{"created":"2025-01-19T00:57:15.612377+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00191367","sets":["1164:4619:9352:9559"]},"path":["9559"],"owner":"11","recid":"191367","title":["多様な動作パターンを有する機器に対応した異常度推定手法の提案"],"pubdate":{"attribute_name":"公開日","attribute_value":"2018-09-13"},"_buckets":{"deposit":"0166f310-5aca-49e3-a482-8c9ad3560a56"},"_deposit":{"id":"191367","pid":{"type":"depid","value":"191367","revision_id":0},"owners":[11],"status":"published","created_by":11},"item_title":"多様な動作パターンを有する機器に対応した異常度推定手法の提案","author_link":["441141","441142","441144","441143","441146","441145"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"多様な動作パターンを有する機器に対応した異常度推定手法の提案"},{"subitem_title":"Anomaly Detection for Various Operations of Machine","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"ディスカッションセッション4","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2018-09-13","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"沖電気工業株式会社経営基盤本部研究開発センター"},{"subitem_text_value":"沖電気工業株式会社経営基盤本部研究開発センター"},{"subitem_text_value":"沖電気工業株式会社経営基盤本部研究開発センター"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"OKI Electric Industry Co., Ltd.","subitem_text_language":"en"},{"subitem_text_value":"OKI Electric Industry Co., Ltd.","subitem_text_language":"en"},{"subitem_text_value":"OKI Electric Industry Co., Ltd.","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/191367/files/IPSJ-CVIM18213022.pdf","label":"IPSJ-CVIM18213022.pdf"},"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-CVIM18213022.pdf","filesize":[{"value":"1.2 MB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"0","billingrole":"20"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_login","version_id":"c4ce5f8c-83bd-4710-92cc-97950ce02321","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2018 by the Institute of Electronics, Information and Communication Engineers This SIG report is only available to those in membership of the SIG."}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"小林, 一樹"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"関根, 理敏"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"伊加田, 恵志"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Kazuki, Kobayashi","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Masatoshi, Sekine","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Satoshi, Ikada","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AA11131797","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":"2188-8701","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"本稿では，多様な動作パターンを持つ機器を対象に，機器の動作異常度を効率的に推定する手法を提案する．提案手法は (1) 1 つの同じモデルから各時刻および各周波数について振動特徴量を抽出するような特徴量抽出モデルを自動作成することと，(2) 時刻および周波数帯ごとに異常度の推移を数値化して表すことの 2 点を特長とする.提案手法の動作手順の概要は次の通りである．まず，振動や音響などの時系列センサデータを周波数解析することでスペクトログラムを取得する．次に，正常と異常の各スペクトログラムを時間軸方向に平均化した情報から深層学習を利用した特徴量抽出モデルを作成する．この特徴量抽出モデルに対し，スペクトログラムを単位時刻および単位周波数帯ごとに取り出した情報を入力することで，時刻 ・ 周波数帯ごとに振動特徴量が得られる．最後に，「時刻と振動特徴量」 および 「周波数と振動特徴量」 の関係を確率モデルで表し，このモデルを利用して異常度推定を行う．また，評価実験として，多様な動作を行う機器から提案手法によって抽出した振動特徴量を用いて動作異常度の推定を行い，その有効性を確認する．評価実験の結果，「時刻と異常度」 および 「周波数と異常度」 の関係を数値化すると同時に，異常度が大きくなる時刻 ・ 周波数が機器の摩耗したギアの動作特性に関係することを確認した．","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"In this paper, we propose an anomaly level estimation method for various operation of machine. Our proposed method has two main functions: 1) making feature extraction models automatically in both time and frequency domains, 2) estimating anomaly level in each time and frequency. The outline of our proposed method is as follows. First, the spectrogram of sensor data such as vibration data and sound data are calculated. Second, the features of them are extracted from the summarized information using deep learning. By inputting the spectrogram divided for each time or each frequency, we can obtain the vibration features for each time or each frequency. Finally, the probability density model for anomaly level estimation is built. It explains the probability of occurrence of the vibration feature at a time or a frequency.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"6","bibliographic_titles":[{"bibliographic_title":"研究報告コンピュータビジョンとイメージメディア（CVIM）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2018-09-13","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"22","bibliographicVolumeNumber":"2018-CVIM-213"}]},"relation_version_is_last":true,"weko_creator_id":"11"},"id":191367,"updated":"2025-01-20T00:43:11.953425+00:00","links":{}}