{"title":"Improvement of activity recognition for child growth monitoring system at kindergarten","authors":"Shuehi Kurashima, Satoshi Suzuki","doi":"10.1109/IECON.2015.7392494","DOIUrl":null,"url":null,"abstract":"Activity recognition is possible to be used to understand behavior of infants and effective for health and development of children. In our previous study, action of children were assumed as sleeping, sitting, hand motion, walking, running, and playing. Revise of action items were, however, required because it was difficult to calculate calorie consumption in the previous. In this study, 11 action items are newly assumed, and accuracies of the activity recognition were investigated. As a result, accuracies of 4 items among 11 items were improved over than 60%, although accuracies of other 7 items were less than 60%.","PeriodicalId":190550,"journal":{"name":"IECON 2015 - 41st Annual Conference of the IEEE Industrial Electronics Society","volume":"90 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2015-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"6","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IECON 2015 - 41st Annual Conference of the IEEE Industrial Electronics Society","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/IECON.2015.7392494","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 6
Abstract
Activity recognition is possible to be used to understand behavior of infants and effective for health and development of children. In our previous study, action of children were assumed as sleeping, sitting, hand motion, walking, running, and playing. Revise of action items were, however, required because it was difficult to calculate calorie consumption in the previous. In this study, 11 action items are newly assumed, and accuracies of the activity recognition were investigated. As a result, accuracies of 4 items among 11 items were improved over than 60%, although accuracies of other 7 items were less than 60%.