ETRI-Knowledge Sharing Plaform



논문 검색
구분 SCI
연도 ~ 키워드


학술대회 A Layered Features Analysis in Smart Farm Environments
Cited 17 time in scopus Download 5 time Share share facebook twitter linkedin kakaostory
박준용, 최장호, 이용주, 민옥기
International Conference on Big Data and Internet of Things (BDIOT) 2017, pp.169-173
17HS4400, (ICBMS-1세부) CoT(Cloud of Things) 환경에서 실시간 반응성 향상을 위한 계층적 데이터 스트림 분석 SW 기술 개발, 민옥기
Calculating or predicting the harvest yield of crops in agriculture has been one of its most important techniques. Smart farms, with the power of IoT, provide invaluable information such as real-time environment data such as temperature, humidity, etc. However, there are situations where not all data can be sent over the internet due to private reasons. In this paper, we propose a scalable data analysis framework: the edges preprocess and analyze the private data and send the results to server, and the server gathers and accumulates the results to estimate and predict the total harvest yield. Based on the results obtained from a real tomato farm, error rate is comparable to the one performed at the server only, but the number of features is reduced significantly.
KSP 제안 키워드
Data analysis, Environment data, Private data, Real-time environment, Smart farm, analysis framework, error rate, features analysis