Open AccessThe Prototype of Hand Gesture Recognition for Elderly People to Control Connected Home Devices
Author(s) -
Shalahudin Al Ayubi,
Dodi Wisaksono Sudiharto,
Erwid Musthofa Jadied,
Endro Aryanto
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1201/1/012042
Subject(s) - gesture , gesture recognition , computer science , controller (irrigation) , human–computer interaction , automation , feature (linguistics) , object (grammar) , home automation , control (management) , artificial intelligence , computer vision , engineering , operating system , mechanical engineering , linguistics , philosophy , agronomy , biology
Nowadays, the technology development makes a human can create a tool which is used to recognize an object and it becomes a popular technology device. It happens because this tool has an important role for interaction between a human and a computer. One example of this technology usage is to recognize a hand gesture for controlling a home automation system. The existing of this technology creates the change related to how the human controls any tools in a house and it also reduces the complexity including an effort when it is used for controlling. This feature is very useful, especially for elderly people who stay in independent living. This study is going to develop a controller prototype by using FAST (Features from Accelerated Segment Test) algorithm to detect hand gesture for operating the connected home devices. This controller uses an embedded system to translate a command which is created by using the hand gesture of senior captured by the cam for controlling the lamps. The lamps itself are represented as several tools in the house. The observation gives a result that the hand gesture is potential to be implemented as a command for controlling the proposed system prototype in the range which is not far than 1 meter with the percentage average recognition accuracy is almost 80%.