A deep learning-based method for Turkish text detection from videos

dc.contributor.authorRasheed, Jawad
dc.contributor.authorJamil, Akhtar
dc.contributor.authorDoğru, Hasibe Büşra
dc.contributor.authorTilki, Sahra
dc.contributor.authorYesiltepe, Mirsat
dc.contributor.authorRasheed, Jawad
dc.contributor.authorTilki, Sahra
dc.contributor.authorAytekin, Hasibe Büşra
dc.date.accessioned2020-12-20T06:49:50Z
dc.date.available2020-12-20T06:49:50Z
dc.date.issued2019
dc.departmentMühendislik ve Doğa Bilimleri Fakültesi
dc.description11th International Conference on Electrical and Electronics Engineering (ELECO) -- NOV 28-30, 2019 -- Bursa, TURKEY
dc.descriptionWOS:000552654100189
dc.description.abstractThe text appearing in videos provides useful information, which can be exploited for developing automatic video indexing and retrieval systems. In this study, we integrated a heuristic and a deep learning-based method using Convolutional Neural Network (CNN) for automatic text extraction from videos. The two independent steps used for text extraction are; candidate text region detection and classification. In first step, rectangular regions were detected that potentially contain text by applying heuristics, which includes morphological processing and geometrical constraints. Then, the obtained candidate text regions were passed through several layers of CNN, that first produced convolutional feature map and then classified the candidate regions into either text or not-text classes. A dataset was prepared by collecting videos from various Turkish channels. 70% of the data was used to train the network while 30% for validation. Experiments showed that our proposed method achieved state-of-the-art performance on our dataset.
dc.description.sponsorshipChamber Elect Engineers Bursa Branch, Bursa Uludag Univ, Dept Elect Elect Engn, Istanbul Tech Univ, Fac Elect & Elect Engn, IEEE Turkey Sect
dc.identifier.doi10.23919/ELECO47770.2019.8990633
dc.identifier.endpage939
dc.identifier.orcidJawad Rasheed |0000-0003-3761-1641
dc.identifier.orcidAkhtar Jamil |0000-0002-2592-1039
dc.identifier.orcidHasibe Büşra Doğru |0000-0002-5944-260X
dc.identifier.scopusqualityN/A
dc.identifier.startpage935
dc.identifier.urihttps://hdl.handle.net/20.500.12436/1837
dc.identifier.urihttps://www.doi.org/10.23919/ELECO47770.2019.8990633
dc.identifier.wosqualityN/A
dc.indekslendigikaynakWeb of Science
dc.indekslendigikaynakScopus
dc.institutionauthorRasheed, Jawad
dc.institutionauthorJamil, Akhtar
dc.institutionauthorDoğru, Hasibe Büşra
dc.institutionauthorTilki, Sahra
dc.language.isoen
dc.publisherIeee
dc.relation.ispartof2019 11Th International Conference On Electrical And Electronics Engineering (Eleco 2019)
dc.relation.publicationcategoryKonferans Öğesi - Uluslararası - Kurum Öğretim Elemanı
dc.rightsinfo:eu-repo/semantics/closedAccess
dc.titleA deep learning-based method for Turkish text detection from videos
dc.typeConference Object
dspace.entity.typePublication
relation.isAuthorOfPublicationf9b9b46c-d923-42d3-b413-dd851c2e913a
relation.isAuthorOfPublication4b84ec9f-70a0-43dd-b9db-561485fcbff1
relation.isAuthorOfPublication64fe8bf9-38f7-4501-b4c2-e31ac4205720
relation.isAuthorOfPublication.latestForDiscoveryf9b9b46c-d923-42d3-b413-dd851c2e913a

Dosyalar

Orijinal paket

Listeleniyor 1 - 1 / 1
Yükleniyor...
Küçük Resim
İsim:
A_Deep_Learning-based_Method_for_Turkish_Text_Detection_from_Videos.pdf
Boyut:
528.39 KB
Biçim:
Adobe Portable Document Format
Açıklama:
Proceeding File