Color image watermarking based on reflectance component modification and guided image filtering

Journal article

Authors/Editors

THUMRONGRAT AMORNRAKSA

Strategic Research Themes

Other (Strategic Research Themes)

Publication Details

Author list: Chotikawanid, Piyanart; Amornraksa, Thumrongrat;

Publisher: Springer

Publication year: 2021

Volume number: 80

Issue number: 18

Start page: 27615

End page: 27648

Number of pages: 34

ISSN: 1380-7501

eISSN: 1573-7721

URL: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85106403157&doi=10.1007%2fs11042-021-10756-9&partnerID=40&md5=a57fef2b3c2f5f7d8ecd50bcfe6ab1ba

Languages: English-Great Britain (EN-GB)

View in Web of Science | View on publisher site | View citing articles in Web of Science

Abstract

Digital images available on the Internet can be effortlessly copied and redistributed. Many image watermarking methods have been developed and are used as technical solutions to trace the ownership. Various watermark embedding and extraction techniques were considered and used in order to obtain reliable extraction and robustness against attacks. This paper presents a new color image watermarking method based on modification of the reflectance component in the Hue-Saturation-Value (HSV) color space. In the watermark embedding process, the reflectance component extracted from the S component is modified in accordance with Just Noticeable Difference (JND) thresholds derived from the V component. Guided image filtering is used in watermark extraction to predict the original reflectance component, and blind extraction is achieved by subtracting the predicted component from the watermarked one. The performances of five watermarking methods, including the proposed method, were evaluated and compared for accuracy and robustness at the equivalent quality of watermarked images. The results demonstrate that the proposed method provides an improved quality of extracted watermark by both objective and subjective quality measures. It is also more robust than the previous methods against various types of image processing based attacks, including the Stirmark benchmark. © 2021, The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.

Keywords

Blind watermark extraction, Just noticeable difference (JND)