Klasifikasi K-NN dan Naive Bayes Terhadap Pelacakan Ujung Jari Berbasis Camera Smartphone

Abstrak

Research vision and computer graphics, motion tracking fingertips of video image sequences automatically has been a very interesting study to be developed. Good tracking method should be able to find back the ends of the finger after the obstacle no longer exists and is also able to predict the fingertips are hindered by the position information of the fingertips that are not obstructed. In practice, there are two ways to track the movement of an object. The first approach is called the approach-by-detection tracking. In this approach, to track an object the object detection performed on each frame of moving pictures or videos are observed, in order to determine the position of the object in each frame. The second approach is referred to as a detection approach-by-tracking. From the results of tracking the movement of a hand camera-based smartphone will use a method of classification algorithm k-Nearest Neighborhood (k-NN) and Naive Bayes (NB). From the research results will be known influence on the movement of fingertips, and penagkapan fingertips as well as the classification method to obtain the arrest of the movement of a fingertip accuracy, algorithms and data from a tracking number will be tested for accuracy. The best results of testing of the algorithms will be the result and purpose of this research.