Search

Computers have slowly begun to replace many of our day to day objects. Something like a laptop can easily become anything, given the right software. Numerous emulators and simulators have been developed in order to make the computer act like something completely different. This study attempts to create a virtual drum system that can be played using a webcam and a computer system alone. It utilizes the many functions of computer vision aided with the OpenCV library. Image processing techniques such as application of a Gaussian filter and thresholding were used in order to track the user's hand and successfully emulate the drum through tracking of its various movements. The contours extracted from the thresholding techniques are tracked by finding their respective bounding rectangles. The result was a system that could emulate a drum and can be played with the hands through striking the different areas of the drum. Although this method was quite effective and did produce the desired effect, the system does not allow the user's head to be present in the video as it may be detected and mistaken for a hand. Further developments to the system should address this, and should also include a solution for supporting foot functionalities such as bass drum and hi-hat control.