講演名 2021-03-04
A Study of Product Identification System Using Optical Character Recognition
陳 仕璽(岡山大), 舩曵 信生(岡山大), 坂上 暢規(岡山大), 土信田 高(アストロラボ), 菅 恒平(アストロラボ),
PDFダウンロードページ PDFダウンロードページへ
抄録(和) Recently, the optical character recognition (OCR) technology has been remarkably progressed due to the advancements of deep learning techniques. Besides, smartphones equipped with cameras have broadly spread among people around the world. As a result, the product identification from the product label photo using OCR becomes possible as the quick way to identify the product. However, the accuracy of OCR is still not 100%. Some characters are incorrectly recognized or missing in the recognition result, which must be considered for use. In this study, we propose a product identification system applying OCR of the label photo taken by a smart phone. The fuzzy search is adopted to improve the accuracy by finding the best-matching record in the database for the possibly incorrect key by OCR. Since this search takes inadmissibly long time when the database has a lot of records, we also propose the speedup method by limiting the matching records. For evaluations, we apply the proposal to 389 label photos. The results show that the CPU time is 15.39sec by the na?ve search, and 0.99sec by the speedup one that limits the number of records to be searched into 0.24% of the na?ve one, where the record hit rate is slightly reduced from 94.3% to 94.1%.
抄録(英) Recently, the optical character recognition (OCR) technology has been remarkably progressed due to the advancements of deep learning techniques. Besides, smartphones equipped with cameras have broadly spread among people around the world. As a result, the product identification from the product label photo using OCR becomes possible as the quick way to identify the product. However, the accuracy of OCR is still not 100%. Some characters are incorrectly recognized or missing in the recognition result, which must be considered for use. In this study, we propose a product identification system applying OCR of the label photo taken by a smart phone. The fuzzy search is adopted to improve the accuracy by finding the best-matching record in the database for the possibly incorrect key by OCR. Since this search takes inadmissibly long time when the database has a lot of records, we also propose the speedup method by limiting the matching records. For evaluations, we apply the proposal to 389 label photos. The results show that the CPU time is 15.39sec by the na?ve search, and 0.99sec by the speedup one that limits the number of records to be searched into 0.24% of the na?ve one, where the record hit rate is slightly reduced from 94.3% to 94.1%.
キーワード(和) product identification / OCR / fuzzy search / regular expression / partial word matching
キーワード(英) product identification / OCR / fuzzy search / regular expression / partial word matching
資料番号 LOIS2020-48
発行日 2021-02-25 (LOIS)

研究会情報
研究会 LOIS
開催期間 2021/3/4(から1日開催)
開催地(和) オンライン開催
開催地(英) Online
テーマ(和) ライフログ活用技術、オフィスインフォメーションシステム、ライフインテリジェンス、および一般
テーマ(英)
委員長氏名(和) 小林 透(長崎大)
委員長氏名(英) Toru Kobayashi(Nagasaki Univ.)
副委員長氏名(和) 戸田 浩之(NTT)
副委員長氏名(英) Hiroyuki Toda(NTT)
幹事氏名(和) 永徳 真一郎(NTT) / 荒井 研一(長崎大学)
幹事氏名(英) Shinichiro Eitoku(NTT) / Kenichi Arai(Nagasaki Univ.)
幹事補佐氏名(和) 藤村 滋(NTT)
幹事補佐氏名(英) Shigeru Fujimura(NTT)

講演論文情報詳細
申込み研究会 Technical Committee on Life Intelligence and Office Information Systems
本文の言語 ENG
タイトル(和)
サブタイトル(和)
タイトル(英) A Study of Product Identification System Using Optical Character Recognition
サブタイトル(和)
キーワード(1)(和/英) product identification / product identification
キーワード(2)(和/英) OCR / OCR
キーワード(3)(和/英) fuzzy search / fuzzy search
キーワード(4)(和/英) regular expression / regular expression
キーワード(5)(和/英) partial word matching / partial word matching
第 1 著者 氏名(和/英) 陳 仕璽 / Shixi Chen
第 1 著者 所属(和/英) 岡山大学(略称:岡山大)
Okayama University(略称:Okayama Univ.)
第 2 著者 氏名(和/英) 舩曵 信生 / Nobuo Funabiki
第 2 著者 所属(和/英) 岡山大学(略称:岡山大)
Okayama University(略称:Okayama Univ.)
第 3 著者 氏名(和/英) 坂上 暢規 / Masaki Sakagami
第 3 著者 所属(和/英) 岡山大学(略称:岡山大)
Okayama University(略称:Okayama Univ.)
第 4 著者 氏名(和/英) 土信田 高 / Takashi Toshida
第 4 著者 所属(和/英) アストロラボ株式会社(略称:アストロラボ)
Astrolab Inc.(略称:Astrolab)
第 5 著者 氏名(和/英) 菅 恒平 / Kohei Suga
第 5 著者 所属(和/英) アストロラボ株式会社(略称:アストロラボ)
Astrolab Inc.(略称:Astrolab)
発表年月日 2021-03-04
資料番号 LOIS2020-48
巻番号(vol) vol.120
号番号(no) LOIS-417
ページ範囲 pp.6-11(LOIS),
ページ数 6
発行日 2021-02-25 (LOIS)