Cargando…

Clothing identification via deep learning: forensic applications

Attribute-based identification systems are essential for forensic investigations because they help in identifying individuals. An item such as clothing is a visual attribute because it can usually be used to describe people. The method proposed in this article aims to identify people based on the vi...

Descripción completa

Detalles Bibliográficos
Autores principales: Bedeli, Marianna, Geradts, Zeno, van Eijk, Erwin
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Taylor & Francis 2018
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6201771/
https://www.ncbi.nlm.nih.gov/pubmed/30483672
http://dx.doi.org/10.1080/20961790.2018.1526251
_version_ 1783365567659376640
author Bedeli, Marianna
Geradts, Zeno
van Eijk, Erwin
author_facet Bedeli, Marianna
Geradts, Zeno
van Eijk, Erwin
author_sort Bedeli, Marianna
collection PubMed
description Attribute-based identification systems are essential for forensic investigations because they help in identifying individuals. An item such as clothing is a visual attribute because it can usually be used to describe people. The method proposed in this article aims to identify people based on the visual information derived from their attire. Deep learning is used to train the computer to classify images based on clothing content. We first demonstrate clothing classification using a large scale dataset, where the proposed model performs relatively poorly. Then, we use clothing classification on a dataset containing popular logos and famous brand images. The results show that the model correctly classifies most of the test images with a success rate that is higher than 70%. Finally, we evaluate clothing classification using footage from surveillance cameras. The system performs well on this dataset, labelling 70% of the test images correctly.
format Online
Article
Text
id pubmed-6201771
institution National Center for Biotechnology Information
language English
publishDate 2018
publisher Taylor & Francis
record_format MEDLINE/PubMed
spelling pubmed-62017712018-11-27 Clothing identification via deep learning: forensic applications Bedeli, Marianna Geradts, Zeno van Eijk, Erwin Forensic Sci Res Original Article Attribute-based identification systems are essential for forensic investigations because they help in identifying individuals. An item such as clothing is a visual attribute because it can usually be used to describe people. The method proposed in this article aims to identify people based on the visual information derived from their attire. Deep learning is used to train the computer to classify images based on clothing content. We first demonstrate clothing classification using a large scale dataset, where the proposed model performs relatively poorly. Then, we use clothing classification on a dataset containing popular logos and famous brand images. The results show that the model correctly classifies most of the test images with a success rate that is higher than 70%. Finally, we evaluate clothing classification using footage from surveillance cameras. The system performs well on this dataset, labelling 70% of the test images correctly. Taylor & Francis 2018-10-17 /pmc/articles/PMC6201771/ /pubmed/30483672 http://dx.doi.org/10.1080/20961790.2018.1526251 Text en © 2018 The Author(s). Published by Taylor & Francis Group on behalf of the Academy of Forensic Science. http://creativecommons.org/licenses/by/4.0/ This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
spellingShingle Original Article
Bedeli, Marianna
Geradts, Zeno
van Eijk, Erwin
Clothing identification via deep learning: forensic applications
title Clothing identification via deep learning: forensic applications
title_full Clothing identification via deep learning: forensic applications
title_fullStr Clothing identification via deep learning: forensic applications
title_full_unstemmed Clothing identification via deep learning: forensic applications
title_short Clothing identification via deep learning: forensic applications
title_sort clothing identification via deep learning: forensic applications
topic Original Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6201771/
https://www.ncbi.nlm.nih.gov/pubmed/30483672
http://dx.doi.org/10.1080/20961790.2018.1526251
work_keys_str_mv AT bedelimarianna clothingidentificationviadeeplearningforensicapplications
AT geradtszeno clothingidentificationviadeeplearningforensicapplications
AT vaneijkerwin clothingidentificationviadeeplearningforensicapplications