06-09-2010, 06:08 PM
SANKET: Interprets your Hand Gestures
INTRODUCTION
Hand gestures provides a very good alternative to cumbersome interface devices
for human-computer interaction (HCI). a real-time system capable of understanding commands given by Hand gestures is described in this article. all basic commands required by a human-computer interface must be given by this technique. The major function provided includes:
-Dual Mouse pointer motion
-More than ten motion-based gestures which is more than enough to operate a browser
-Static gestures enabling the input of digits etc.
-interfacing of external hardware can be done via ports
BENEFITS OF THE PROJECT
continuous and improper use of keyboard and mouse can cause many injuries. This system allows to do the basic operations on a computer with ease. physically challenged persons will greatly benefit from this.
INNOVATION
Products offering hand gesture recognition is very less in the market although there exist several algorithms for this purpose.
SYSTEM ORGANIZATION
a camera(supports streaming video input ) is the only hardware connected externally. OpenCV2 library is used. The The C++ code can be used on any platform.
PRINCIPLES OF OPERATION
both static and motion-based gestures are intended to be recognised.
The various steps of the algorithm are described below:
1)Hand Segmentation:
The region of interest, i.e., hands have to be localized.
2) Affine transforms for hand regions between successive image frames are estimated using local descent criterion.
3)Determining intended hand gesture from the feature vector.
4)Static hand Gestures:
Gestures like OK, Cancel and numbers for m 1 through 5 can be recognised,
For more details, visit these pdf:
http://stanford.edu/~asaxena/resources/p...sanket.pdf
http://iaengpublication/IMECS2008/IMECS2008_pp388-393.pdf
