{"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00242217","sets":["1164:6757:11881:11882"]},"path":["11882"],"owner":"44499","recid":"242217","title":["ウェアラブルセンサのスペクトル可視化による転倒検出AIの開発と評価"],"pubdate":{"attribute_name":"公開日","attribute_value":"2025-01-16"},"_buckets":{"deposit":"cfcce760-e115-4fea-adde-f9cea23d414f"},"_deposit":{"id":"242217","pid":{"type":"depid","value":"242217","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"ウェアラブルセンサのスペクトル可視化による転倒検出AIの開発と評価","author_link":["668495","668496","668494","668493"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"ウェアラブルセンサのスペクトル可視化による転倒検出AIの開発と評価"},{"subitem_title":"Development and Evaluation of Fall Detection AI Using Spectral Visualization of Wearable Sensors","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"自然環境，農業および支援技術","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2025-01-16","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"静岡大学大学院総合科学技術研究科情報学専攻／ローランド ディー.ジー.株式会社"},{"subitem_text_value":"静岡大学学術院情報学領域／静岡大学グリーン科学技術研究所"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Department of Informatics, Graduate School of Integrated Science and Technology, Shizuoka University / Roland DG Corporation","subitem_text_language":"en"},{"subitem_text_value":"College of Informatics, Academic Institute, Shizuoka University / Research Institute of Green Science and Technology, Shizuoka University","subitem_text_language":"en"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_publisher":{"attribute_name":"出版者","attribute_value_mlt":[{"subitem_publisher":"情報処理学会","subitem_publisher_language":"ja"}]},"publish_status":"0","weko_shared_id":-1,"item_file_price":{"attribute_name":"Billing file","attribute_type":"file","attribute_value_mlt":[{"url":{"url":"https://ipsj.ixsq.nii.ac.jp/record/242217/files/IPSJ-DCC25039058.pdf","label":"IPSJ-DCC25039058.pdf"},"date":[{"dateType":"Available","dateValue":"2027-01-16"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-DCC25039058.pdf","filesize":[{"value":"1.7 MB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"660","billingrole":"5"},{"tax":["include_tax"],"price":"330","billingrole":"6"},{"tax":["include_tax"],"price":"0","billingrole":"50"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"da38e5f4-bca2-4342-9448-91ad1148caac","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2025 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"束野, 通洋"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"峰野, 博史"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Michihiro, Tsukano","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Hiroshi, Mineno","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AA12628338","subitem_source_identifier_type":"NCID"}]},"item_4_textarea_12":{"attribute_name":"Notice","attribute_value_mlt":[{"subitem_textarea_value":"SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc."}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_18gh","resourcetype":"technical report"}]},"item_4_source_id_11":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"2188-8868","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"高齢者の転倒は死に至る重大な要因であり，転倒後，早急に対応することが求められる．近年，深層学習を用いた転倒検出の研究が盛んである．精度向上のため，教師データを大量に収集することが求められるが，転倒データを収集することは危険を伴い，現実的に困難である．また，カメラ画像を使用する場合，設置環境に依存するため運用が制限されてしまうことや，プライバシーの問題が発生する．本研究では，異常検知手法を利用することで，正常な行動データのみを学習し，認識された異常な行動を転倒と分類する手法を検討する．教師データが正常な行動データのみであれば，膨大な量の教師データが期待できる．また，ウェアラブルセンサデータを用いることで，プライバシーを保護しつつ，突発的な転倒に対応することを目指す．本論文では，加速度センサとジャイロセンサのデータをスペクトル可視化し，畳み込みオートエンコーダを用いて特徴量を抽出後，OC-SVM で正常と転倒を分類した．また，複数のセンサに対応するために，センサごとに畳み込みオートエンコーダを構築した．評価では，元のセンサデータを画像化したものやスペクトログラム，スカログラムを用いるとともに，センサごとに構築した単体モデル，あるいは複数モデルによる性能を比較した．結果として，スカログラムを用いた複数モデルによる方法の F1-score が 0.915 で，最も有効であることを示した．","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"8","bibliographic_titles":[{"bibliographic_title":"研究報告デジタルコンテンツクリエーション（DCC）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2025-01-16","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"58","bibliographicVolumeNumber":"2025-DCC-39"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"updated":"2025-01-19T07:24:26.868381+00:00","created":"2025-01-19T01:47:17.598440+00:00","links":{},"id":242217}