Key-device based place recognition using similarity measure between IoT spaces = 사물인터넷 공간 간 유사도를 이용한 키 디바이스 기반 장소 인식 기법

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 323
  • Download : 0
DC FieldValueLanguage
dc.contributor.advisorLee, Young Hee-
dc.contributor.authorHong, In Taek-
dc.description학위논문(석사) - 한국과학기술원 : 전산학부, 2016.2 ,[iv, 17 p. :]-
dc.description.abstractPlace is one of the important context information to provide context aware services to the users in smart environments. Many researchers in mobile robotics community or context-aware service system have suggested various solutions using vision data processing or additional geographical information such as Google Maps. However, recognizing the place by processing vision and image data must be too big burden to IoT devices. Geographical information such as a building structure drawing or a blueprint of a building may not available for IoT devices. As the number of Internet of Things (IoTs) grows explosively, limitations of existing approaches can be solved with the IoT environment since IoT devices in a place can detect each other easily. In this paper, we propose a novel approach that utilizes Key-Device Values and similarity measure between the devices to recognize a place. Our proposed approach measures similarity between an IoT space input and representative feature set of each known place a priory and determines the most probable place of the IoT space. We evaluate with Euclidean, cosine and weighted cosine similarity. Experimental results show that Key-Device concept remarkably improves precision and recall of place recognition.-
dc.subjectInternet of Things-
dc.subjectPlace recognition-
dc.subjectCosine similarity-
dc.subjectSimilarity measure-
dc.subjectKey-Device value-
dc.subject장소 인식-
dc.subject코사인 유사도-
dc.subject유사도 측정 기법-
dc.subject키 디바이스 값-
dc.titleKey-device based place recognition using similarity measure between IoT spaces = 사물인터넷 공간 간 유사도를 이용한 키 디바이스 기반 장소 인식 기법-
dc.description.department한국과학기술원 :전산학부,-
Appears in Collection
Files in This Item
There are no files associated with this item.


  • mendeley


rss_1.0 rss_2.0 atom_1.0