{"id":199048,"updated":"2025-01-19T21:48:15.606965+00:00","links":{},"created":"2025-01-19T01:03:10.154919+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00199048","sets":["6164:6165:8377:9886"]},"path":["9886"],"owner":"44499","recid":"199048","title":["敵対的学習に基づくドメイン適応によるドライブレコーダを用いたヒヤリハット検出及び分類"],"pubdate":{"attribute_name":"公開日","attribute_value":"2019-09-01"},"_buckets":{"deposit":"8eade7fb-6073-42d8-9abb-a9141267b132"},"_deposit":{"id":"199048","pid":{"type":"depid","value":"199048","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"敵対的学習に基づくドメイン適応によるドライブレコーダを用いたヒヤリハット検出及び分類","author_link":["481329","481333","481330","481332","481331"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"敵対的学習に基づくドメイン適応によるドライブレコーダを用いたヒヤリハット検出及び分類"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"モビリティとデータ処理","subitem_subject_scheme":"Other"}]},"item_type_id":"18","publish_date":"2019-09-01","item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_18_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"日本電信電話株式会社NTTサービスエボリューション研究所"},{"subitem_text_value":"日本電信電話株式会社NTTサービスエボリューション研究所"},{"subitem_text_value":"日本電信電話株式会社NTTサービスエボリューション研究所"},{"subitem_text_value":"日本電信電話株式会社NTTサービスエボリューション研究所"},{"subitem_text_value":"日本電信電話株式会社NTTサービスエボリューション研究所"}]},"item_18_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"NTT Service Evolution Laboratories, NTT Corporation","subitem_text_language":"en"},{"subitem_text_value":"NTT Service Evolution Laboratories, NTT Corporation","subitem_text_language":"en"},{"subitem_text_value":"NTT Service Evolution Laboratories, NTT Corporation","subitem_text_language":"en"},{"subitem_text_value":"NTT Service Evolution Laboratories, NTT Corporation","subitem_text_language":"en"},{"subitem_text_value":"NTT Service Evolution Laboratories, NTT Corporation","subitem_text_language":"en"}]},"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":44499,"item_file_price":{"attribute_name":"Billing file","attribute_type":"file","attribute_value_mlt":[{"url":{"url":"https://ipsj.ixsq.nii.ac.jp/record/199048/files/IPSJ-WebDBF2019003.pdf","label":"IPSJ-WebDBF2019003.pdf"},"date":[{"dateType":"Available","dateValue":"2021-09-01"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-WebDBF2019003.pdf","filesize":[{"value":"1.9 MB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"330","billingrole":"5"},{"tax":["include_tax"],"price":"0","billingrole":"6"},{"tax":["include_tax"],"price":"0","billingrole":"13"},{"tax":["include_tax"],"price":"0","billingrole":"39"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"10307d84-c85e-432e-a374-66de0f4d159b","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2019 by the Information Processing Society of Japan"}]},"item_18_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"瀧本, 祥章"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"山本, 修平"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"松林, 達史"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"倉島, 健"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"戸田, 浩之"}],"nameIdentifiers":[{}]}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_5794","resourcetype":"conference paper"}]},"item_18_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"ドライブレコーダデータから交通事故やヒヤリハットなどの危険な状態を有用なデータとして抽出，分類し，その利活用が行われている．しかしながら，それらのデータを自動的に抽出，分類するには多量の教師データを人手で用意する必要があり，高コストである．また，既存の教師ありデータを用いて学習したモデルを適用することも考えられるが，車種やカメラの種類などの収集時の条件の違いによって，データ集合全体の傾向が大きく変化し，その精度は制限される．そこで本稿では，分類対象のデータと異なる環境で収集されたデータセットとその教師データを元に，高精度な分類モデルの構築を目指す．具体的には，既存の Convolutional Recurrent Neural Networks ベースのヒヤリハット分類手法をベースに，収集した環境を推定する層を追加することで，敵対的学習によるドメイン適応を行う．これによって，従来は収集した環境に大きく依存した特徴量を抽出する CNN 部分について，環境に依存しない特徴量の抽出が期待できる．実験では，実際のドライブレコーダデータを用いて検証を行い，ドメイン適応によって教師データがない環境下のデータ集合に対しても，高い分類精度を持つモデルを構築できることを示した．","subitem_description_type":"Other"}]},"item_18_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"12","bibliographic_titles":[{"bibliographic_title":"WebDB Forum 2019論文集"}],"bibliographicPageStart":"9","bibliographicIssueDates":{"bibliographicIssueDate":"2019-09-01","bibliographicIssueDateType":"Issued"},"bibliographicVolumeNumber":"2019"}]},"relation_version_is_last":true,"weko_creator_id":"44499"}}