Currently in the health environment, medical
images are a very crucial and important part of the medical
information because of the large amount of information and
their disposal two-dimensional. Medical images are stored,
transmitted and recovered on the network. The images users
await efficient solutions to preserve the quality and protect
the integrity of images exchanged. In this context,
watermarking medical image has been widely recognized as
an appropriate technique to enhance the security,
authenticity and content verification. Watermarking image
may bring elements of complementary research methods of
classical cryptography. The objective of this paper is to
develop a method to authenticate medical images to
grayscale, detect falsified on these image zones and retrieve
the original image using a blind fragile watermarking
technique. We propose a method based on the discrete
wavelet transform (DWT) for the application of content
authentication. In our algorithm, the watermark is
embedded into the sub-bands detail coefficient. The subbands
coefficients are marked by adding a watermark of the
same size as three sub-bands and a comparison of
embedding a watermark at vertical (LH), horizontal (HL)
and diagonal (HH) details. We tested the proposed
algorithm after applying some standard types of attacks and
more interesting. The results have been analyzed in terms of
imperceptibility and fragility. Tests were conducted on the
medical images to grayscale and color size 512 × 512.
Published In:IJCSN Journal Volume 5, Issue 5
Date of Publication : October 2016
Pages : 734-740
Figures :04
Tables : 02
Boujemaa Nassiri : was born in El Jadida, Morocco on January 1,
1974. He received the Master’s degree in electronic systems from Ibn
Zohr University, Agadir, Morocco, in 2009. He received the PhD
degree in data processing in 2015, Currently, he is a teacher
researcher in the international university of Agadir. His research
interests include Real time Acquisition and protected transmission of
medical data.
Abdelaziz El Aissaoui : was born in Agadir in Morocco on June 3,
1967. He holds a Master in Business Intelligence from the University
of Lorraine in France in 2010. Currently he is a preparatory course
coordinator in the international university of Agadir . His research
interests relate to the processing of medical image.
Yousef EL Mourabit : was born in Agadir, Morocco in 1986.
He is graduate as a computer engineer in 2011 from National
school of applied sciences (ENSA) of Agadir in IBN ZOHR
University. He joined the Laboratory of Systems Engineering ans
Information Technology (LISTI) in the ENSA of Agadir in IBN ZOHR
University as a PhD Student in 2012. His current research interests
include information technology, security of sensor networks and
biomedical image.
Rachid Latif : was born in Agadir, Morocco, on December 8, 1968.
He received the PhD degree in signal processing in 2000 and the
Habilitation degree in 2005, from Ibn Zohr University, Morocco.
Currently, he is a Professor with the Department of Industrial
Engineering, Ibn Zohr University, Agadir, Morocco. His research
interests include biomedical signal processing, fuzzy logic, timefrequency
signal processing and he is working on the modeling,
filtering, and analysis of fetal cardiac signals. Prof. Latif is the head of
the Laboratory of Systems Engineering ans Information Technology
(LISTI) in the ENSA of Agadir in IBN ZOHR University and is a
member of the Marocain Acoustical Society (SMA).
Bsiss Mohammed Aziz : Doctor in Medicine,Assistant Professor of
Biophysics and Nuclear Medicine University Cady Ayyad Marrakech
CHU Mohammed VI Marrakech, Faculty of Medicine and Pharmacy
Marrakech, Biophysiqe Laboratory Department of Nuclear Medicine.
Watermarking, DWT, Imperceptibility,
Authenticity, Fragility, Normalized Hamming Distance, Medical
Image
The proposed scheme is very sensitive. It can detect even
one bit of distortion in the image. The image doctor logo
used as watermark can easily detect the authenticity of the
image. One shortcoming of the proposed technique is that
it cannot distinguish between the intentional and unintentional
attacks. During transmission from one
hospital to the other, medical images are generally
compressed in order to save the bandwidth memory. In
this case compression can be considered as un-intentional
attack and the authentication system should deem the
image authentic. However the proposed scheme declares
the image as unauthentic in this scenario as shown in
(table2). The future work will deal with the semi-fragile
watermarking technique for medical images that can
survive against the legitimate attacks like common signal
and image processing operation where as declare the
image unauthentic if an instance of illegitimate attack
occurs.
[1] J.L. Dugelay and S. Roche, 1999 , “Introduction au
tatouage d’image,” in Ann. Télécommunication, vol.
54, pp. 427–437.
[2] P. Bas, 2000, ‘‘Méthodes de tatouages d’images
fondées sur le contenu’’. Thèse de Doctorat, Institut
National Polytechnique de Grenoble.
[3] S. Katzenbeisser and F. A. Petitcolas, 2000,
‘‘Information Hiding, Techniques for Steganography
and Digital Watermarking’’, Artech House.
[4] Nikolaidis and I. Pitas, 1999, “Digital Image
Watermarking : An Overview’’. In ICMCS, volume 1,
pages 1_6.
[5] B. Mathon, 2011, ‘‘Développement de méthodes de
tatouage sûres pour le traçage de contenus
multimédia’’, thèse en cotutelle internationale,
université de Grenoble et catholique de Louvain.
[6] G .Coatrieux, , H. Maitre, B .Sankur, Y .Rolland, and
Collorec, R., 2000 “Relevance of watermarking in
medical imaging’’. Proceedings of International
Conference on Information , Technology Applications
in Biomedicine, IEEE-EMBS, pp. 250–255.
[7] I. J .Cox, M. L .Miller and J. A. Bloom, 2002 ,‘‘
Digital watermarking’’.San Francisco, CA: Morgan
Kauffman.
[8] G. Coatrieux, L. Lecornu, and B .Sankur, 2006, ‘‘A
review of image watermarking applications in
healthcare’’, Proceedings of the 28th IEEE.
[9] H. Yuan, X.-P. Zhang, 2004, "A Multiscale Fragile
Watermark Based on the Guassian Mix-ture Model in
Wavelet Domain", Proceedings of IEEE ICASSP
Confernce, Vol. 3, pp.17-21, May.
[10] N.F. Johnson, Z. Duric, S. Jajodia, 2000, "Information
Hiding: Stegnography and Watermarking Attacks and
Counterattacks", Kluwer Academic Publishers,
Dordrechet.
[11] S.Mallat, 1989, “Multiresolution Approximation and
Wavelet Orthonormal Bases of L2 (R)”, Trans. Amer.
Math. Soc, pp. 69-87.
[12] Z. Dezhong, C. Fayi, 2008, “Face Recognition based
on Wavelet Transform and Image
Comparison”,International Symposium on
Computational Intelligence and Design.
[13] S.Chikhi, 2008 ,‘‘Contribution to the supple
authentication by digital techniques watermarkings
digital image: application to medical image'', PhD
thesis, Univesity of Constantine Algeria .
[14] N. V. Rao & V. Meena Kumari, 2011, “Watermarking
in Medical Imaging for Security and Authentication’’, Information Security Journal: A Global Perspective,
20:3, 148-155.
[15] C.Shien Lu, 2005, “Multimedia security :
steganography and digital watermarking techniques
for protection of intellectual property” ,Institue of
Information Science Academie Sinica,Taiwan.
[16] C.R. Rodriguez, F. Uribe Claudia, T. Blas Gershom,
2007, “Data Hiding Scheme for Medical Images”,
Proceedings of IEEE 17thInternational Conference on
Electronics, Communications and Computers
(CONIELECOMP).
[17] F. Autrusseau, 2002, ‘‘Watermarking based on
modeling the human visual system and the
transformation Mojette’’, Doctoral thesis, University
of Nantes.
[18] H.Y. Leung, L.M. Cheng, and L.L. Cheng, 2009,
“Digital Watermarking Schemes Using Multiresolution
Curvelet and HVS Model” Proceedings of
Springer , 8th International Workshop, IWDW,
Guildford, UK, , August 24-26.
