Mostrar el registro sencillo del ítem
dc.contributor | Escuela de Ingenierias Industrial, Informática y Aeroespacial | es_ES |
dc.contributor.author | Blanco Medina, Pablo | |
dc.contributor.author | Fidalgo Fernández, Eduardo | |
dc.contributor.author | Alegre Gutiérrez, Enrique | |
dc.contributor.author | Alaiz Rodríguez, Rocío | |
dc.contributor.author | Jáñez Martino, Francisco | |
dc.contributor.author | Bonnici, Alexandra | |
dc.contributor.other | Ingenieria de Sistemas y Automatica | es_ES |
dc.date | 2020-10-16 | |
dc.date.accessioned | 2024-05-06T09:12:01Z | |
dc.date.available | 2024-05-06T09:12:01Z | |
dc.identifier.citation | Blanco-Medina, P., Fidalgo, E., Alegre, E., Alaiz-Rodríguez, R., Jáñez-Martino, F., & Bonnici, A. (2020). Rectification and super-resolution enhancements for forensic text recognition. Sensors (Switzerland), 20(20), 1-17. https://doi.org/10.3390/S20205850 | es_ES |
dc.identifier.other | https://www.mdpi.com/1424-8220/20/20/5850 | es_ES |
dc.identifier.uri | https://hdl.handle.net/10612/20387 | |
dc.description.abstract | [EN] Retrieving text embedded within images is a challenging task in real-world settings. Multiple problems such as low-resolution and the orientation of the text can hinder the extraction of information. These problems are common in environments such as Tor Darknet and Child Sexual Abuse images, where text extraction is crucial in the prevention of illegal activities. In this work, we evaluate eight text recognizers and, to increase the performance of text transcription, we combine these recognizers with rectification networks and super-resolution algorithms. We test our approach on four state-of-the-art and two custom datasets (TOICO-1K and Child Sexual Abuse (CSA)-text, based on text retrieved from Tor Darknet and Child Sexual Exploitation Material, respectively). We obtained a 0.3170 score of correctly recognized words in the TOICO-1K dataset when we combined Deep Convolutional Neural Networks (CNN) and rectification-based recognizers. For the CSA-text dataset, applying resolution enhancements achieved a final score of 0.6960. The highest performance increase was achieved on the ICDAR 2015 dataset, with an improvement of 4.83% when combining the MORAN recognizer and the Residual Dense resolution approach. We conclude that rectification outperforms super-resolution when applied separately, while their combination achieves the best average improvements in the chosen datasets. | es_ES |
dc.language | eng | es_ES |
dc.publisher | MDPI | es_ES |
dc.rights | Atribución 4.0 Internacional | * |
dc.rights.uri | http://creativecommons.org/licenses/by/4.0/ | * |
dc.subject | Cibernética | es_ES |
dc.subject | Informática | es_ES |
dc.subject.other | Text spotting | es_ES |
dc.subject.other | Text recognition | es_ES |
dc.subject.other | Super-resolution | es_ES |
dc.subject.other | Tor Darknet | es_ES |
dc.subject.other | Computer forensics | es_ES |
dc.title | Rectification and Super-Resolution Enhancements for Forensic Text Recognition | es_ES |
dc.type | info:eu-repo/semantics/article | es_ES |
dc.identifier.doi | 10.3390/s20205850 | |
dc.description.peerreviewed | SI | es_ES |
dc.rights.accessRights | info:eu-repo/semantics/openAccess | es_ES |
dc.identifier.essn | 1424-8220 | |
dc.journal.title | Sensors | es_ES |
dc.volume.number | 20 | es_ES |
dc.issue.number | 20 | es_ES |
dc.page.initial | 5850 | es_ES |
dc.type.hasVersion | info:eu-repo/semantics/publishedVersion | es_ES |
dc.subject.unesco | 3304.05 Sistemas de Reconocimiento de Caracteres | es_ES |
dc.subject.unesco | 1207.03 Cibernética | es_ES |
dc.subject.unesco | 1203.17 Informática | es_ES |
dc.description.project | Instituto Nacional de Ciberseguridad | es_ES |
dc.description.project | This research has been funded with support from the European Commission under the 4NSEEK project with Grant Agreement 821966. This publication reflects the views only of the author, and the European Commission cannot be held responsible for any use that may be made of the information contained therein. | es_ES |
dc.description.project | This research has been supported by the grant ’Ayudas para la realización de estudios de doctorado en el marco del programa propio de investigación de la Universidad de León Convocatoria 2018’ and by the framework agreement between Universidad de León and INCIBE (Spanish National Cybersecurity Institute) under Addendum 01. We acknowledge NVIDIA Corporation with the donation of the Titan Xp GPU used for this research. | es_ES |
Ficheros en el ítem
Este ítem aparece en la(s) siguiente(s) colección(ones)
-
Artículos [5530]