TY - GEN
T1 - Image quality assessment in reversible data hiding with contrast enhancement
AU - Wu, Hao Tian
AU - Tang, Shaohua
AU - Shi, Yun Qing
N1 - Publisher Copyright:
© Springer International Publishing AG 2017.
PY - 2017
Y1 - 2017
N2 - In this paper, image quality assessment (IQA) in reversible data hiding with contrast enhancement (RDH-CE) is studied. Firstly, the schemes of RDH-CE are reviewed, with which image contrast can be enhanced without any information loss. Secondly, the limitations of using the peak signal-to-noise ratio (PSNR) to indicate image quality in the scenario of RDH-CE are discussed. Subsequently, three no-reference IQA metrics and four metrics specially designed for contrast-changed images are adopted, in addition to PSNR and structural similarity (SSIM) index. By using these metrics, the evaluation results on the contrast-enhanced images generated with two RDH-CE schemes are obtained and compared. The experimental results have shown that the no-reference IQA metrics, the blind/referenceless image spatial quality evaluator (BRISQUE) for instance, are more suitable than PSNR and SSIM index for the images that have been enhanced by the RDH-CE schemes. Furthermore, how to use the suitable IQA metrics has been discussed for performance evaluation of RDH-CE schemes.
AB - In this paper, image quality assessment (IQA) in reversible data hiding with contrast enhancement (RDH-CE) is studied. Firstly, the schemes of RDH-CE are reviewed, with which image contrast can be enhanced without any information loss. Secondly, the limitations of using the peak signal-to-noise ratio (PSNR) to indicate image quality in the scenario of RDH-CE are discussed. Subsequently, three no-reference IQA metrics and four metrics specially designed for contrast-changed images are adopted, in addition to PSNR and structural similarity (SSIM) index. By using these metrics, the evaluation results on the contrast-enhanced images generated with two RDH-CE schemes are obtained and compared. The experimental results have shown that the no-reference IQA metrics, the blind/referenceless image spatial quality evaluator (BRISQUE) for instance, are more suitable than PSNR and SSIM index for the images that have been enhanced by the RDH-CE schemes. Furthermore, how to use the suitable IQA metrics has been discussed for performance evaluation of RDH-CE schemes.
KW - Contrast enhancement
KW - Image quality assessment
KW - Reversible data hiding
KW - Visual quality
UR - http://www.scopus.com/inward/record.url?scp=85028454212&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85028454212&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-64185-0_22
DO - 10.1007/978-3-319-64185-0_22
M3 - Conference contribution
AN - SCOPUS:85028454212
SN - 9783319641843
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 290
EP - 302
BT - Digital Forensics and Watermarking - 16th International Workshop, IWDW 2017, Proceedings
A2 - Shi, Yun-Qing
A2 - Kim, Hyoung Joong
A2 - Kraetzer, Christian
A2 - Dittmann, Jana
PB - Springer Verlag
T2 - 16th International Workshop on Digital Forensics and Watermarking, IWDW 2017
Y2 - 23 August 2017 through 25 August 2017
ER -