{"created":"2025-01-19T00:58:06.558917+00:00","updated":"2025-01-20T00:11:13.172505+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00192394","sets":["1164:3616:9424:9602"]},"path":["9602"],"owner":"44499","recid":"192394","title":["独立成分分析を用いたパーティクルフィルタによる歩行者の追跡"],"pubdate":{"attribute_name":"公開日","attribute_value":"2018-11-22"},"_buckets":{"deposit":"f89ac035-2abf-406b-ba9f-a97d63e8a194"},"_deposit":{"id":"192394","pid":{"type":"depid","value":"192394","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"独立成分分析を用いたパーティクルフィルタによる歩行者の追跡","author_link":["448628","448627","448629","448626"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"独立成分分析を用いたパーティクルフィルタによる歩行者の追跡"},{"subitem_title":"A pedestrian tracking system using ICA and a particle filter","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2018-11-22","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"室藺工業大学大学院工学研究科"},{"subitem_text_value":"室藺工業大学大学院工学研究科"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Muroran Institute of Technorogy","subitem_text_language":"en"},{"subitem_text_value":"Muroran Institute of Technorogy","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/192394/files/IPSJ-AVM18103024.pdf","label":"IPSJ-AVM18103024.pdf"},"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-AVM18103024.pdf","filesize":[{"value":"985.8 kB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"0","billingrole":"27"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_login","version_id":"48291b0c-9cca-4f85-b097-f521ac136559","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":"友成, 芳宏"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"鈴木, 幸司"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Yoshihiro, Tomonari","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Yukinori, Suzuki","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN10438399","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":"2188-8582","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"歩行者追跡システムは監視システムや交通モニタリングシステムなどのコンピュータビジョンの応用として広く研究されている．パーティクルフィルタは非線形 ・ 非ガウス状態空間モデルを対象としており歩行者追跡に有効である．動画像中の歩行者の特徴を捉える事が重要であり，本論文では ICA によって歩行者の幾何学的特徴を捉えることによりパーティクルフィルタを用いて歩行者を追跡するシステムを開発した．様々な人の画像から ICA の基底を求めた．人の画像ブロックはその基底の線形結合によって復元することが可能である．パーティクルフィルタのパーティクル位置の画像ブロックを復元する ICA の結合係数から構成される行列を作成し特異値ベクトルを求める．この特異値ベクトルからパーティクルフィルタの尤度を求めフィルタリングを行う．複数の歩行者から特定の歩行者を追跡する実験を行った結果，有効性を示すことはできなかったが，改善点を見出すことができた．","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"A pedestrian tracking system has been studied for various systems such as surveillance systems and traffic monitoring systems, which are application studies of computer vision. Since a particle filter is applicable to state space models with nonlinear and non-Gaussian, it is useful for tracking a pedestrian. To implement robust tracking, it is necessary to obtain features of the appearance of pedestrian in a video. ICA (independent component analysis) is able to obtain geometrical features of a pedestrian and is therefore useful for implementing a tracking system. ICA bases were computed from a variety of pedestrian images. An image block of a pedestrian image is reconstructed by linear combination of ICA bases. Weight coefficients to combine ICA bases can be features for the appearance of a pedestrian in a video. Image blocks were extracted from the image where a particle located. A matrix was constructed by weight coefficients for these image blocks. Vector consisting of singular values of the matrix were computed and they were used to compute the likelihood of a particle. A filtering operation was carried out for each particle using the likelihood. We conducted experiments in which a specific pedestrian among a number of pedestrians was tracked. Experimental results showed that the system was not able to accurately track pedestrians. However, we found how to improve the system.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"6","bibliographic_titles":[{"bibliographic_title":"研究報告オーディオビジュアル複合情報処理（AVM）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2018-11-22","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"24","bibliographicVolumeNumber":"2018-AVM-103"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":192394,"links":{}}