Privacy preserving content based image retrieval

Maemoona Kayani, M Mohsin Riaz, Abdul Ghafoor, Fawad Khan

Research output: Contribution to journalArticlepeer-review


The rapid and tremendous growth of multimedia content has motivated users and organizations to upload their large scale digital content on the cloud server which makes the security of data an important issue. To overcome the problem of security in content based image retrieval (CBIR), we have proposed a novel privacy preserving image retrieval technique that efficiently retrieves similar images in an encrypted domain. Initially features of all the images in the database are extracted and then encrypted using secret key. In order to enhance the security, images are encrypted using Gauss-Square chaotic (GSC) map prior to outsourcing it to cloud server. Cloud uses modified euclidian distance (for encrypted features) to retrieve images similar to user encrypted query. Simulation results on Corel-1K and GHIM-10K illustrates that proposed technique efficiently retrieves encrypted images without compromising the retrieval accuracy. Moreover, the security analysis demonstrates the high security of proposed image encryption scheme.
Original languageEnglish
Pages (from-to)44955-44978
Number of pages24
JournalMultimedia Tools and Applications
Issue number15
Early online date19 Oct 2023
Publication statusPublished - May 2024
Externally publishedYes

Bibliographical note

Publisher Copyright:
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2023.


  • Multimedia security
  • Privacy
  • Image retrieval
  • Chaotic map


Dive into the research topics of 'Privacy preserving content based image retrieval'. Together they form a unique fingerprint.

Cite this