Human activity recognition using skeleton data and support vector machine

In this paper, we propose a method for recognizing human activities using skeleton data by RGB-D camera, namely Kinect device. The human activity recognition is a learning in the field of computer vision. In its application, the recognition of human activity can be used for a sign language learning,...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Journal of physics. Conference series Ročník 1192; číslo 1; s. 12044 - 12051
Hlavní autoři: Komang, Mandira G A, Surya, Michrandi N, Ratna, Astuti N
Médium: Journal Article
Jazyk:angličtina
Vydáno: Bristol IOP Publishing 01.03.2019
Témata:
Algorithms
Computer vision
Human activity recognition
Image processing
Learning
Moving object recognition
Support vector machines
ISSN:1742-6588, 1742-6596
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:In this paper, we propose a method for recognizing human activities using skeleton data by RGB-D camera, namely Kinect device. The human activity recognition is a learning in the field of computer vision. In its application, the recognition of human activity can be used for a sign language learning, human-computer interaction, surveillance of the elderly, image processing and etc. Our approach is based on skeleton data with coordinate value of each joints in human body, that will be classified using support vector machine algorithm when performing a movement to predict the activities name. Experiments were performed with a new training data that we've create manual from capturing movement while human target are doing activities. Experiments result show that the system best average accuracy is 93.75% of all activities prediction with the optimal distance of object to the devices is 2 meters.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1742-6588
1742-6596
DOI:10.1088/1742-6596/1192/1/012044