Secure content based image retrieval in medical databases

Abstract : In this paper, we propose an implementation in the encrypted domain of a content based image retrieval (CBIR) method. It allows a physician to retrieve the most similar images to a query image in an outsourced database while preserving data confidentiality. Image retrieval is based on image signatures we build in the hormomorphically encrypted wavelet transform domain. Experimental results show it is possible to achieve retrieval performance as good as if images were processed nonencrypted.
Document type :
Conference papers
Complete list of metadatas

https://www.hal.inserm.fr/inserm-01154957
Contributor : Frédérique Frouin <>
Submitted on : Monday, May 25, 2015 - 7:00:51 PM
Last modification on : Thursday, October 17, 2019 - 12:33:26 PM
Long-term archiving on: Thursday, April 20, 2017 - 8:02:40 AM

File

ACTES-RITS2015-HAL-p114-5.pdf
Files produced by the author(s)

Licence


Distributed under a Creative Commons Attribution 4.0 International License

Identifiers

  • HAL Id : inserm-01154957, version 1

Citation

Reda Bellafqira, Gouenou Coatrieux, Dalel Bouslimi, Gwénolé Quellec. Secure content based image retrieval in medical databases. Journées RITS 2015, Mar 2015, Dourdan, France. pp 114-115. ⟨inserm-01154957⟩

Share

Metrics

Record views

241

Files downloads

103