative, Discriminative and Ensemble Methods via Fisher Kernels", Proc. Int’l Workshop Multimedia Content Representation, Classification and Security (MCRCS ’06), pp.159~166 (2006)
  18. C. Bahlmann, "Online Handwriting Recognition with Support Vector Machines- A Kernel Approach", Proc. Eighth Int’l Workshop Frontiers in Handwriting Recognition (IWFHR ’02), pp.49~54 (2002)
  19. B. Strle, "Qualitative approximation to Dynamic Time Warping similarity between time series data", 23rd Annual Workshop on Qualitative Reasoning, pp.22~24 (2009)
  20. S.H. Kim, "Temporal Sign Language Analysis Based on DTW and Incremental Model", MICC 2009, pp.14~16 (2009)
  21. E.Keogh, "Exact Indexing of Dynamic Time Warping", Very Large Database, pp.406~417 (2002)
  22. H.T. Huynh, "Hematocrit Estimation from compact single hidden layer feedforward neural networks trained by evolutionary algorithm", IEEE Congress on Evolutionary Computation, pp.2962~2966 (2008)
  23. G.B, "Extreme learning machine: A new learning scheme of feedforward neural networks", Proc. of International Joint Conference on Neural Networks (IJCNN2004), pp.489~501 (2004)
  24. T. Starner, "Visual Recognition of American Sign Language Using Hidden Markov Models", 박사, Media Arts and Sciences, pp.1995

관련 논문AI추천

Korean Sign Language Recognition Using LSTM and Video Datasets임선영, 정수연, Ho-Yeon Jeong한국정보처리학회 · JIPS(Journal of Information Processing Systems)2025 · 제21권 제4호Effective Hand Gesture Recognition by Key Frame Selection and 3D Neural Network김수형, 양형정, 이귀상, 누엔호황(사)한국스마트미디어학회 · 스마트미디어저널2020 · 제9권 제1호수화번역시스템을 위한 인공신경망의 응용김형훈, 조정란한국컴퓨터정보학회 · 한국컴퓨터정보학회논문지2019 · 제24권 제2호An Efficient Approach for Eigen-joints-based 3D Activity Recognition서호, 이칠우한국차세대컴퓨팅학회 · 한국차세대컴퓨팅학회 논문지2015 · 제11권 제4호인공신경망을 사용한 수화영상인식시스템김형훈, 조정란한국컴퓨터정보학회 · 한국컴퓨터정보학회논문지2019 · 제24권 제2호Dynamic Time Warping 기반의 특징 강조형 제스처 인식 모델권혁태, 이석균한국정보처리학회 · 정보처리학회 논문지2015 · 제4권 제3호

이 간행물의 다른 논문 (9건)

Implementation of low power algorithm for near distance wireless communication and RFID/USN systems황문수, 김송주, 김영민pp. 1-7An Energy Awareness Congestion Control Scheme in Wireless Sensor Networks곽동원, 박준호, 유재수, 성동욱, 김미경pp. 8-13Analysis of Verification Methodologies Based on a SoC Platform Design김상춘, 이제훈pp. 23-28A Development of Gesture Interfaces using Spatial Context Information배기태, 권두영pp. 29-36Creative Analysis of Brand Placement in Game Contents이용재pp. 37-44
원문 보기