복잡한 영상에 강인한 손동작 인식 방법

Hand Gesture Recognition Algorithm Robust to Complex Image

  • 박상윤 (동명대학교 대학원 정보통신공학과) ;
  • 이응주 (동명대학교 정보통신공학과)
  • 투고 : 2009.11.12
  • 심사 : 2010.04.15
  • 발행 : 2010.07.31

초록

본 논문에서는 손동작 인식을 위한 새로운 방법을 제안한다. 손 추출을 위한 방법으로는 피부색과 boundary energy 정보를 이용하고 moment method로 손바닥의 중심을 구하게 된다. 손동작 인식은 두 단계로 나눌 수 있다. 첫 번째 단계는 손 형상 인식으로 병렬 신경망을 이용하였다. 손 형상의 패턴을 추출하기 위해서 fitting ellipses method를 이용하였다. fitting ellipses method는 추출된 손 영역을 12개의 타원형으로 분류하고 타원 외곽선의 흰 픽셀 비율을 계산한다. 패턴은 12개의 입력 노드로 신경망에 입력되고 4개의 출력 노드로 출력되는데 각 출력 노드는 0~1사이의 값을 갖게 된다. 손 형상은 4개의 출력 노드의 구성으로 나타낼 수 있다. 두 번째 단계는 손동작 추적과 인식이다. 손동작 추적과 인식을 위해서는 손동작의 위치 정보를 예측 할 수 있는 Kalman Filter를 이용하였다. 실험은 Windows XP상에서 수행되었고 제안한 알고리즘의 효율성을 평가하였다. 손 형상을 인식하기 위해서 300개의 이미지를 인식기에 훈련시키고 200개의 이미지를 실험에 사용하였다. 194개 이상의 이미지가 정확하게 인식 되었다. 그리고 손동작 추적 인식을 실험하기 위해서 1200번의 손동작(각 동작은 400번)을 사용하였고 그 중 1002번의 손동작이 정확하게 인식 되었다.이러한 결과는 제안된 방법이 손 영역을 추출하고 손 동작을 인식하는데 유용함을 확인 할 수 있었다.

In this paper, we propose a novel algorithm for hand gesture recognition. The hand detection method is based on human skin color, and we use the boundary energy information to locate the hand region accurately, then the moment method will be employed to locate the hand palm center. Hand gesture recognition can be separated into 2 step: firstly, the hand posture recognition: we employ the parallel NNs to deal with problem of hand posture recognition, pattern of a hand posture can be extracted by utilize the fitting ellipses method, which separates the detected hand region by 12 ellipses and calculates the white pixels rate in ellipse line. the pattern will be input to the NNs with 12 input nodes, the NNs contains 4 output nodes, each output node out a value within 0~1, the posture is then represented by composed of the 4 output codes. Secondly, the hand gesture tracking and recognition: we employed the Kalman filter to predict the position information of gesture to create the position sequence, distance relationship between positions will be used to confirm the gesture. The simulation have been performed on Windows XP to evaluate the efficiency of the algorithm, for recognizing the hand posture, we used 300 training images to train the recognizing machine and used 200 images to test the machine, the correct number is up to 194. And for testing the hand tracking recognition part, we make 1200 times gesture (each gesture 400 times), the total correct number is 1002 times. These results shows that the proposed gesture recognition algorithm can achieve an endurable job for detecting the hand and its' gesture.

키워드

과제정보

연구 과제 주관 기관 : 중소기업청

참고문헌

  1. K. Fukushima, "Neural Network Model for Selective Attention in Visual Pattern Recognition and Associative Recall," Applied Optics 26, pp. 4985-4992, 1987. https://doi.org/10.1364/AO.26.004985
  2. P. Dayan, S. Kakade, and P. R. Montague, "Learning and selective attention," Nature Neuroscience 3, pp. 1218-1223, 2000. https://doi.org/10.1038/81504
  3. Francis K. H. Quek, "Unencumbered gestural interaction," IEEE MultiMedia, Vol.3, No.4, pp. 36-47, Winter, 1996. https://doi.org/10.1109/93.556459
  4. R. P. N.Rao and A. N. Meltzoff, "Imitation Leaning in Infants and Robots: Towards Probabilistic Computational Models," Proceedings of Artificial Intelligence and Simulation of Behaviors, 2003.
  5. S. Cali non and A. Billard, "Stochastic Gesture Production and Recognition Model for a Humanoid Robot," Proc. of IEEE/RSJ International Conference on Intelligent Robots and Systems, pp. 2769-2774, 2004.
  6. Y. Hamada, N. Shamada, and Y. Shirai, "Hand Shape Estimation Using Image Transition Network," Proc. Of Workshop on Human Motion, pp. 161-166, 2000.
  7. E. Ong and R. Bowden, "A Boosted Classifier Tree for Hand Shape Detection," Sixth IEEE International Conference on Automatic Face and Gesture Recognition, pp. 889-894, 2004.
  8. V. Athitsos and S. Sclaroff, "An Appearance-Based Framework for 3D Hand Shape Classification and Camera Viewpoint Estimation," Proc. of Face and Gesture Recognition, 2002.
  9. Ihanb Zaqout, Roziati Zainuddin, and Sapian Baba, "Pixel-based skin color detection technique," Machine Graphics & Vision International Journal Volume 14, Issue 1 pp. 61-70, 2005.
  10. P. Peer, J. Kovac, and F. Solina, "Human skin color clustering for face detection," In submitted to EUROCON 2003-International Conference on Computer as a Tool, 2003.
  11. W. Sharbek and A. Koshan, "Color Image Segmentation - a survey-," Tech. Rep., Institute for Technical Informatics. Technical University of Berlin, October, 1994.
  12. I. T. Young, J. E. Walker, and J. E. Bowie, "An Analysis Technique for Biological Shape. I," Information and Control, 25: pp. 357-370, 1974. https://doi.org/10.1016/S0019-9958(74)91038-9
  13. M. K. Hu, "Pattern Recognition by Moment Invariants," Proc. IEEE, Vol 49, No.9, pp. 1428, Sept. 1961.
  14. M. K. Hu, "Visual Pattern Recognition by Moment Invariants," IRE Transactions on Information Theory, Vol.17-8, No.2, pp. 179-187, Feb. 1962.
  15. A. P. Reeves and A. Rostampour, "Shape Analysis of Segmented Objects Using Moments," Proc. IEEE Conf on Pattern Recognition and Image Processing, pp. 171-174, 1981.
  16. R. Y. Wong and E. L. Hall, "Scene Matching with Invariant Moments," Computer Graphics and Image Processing, Vol.8, No.1, pp. 16-24, Aug. 1978. https://doi.org/10.1016/S0146-664X(78)80028-8
  17. Caglar, M.B. and Lobo, N., "Open Hand Detection in a Cluttered Single Image using Finger Primitives," 2006 Conference on Computer Vision and Pattern Recognition Workshop, pp. 17-22, June 2006.
  18. Cheung-Wen Chang and Yung-Nien Sun, "Hand Detections Based on Invariant Skin-Color Models Constructed Using Linear and Nonlinear Color Spaces," Intelligent Information Hiding and Multimedia Signal Processing, pp. 577-580, pp. 15-17, Aug. 2008.
  19. R. Setiono, "Feedforward neural network construction using cross validation," Neural Computer., Vol.13, pp. 2865-2877, 2001. https://doi.org/10.1162/089976601317098565