Call For Papers
Contact Us

  Performance Improvement in Hand Gesture Recognition Using Adaptive Kalman Filter  
  Authors : Amol Mahadik; Madhuri Joshi
  Cite as:


This paper suggests method to improve hand gesture recognition rate in limited environment. The system consists of two main stages which are Hand detection and tracking. In Hand detection, hand region is first detected by combining motion and skin color pixels. A region of interest (ROI) is then created in the detected hand region. In tracking stage, skin and motion pixels are scanned around top, left and right corners of the ROI to detect the moving hand in consecutive video frames. These pixels are used to actually measure the ROI position and fed into measurement update of Adaptive Kalman Filter (AKF) operation. The experimental result shows the proposed method has the robust ability to track the moving hand under real life scenarios with average 96.66% result.


Published In : IJCSN Journal Volume 3, Issue 4

Date of Publication : 01 August 2014

Pages : 156 - 161

Figures : 06

Tables : 03

Publication Link : Performance Improvement in Hand Gesture Recognition Using Adaptive Kalman Filter




Amol Mahadik : completed his B.E. in Information Technology from B.A.M. University Aurangabad. Currently he is perusing his master’s degree from same university. His research interest lies in Human computer Interaction.

Madhuri S Joshi : completed her BE from College of Engineering, Pune (1985), M.Tech. (CS) (1993) from IIT, Madras and Ph.D. from SRT University, Maharashtra, India. She has published 29 research papers in various International Journals, International and National Conferences. Her areas of interest are Data Mining ,Image Processing and Pattern Recognition.








Hand detection

Hand tracking

Hand gesture

Adaptive Kalman Filter

Region of Interest

Adaptive Kalman Filter (AKF) gives advantages for Hand Gesture Recognition as compared to HMM. AKF doesn’t require any kind of training. It can work without sensors to track the hand. It is simple to use. It gives more accuracy than HMM. Practical accuracy might vary as per the available illumination. The proposed system can be made even more robust so that it will give same performance for all types of users irrespective of their expertise. Due to extensive training sets for modeling HMM is computationally expensive. Authors are willing to continue the experimentation further.










[1] Sushmita Mitra, and Tinku Acharya, “Gesture Recognition: A Survey” IEEE VOL. 37, NO. 3, MAY 2007

[2] Mohd Shahrimie Mohd Asaari, “Hand Gesture Tracking System Using Kalman Filter”, 2010 IEEE, 10th International Conference on Intelligent Systems Design and Applications

[3] F.S. Chen, C.M. Fu and C.L. Huang, “Hand gesture recognition using a real-time tracking method and hidden Markov models”, Image Vision Computer, IEEE v 21 n 8, pp 745- 758, March 2003.

[4] M. Isard and A. Blake, “Conditional Density Propagation for Visual Tracking,” IJCV, v 29, n 1, pp 5-28, Aug. 2008.

[5] G. McAllister, S. McKenna and I. Ricketts, “Hand Tracking for Behavior Understanding,” Image and Vision Computing, VOL 20, NO 12,Oct. 2009.

[6] Q.Y. Zhang, M.Y. Zhang and J.Q. Hu, “Hand Gesture Contour Tracking Based on Skin Color Probability and State Estimation Model,” IEEE, VOL 4, NO 6, Dec. 2009.

[7] K.Imagawa, S.Lu and S.Igi, “Color-Based Hands Tracking System for Sign Language Recognition,” Proc. International Conference on Face and Gesture Recognition, Apr. 14-16, 1998, Nara, Japan.

[8] N.D.Binh, E.Shuichi, and T.Ejima, Real-Time Hand Tracking and Gesture Recognition System,” Proc. GVIP05, Dec.19-21, 2005, Cairo, Egypt.

[9] Byung-Woo Min, Ho-Sub Yoon, Jung Soh, Yun-Mo Yang, ”Hand Gesture Recognition Using HMM ”, IEEE International Conference on computational Cybernetics and Simulation 1997,Vol 5 pp 745- 758, March 1997.