Wavelet Based Data Hiding of DEM in the Context of Realtime 3D Visualization - PowerPoint PPT Presentation

1 / 23
About This Presentation
Title:

Wavelet Based Data Hiding of DEM in the Context of Realtime 3D Visualization

Description:

In Electronic Imaging, Wavelet Applications in Signal and Image Processing, SPIE, ... Electronic Imaging, Security and Watermarking of Multimedia Contents, SPIE, IS&T, ... – PowerPoint PPT presentation

Number of Views:75
Avg rating:3.0/5.0
Slides: 24
Provided by: lir7
Category:

less

Transcript and Presenter's Notes

Title: Wavelet Based Data Hiding of DEM in the Context of Realtime 3D Visualization


1
Wavelet Based Data Hiding of DEM in the Context
of Real-time 3D Visualization
  • Presented By
  • Khizar Hayat

2
The Team
  • Student
  • Khizar Hayat
  • Supervisor
  • William Puech (LIRMM/ROB/ICAR)
  • Co-supervisors
  • Marc Chaumont (LIRMM/ROB/ICAR)
  • Gilles Gesquiere (LSIS - University of Marseille)
  • Concerned PhD student
  • Philippe Amat

3
The Problem
  • How to achieve optimal real-time 3D visualization
    in a client server environment in a scalable and
    synchronized way compatible with the clients
    computing resources and requirements?

100 Data
A fraction of Data
4
The Proposed Solution
  • Our Approach
  • Discrete Wavelet Transform (DWT) of JPEG2000
  • Data Hiding LSB based embedding
  • Advantages
  • Compression to reduce the amount of data
  • Scalability - different levels of detail for
    various
  • Platforms
  • Users
  • contexts
  • Synchronization reduce the number of files

5
Essential Concepts
  • JPEG2000
  • Data Hiding/ Steganography
  • 3D Visualization

6
JPEG2000


7
DWT in JPEG2000
Let Sj 1D input signal (pixel row or pixel
column) Lk lowpass subband signal Hk
highpass subband signal Where 1 j n and 1
k n/2 Then
  • Lossless (Daubechies 5/3)

Lossy (Daubechies 9/7)
Where a - 1.586134 b - 0.052980 c
0.882911 d 0.443506 With ß and ß, the
scaling parameters ß 0.812893 ß 1/ ß
S2n1 the first stage lifting outcomes S2n
the second stage lifting outcomes
8
LSB Based Data Hiding
  • Embed the message bits directly into the
    least-significant bit plane of the cover image in
    a deterministic sequence, e.g. PRNG
  • Advantages
  • Easy to understand and comprehend ?popular
  • High perceptual transparency
  • Low degradation in the image quality
  • High Hiding Capacity
  • Disadvantages
  • Low robustness to malicious attacks
  • Vulnerable to accidental or environmental noise

9
3D Visualization
  • Three files are necessary
  • Texture
  • Altitude DEM ? Fig. a
  • (50m by IGN France)
  • Geo-referential coordinates (longitude /
    latitude)
  • Two steps for visualization
  • Triangulation (Fig. b)
  • Aerial Photograph draped onto the triangles for
    3D visualization (Fig. c)

(a)
(b)
(c)
10
The Proposed Method
11
The Proposed Method
  • RGB to YCrCb
  • Texture.ppm
  • Wavelet Transformation
  • Lossy DWT of the YCrCb of Texture
  • Lossless DWT of altitude (DEM)
  • Insertion
  • Trans DEM Trans Y plane of texture
  • Correspondence
  • Same level
  • Factor of insertion (E)
  • E m²/ N² (coefficients/ pixel)
  • Block size 1/E ? 2 bytes per 32x32 block in our
    case
  • LSB based Insertion with running a PRNG for pixel
    allocation
  • Final Interleaved Image

12
Results
Altitude Image (64x64 coefficients) 1 coefficient
16 bits
Texture Image (2048x2048 pixels) 1 Pixel 24 bits
Texture Image (A part magnified)
13
Level 1 Transformation
Texture (2048x2048 pixels)
Altitude (64x64 coefficients)
14
Level 3 Transformation
Texture (2048x2048 pixels)
Altitude (64x64 coefficients)
15
Observations

Where
16
Reconstruction from the Image of Approximation at
Level 3
Altitude Image (64x64 coefficients) 1 coefficient
16 bits
Texture Image (2048x2048 pixels) 1 Pixel 24 bits
Texture Image (A part magnified)
ATTENTION Only 1.6 of the initial data utilized
17
Results
Altitude Image (64x64 coefficients) 1 coefficient
16 bits
Texture Image (2048x2048 pixels) 1 Pixel 24 bits
Texture Image (A part magnified)
18
3D visualization of the Altitude from the Image
of Approximaton
Level 1
Level 0 - all the information
Level 2
Level 3
19
3D navigation of the Constructed Images
All the data
Level 3 lowest suband data
ATTENTION Only 1.6 of the initial data utilized
20
Conclusion
  • Encouraging Results in the following contexts
  • Compression
  • Scalability and portability
  • Synchronization
  • Future Perspective
  • Other aspects of JPEG2000
  • Standard Software, e.g. OpenJPEG
  • Geometric Wavelets
  • Non uniform grid to decrease the number of
    triangles
  • Exploration of Chrominance planes

21
Research Papers
  • Papers expected to be published
  • K. Hayat, M. Chaumont, G. Gesquiere and W. Puech.
    Visualisation 3D Temps-Réel à Distance de MNT par
    Insertion de Données Cachées Basée Ondelettes. In
    CORESA06, Caen, France, Nov. 2006. (Submitted in
    May 2006).
  • K. Hayat, W. Puech, G. Gesquiere and M. Chaumont.
    Wavelet Based Data Hiding of DEM In the Context
    of Real-time 3D Visualization. In IST and SPIE
    07 - Visualization and Data Analysis 2007
    (EI108), San Jose, California, USA (to be
    submitted before 17 July 2006).

22
References
  • A. Martin, G. Gesquiere, W. Puech and S. Thon.
    Real Time 3D Visualisation of DEM Combined with a
    Robust DCT Based Data-Hiding Method. In
    Electronic Imaging, Visualization and Data
    Analysis, SPIE, IST, San Jose, CA, USA, volume
    6060, pages 60600G160600G8, Jan 2006.
  • W. Sweldens. The Lifting Scheme a New Philosophy
    in Biorthogonal Wavelet Constructions. In
    Electronic Imaging, Wavelet Applications in
    Signal and Image Processing, SPIE, IST, San
    Diego, CA, USA, volume 2569, pages 6879, Sep.
    1995.
  • I. Daubechies and W. Sweldens. Factoring Wavelet
    Transforms into Lifting Steps. Fourier Anal.
    Appl., 4(3), 1998.
  • ISO/IEC 14492-1 Lossy/lossless coding of
    bi-level images, 2000.
  • R. Raffin S. Thon and G. Gesquiere. Visualisation
    3D de Feux de Forêts sur des Modèles Numériques
    de Terrain de lIGN. In Rencontre GEO-RISQUE 2006
    La cartographie des risques naturels,
    Montpellier (France), Feb. 2006.
  • P. Meerwald and A. Uhl. A Survey of
    Wavelet-Domain Watermarking Algorithms (2001).
    Electronic Imaging, Security and Watermarking of
    Multimedia Contents, SPIE, IST, San Jose, CA,
    USA, Jan 2001.
  • A. Piva, F. Bartolini and R. Caldelli. Self
    Recovery Authentication of Images in the DWT
    Domain. In International Journal of Image and
    Graphics, pages 149-165, Vol. 5, No. 1(2005).
  • J. Li. Image Compression - The Mechanics of the
    JPEG2000. Microsoft Research, Signal Processing
    2002. (Available from http//research.microsoft.c
    om/jinl/paper_2002/msri_jpeg.htm).
  • M.D. Adams. The JPEG-2000 Still Image Compression
    Standard (Last Revised 2005-12-03). ISO/IEC JTC
    1/SC 29/WG 1 (N 2412) (Available from
    www.ece.uvic.ca/mdadams)

23
Thank You
  • Questions ?
Write a Comment
User Comments (0)
About PowerShow.com