Hand Gesture Recognition using Leap Motion Controller for Recognition of Javanese Script

Abstract

Javanese Script is The cultural heritage of ancient Javanese people used to write in the manufacture

of books, manuscripts, song Java, inscriptions, or correspondence. It is important to maintain and preserve this

heritage, so that the Indonesian people can understand the cultural values and art writing. The development of

technology is currently increasing rapid, making some researchers use Leap Motion as a translator device. In

this study, a hand gesture recognition system using a leap motion controller is proposed for the recognition of

Javanese Script. By using a leap motion controller, every bone coordinate point on the hand will be detected, so

that these coordinate points can be used as input features. The use of the feature of the distance between the

palm position and distal phalanges measured using the Euclidean distance. These features will then be classified

using the k-nearest neighbor classification method. System testing was carried out using 20 Javanese characters

consisting of /ha/, /na/, /ca/, /ra/, /ka/, /da/, /ta/, /sa/, /wa/, /la/ , /pa/, /dha/, /ja/, /ya/, /nya/, /ma/, /ga/, /ba/, /tha/,

/nga/. This research was successful with an average accuracy rate of 95% in Javanese script.