Selective concealment of characters for privacy protection

Kohei Inai, Mårten Pålsson, Volkmar Frinken, Yaokai Feng, Seiichi Uchida

Research output: Chapter in Book/Report/Conference proceedingConference contribution

12 Citations (Scopus)

Abstract

A method of concealing characters is proposed for degrading legibility of privacy sensitive textual information in natural scene, such as car license plate numbers and name tags. An important property of the proposed method is that it realizes selective concealing of characters, that is, the proposed method degrades legibility of character regions without degrading the quality of non-character regions. This selective concealment is realized because characters have special concealment characteristics. Specifically, character legibility can be degraded by damaging stroke structure by using exemplar-based image in painting, which does not affect non-character regions. Experimental results of qualitative and quantitative evaluations have proven that the selective concealment is practically possible. Furthermore, the quantitative evaluation through a subjective experiment revealed appropriate setups of image in painting for maximizing selective concealment performance.

Original languageEnglish
Title of host publicationProceedings - International Conference on Pattern Recognition
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages333-338
Number of pages6
ISBN (Electronic)9781479952083
DOIs
Publication statusPublished - Dec 4 2014
Event22nd International Conference on Pattern Recognition, ICPR 2014 - Stockholm, Sweden
Duration: Aug 24 2014Aug 28 2014

Publication series

NameProceedings - International Conference on Pattern Recognition
ISSN (Print)1051-4651

Other

Other22nd International Conference on Pattern Recognition, ICPR 2014
Country/TerritorySweden
CityStockholm
Period8/24/148/28/14

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition

Fingerprint

Dive into the research topics of 'Selective concealment of characters for privacy protection'. Together they form a unique fingerprint.

Cite this