Infrared Camera-Based Hand Gesture Space Touch System Implementation of Smart Device Environment

This paper proposes a method to recognize the tip of a finger and space touch hand gesture using an infrared camera in a smart device environment. The proposed method estimates the tip of a finger with a curvature-based ellipse fitting algorithm, and verifies that the estimated object is indeed a finger with an ellipse fitting rectangular area. The feature extracted from the verified finger tip is used to implement the movement of a mouse and clicking gesture. The proposed algorithm was implemented with an actual smart device to test the proposed method. Empirical parameters were obtained from the keypad software and an image analysis tool for the performance optimization, and a comparative analysis with conventional research showed improved performance with the proposed method.





References:
[1] Y. Hirobe, T. Niikura, Y. Watanabe, T. Komuro, M. Ishikawa,
“Vision-based Input Interface for Mobile Devices with High-speed
Fingertip Tracking,“ Adj. Proc. ACM UIST 2009, pp. 7-8.
[2] Y. Takeoka et al.: Z-touch: an infrastructure for 3d gesture interaction in
the proximity of tabletop surfaces, Proceedings of ITS’10, 2010.
[3] Y. Tsukada et al.: Layerd touch panel: the input device with touch layers,
Proceedings of CHI’02, 2002, pp. 584-585.
[4] Intel Corporation. Open Source Computer Vision Library reference
manual. December 2000.
[5] T. Lee and T. Höllerer, "Handy AR: Markerless inspection of augmented
reality objects using fingertip tracking," International Symposium on
Wearable Computers, Citeseer, 2007, pp. 83-90.
[6] J. L. Rodgers and W. A. Nicewander, "Thirteen ways to look at the
correlation coefficient," American Statistician 42, 1988, pp. 59-66.
[7] Baker, S. and Matthews, I. Lucas-Kanade 20 Years On: A Unifying
Framework, International Journal of Computer Vision, 2004, vol.56, No3,
pp. 221-255.
[8] "List of Pangrams", http://en.wikipedia.org/wiki/Li st_of_pangrams.