Lightdraw allows one or more users to communicate with the computer by moving light points on a display surface. The motions of these light points are then interpreted by the computer system to trigger appropriate responses on the display. The light tracking is done with a commodity-off-the-shelf webcam and the light points can be generated by a number of devices including laser pointers, torchlight and bright hand phone screens.
Please make sure you have read and understand our flexible licensing scheme before you download.
The preferred embodiment of the this invention is as illustrated in the video below.
A camera is positioned to face the large screen in such a position to view the whole display. The video stream is
then fed into a laser detection module that is able to separate light spots from the display projection.
The coordinates of the spots are then fed to any listening applications to be interpreted as pointers on the
This laser detection module is crucial in the Lightdraw project. To detect incident laser points on a projector
image, the raw image, captured by the camera, has to undergo a few stages of processing before it can be binarized.
Firstly, thresholding removes unwanted noise from the image before a colour conversion to a Hue, Saturation,
and Brightness (HSV) model is done. Working with a HSV color model is easier as the laser appears as brightly
colour pixels and has a very narrow range of hue. Next, a closing operation is used to repair gaps within the
laser stroke. Lastly, the blobs (a blob is a region of touching pixels) remaining on the image go through another
round of thresholding on their characteristics such size and perimeter. In essence this module helps the system
differentiate between the laser point and the light from the projector.
Post a Comment
Could not execute select query.