Slideshow presentation control through hand pattern recognition using web camera

Slideshow presentations are an efficient and effective way to present lectures of all topics. Unfortunately, the lecturer has to stay in close proximity to the device they are using in order to manipulate the presentation. The lecturer may choose to obtain a clicker, but it is then bound by dependence on batteries, and has an additional cost. In order to solve this problem, the researchers propose a system which allows the user to manipulate slideshow presentations with the use of hand patterns captured through the device's built-in camera. The hand in use is to wear a green glove to isolate the region of interest by color segmentation. The hand detection and recognition system are executed in MATLAB and its features are extracted using Principal Component Analysis (PCA). The classification of the pattern is done through the use of the K-Nearest Neighbor Algorithm (KNN). Once the hand pattern is identified, it will be converted to a specific keyboard shortcut in the slideshow presentation software and the command will be executed.