ELE 488 Fall 2006

peachpuceAI and Robotics

Nov 6, 2013 (3 years and 9 months ago)

115 views

ELE 488 F06

ELE 488 Fall 2006

Image Processing and Transmission


(12


12


06)





Digital Watermarking


What?
secondary information in media data

Why? to convey additional information


to detect alteration

How? insertion and detection


Attacks, Legal issues


Binary Images


Self Embedding


12/12

ELE 488 F06

Embedding in Binary Images

ELE 488 F06

Document Authentication


Embed prescribed pattern or content features beforehand


Verify hidden data’s integrity to decide on authenticity

(f)

alter

(a)

(b)

(g)

after
alteration

(e)

(c)

(d)

ELE 488 F06

How to Embed Information


Flip pixels from black to white or white to black


Pixel is
flippable

if changing it causes no noticeable artifacts


Construct flippable score of all patterns.

Compute: smoothness


# of transition


connectivity


# of clusters

Compare smoothness and connectivity
before and after flipping

ELE 488 F06

Flippable Score

ELE 488 F06

How to Embed Information


Odd

Even Embedding


Partition image into blocks


To embed “0” in a block, enforce the total number of black pixels
in that block to be even.


Change that pixel with the highest flippable score


To embed “1”, force total number of black pixels to be odd.



Generalization


Choose a step size of Q and enforce the total number of black
pixels in a block to be 2kQ (for some k) to embed a “0”, and to be
(2k+1)Q otherwise.


Enhances robustness


Q = 1


Odd

Even

ELE 488 F06

Not all blocks have flippable pixels

ELE 488 F06


embedding rate


… >= 1 bit / block

Wu
-
Liu Scheme: shuffling
(cont’d)

ELE 488 F06

ELE 488 F06

Summary

Extraction of watermark does not require original

Geometric Attacks

Variations and Generalizations

ELE 488 F06

Fridrich & Goljan ICIP 99

ELE 488 F06

Self


Correcting Images


Divide image into 8 x 8 blocks; DCT; Zig


Zag


Quantize DCT coefficients
using quantization matrix corresponding
to a 50% quality JPEG


Encode each coeff with a fixed # of bits, so that total # of bits is 64
or 128.


Insert the 64
-
bit string of block B into the LSB of the block
B
+ p,
where p is a vector of length approximately 3/10 of the image size
with a randomly chosen direction
.




Q 16 11 10 16 24 40 51 61 L

7 7 7 5 4 3 2 1
12 12 14 19 26 58 60 55
7 6 5 5 4 2 1 0

14 13 16 24 40 57 69 56
6 5 5 4 3 1 0 0

14 17 22 29 51 87 80 62
5 5 4 3 1 0 0 0

18 22 37 56 68 109 103 77
4 4 3 1 0 0 0 0



24 35 55 64 81 104 113 92 3 2 1 0 0 0 0 0


49 64 78 87 103 121 120 101 2 1 0 0 0 0 0 0


72 92 95 98 112 100 103 99 1 0 0 0 0 0 0 0


ELE 488 F06

“Self Embedding”

ELE 488 F06

Reconstruction

Original

Watermarked

ELE 488 F06

Encoding


First 3 coefficients encoded using the same # of bits as
in Table.


Next 18 bits indicate which of coefficients No. 4

21 are
not zero. Followed by values of nonzero coefficients.


Continue same way if under bit budget, 2 coeff at a time


Average code length ~100 bits (1.55 bits / pixel)


Vulnerable to attacks

Generalizations and Variations?


Trade off vs payload

ELE 488 F06

Example

ELE 488 F06

Issues and Challenges


Tradeoff among conflicting
requirements


Imperceptibility


Robustness & security


Capacity



Key elements of data hiding


Perceptual model


Embedding one bit


Multiple bits


Uneven embedding capacity


Robustness and security


What data to embed




Upper
Layers

Uneven capacity equalization

Error correction

Security

……


Lower
Layers

Imperceptible embedding

of one bit

Multiple
-
bit embedding

Coding of embedded data

Robustness

Capacity

Imperceptibility

UMCP ENEE631 Slides (created by M.Wu
©
based on Research Talks ’98
-
’04)

ELE 488 F06

Watermark Attacks: What and Why?


Attacks: intentionally obliterate watermarks


remove a robust watermark


make watermark undetectable
(e.g., miss synchronization)


uncertainty in detection
(e.g., multiple ownership claims)


forge a valid (fragile) watermark


bypass watermark detector



Why study attacks?


identify weaknesses


propose improvement


understand pros and

limitation of tech. solution

UMCP ENEE631 Slides (created by M.Wu
©
based on Research Talks ’98
-
’04)

ELE 488 F06

“Innocent Tools” Exploited by Attackers


Recovery of lost blocks


for resilient multimedia transmission of JPEG/MPEG


good quality by edge
-
directed interpolation:
Jung et al; Zeng
-
Liu


Remove robust watermark by block replacement

edge
estimation

edge
-
directed
interpolation

UMCP ENEE631 Slides (created by M.Wu
©
based on Research Talks ’98
-
’04)

ELE 488 F06


Potential civilian use for digital rights management (DRM)


Copyright industry


$500+ Billion business ~ 5% U.S. GDP


Alleged Movie Pirate Arrested (23 January 2004)


A real case of a successful deployment of 'traitor
-
tracing'
mechanism in the digital realm


Use invisible fingerprints to protect screener copies of pre
-
release movies

Carmine Caridi

Russell

friends

… Internet

w1

Last Samurai

Hollywood studio traced pirated version

http://www.msnbc.msn.com/id/4037016/

Case Study: Tracing Movie Screening Copies

UMCP ENEE631 Slides (created by M.Wu
©
based on Research Talks ’98
-
’04)

ELE 488 F06

Collusion Attacks by Multiple Users

. . .

Averaging Attack

Interleaving Attack


Collusion: A cost
-
effective attack against MM fingerprints


Users with same content but different fingerprints come together
to produce a new copy with diminished or attenuated
fingerprints









Result of fair collusion:


Each colluder contributes equal share through averaging,
interleaving, and nonlinear combining


Energy of embedded fingerprints may decrease


=> Need for Collusion
-
resistant Fingerprinting

UMCP ENEE631 Slides (created by M.Wu
©
based on Research Talks ’98
-
’04)

ELE 488 F06

References


F. Mintzer, G.W. Braudaway, M.M. Yeung, “Effective and Ineffective Digital
Watermarks”, IEEE ICIP 97


Cox, J. Kilian, T. Leighton, T. Shamoon: “Secure Spread Spectrum
Watermarking for Multimedia'', IEEE Trans Image Processing, Dec 1997


M Wu, B Liu, “Watermarking for image authentication”, ICIP 98.


M. Wu, B. Liu, “Data Hiding in Binary Images for Authentication and
Annotation", IEEE Trans Image Processing, August 2004.


M. Wu, W. Trappe, Z.J. Wang, and K.J.R. Liu: “Collusion
-
resistant
fingerprinting for Multimedia,” IEEE Signal Proc Magazine, March 2004.


J. Fridrich and M. Goljan, “Images with Self
-
Correcting Capabilities”, ICIP
1999.

ELE 488 F06

ELE 488 Fall 2006

Image Processing and Transmission

Bede Liu
, B330, x4628, liu@princeton.edu


Bryan Conroy
, F
-
2G2, bconroy@princeton.edu



Office hours, help sessions, reviews, etc will be
announced on blackboard


image = picture; processing = working on

Working on Pictures

ELE 488 F06

Why Image Processing?


Enhancement



improve appearance, reveal more details


Analysis and extraction of content


face, motion, patterns,
object tracking


Storage and distribution



encoding, compression, transmission


Digital Library



indexing, searching


Medical imaging



diagnosis, exploration



Remote sensing



weather map, terrain mapping, monitoring of
environment


Radar and microwave imaging


mapping of sky, moon


Security


biometrics, detection of events


Watermarking


data hiding


Composition



Magazines, Movies


Display and printing



ELE 488 F06

Some Logistics


Labs, homeworks, project suggestions will be posted on
blackboard.


Assignments, Labs and Project are based on Paintshop
Pro and MATLAB. You can use any computer loaded
with MATLAB and the Signal and Image Processing
Toolboxes.


A Lab Room (F
-
113) for ELE 488 has been reserved on
Wednesday and Thursday evening, except next week.


Bryan will be in the Lab to answer questions on those
evenings.


ELE 488 F06

Grading

(tentative)


Midterm I


(Thursday, midterm week)


20%


Assignments and Labs




20%


Midterm II (last week of classes in Dec.)

20%


Final project: 4 parts, all graded:


40%


Project Proposal

(due week after fall break)


Progress Report

(due last day of classes Dec.)


Oral Presentation

using power point (due reading period)


**
Note that we will meet 2 or 3 times during reading period


Final Report

(due Dean’s date).








ELE 488 F06

Books and References


Primary


A.K. Jain:
Fundamentals of Digital Image Processing
, Prentice
-
Hall, 1989.


R.C. Gonzalez and R.E. Woods:
Digital Image Processing
,
Prentice Hall, 2001.


R.C. Gonzalez, R.E. Woods and S.L. Eddins:
Digital Image
Processing using Matlab
, Prentice Hall, 2004.


Y. Wang, J. Ostermann, Y
-
Q. Zhang:
Digital Video Processing
and Communications
, Prentice
-
Hall, 2001.


Introductory sections in Matlab Image Processing Toolbox
http://www.mathworks.com/access/helpdesk/help/toolbox/images/image
s.shtml



Other references


To be announced

ELE 488 F06

ELE 488 Fall 2006

Image Processing and Transmission





Syllabus


1. Human Visual System

2. Image Representations (gray level, color)

3. Simple Processing: point operations and filtering

4. Still Image Coding

5. Resampling, Resizing, Interpolation, and Registration

6. Probability Models, Quantization, Estimating Densities

7. Synthesizing Pixels, Segmentation

8. Radon Transform, Other imaging modes

9. Video, Video Compression

10. Selected Topics: watermarking, feature description,


face recognition, . . .


9/19/06

ELE 488 F06

ELE 488 Fall 2006

Image Processing and Transmission





09
-
21
-
06


1.
Generate and Display of Gray Scale images
in Matlab


2.
Histogram of Gray Scale Image


3.
Point Operations: brightness, contrast,
gamma


correction


9/21/06

ELE 488 F06

ELE 488 Fall 2006

Image Processing and Transmission





09
-
26
-
06



Linear 2
-
D Image Filtering



1
-
D discrete convolution


2
-
D discrete convolution


2
-
D spatial masks


Mask filtering


Mask filtering and 2
-
D convolution


Spatial Averaging, Blurring, Image Sharpening



Edge Map
-

gradient


9/26/06

ELE 488 F06

ELE 488 Fall 2006

Image Processing and Transmission





09
-
28
-
06



Edge Map



Laplacian



Median Filter



Filtering Images in Frequency Domain



Image Restoration


9/28/06

ELE 488 F06

ELE 488 Fall 2006

Image Processing and Transmission





10
-
3
-
06


Image Restoration



distortion


noise


Inverse Filtering


Wiener Filtering


Ref:
Jain, Sec 8.1


8.3
.


Gonzalez

Woods, Sec 5.5


5.8



10/3/06

ELE 488 F06

ELE 488 Fall 2006

Image Processing and Transmission


(10
-
5
-
06)






Wiener Filtering



Derivation


Comments


Re
-
sampling and Re
-
sizing



1D


2D


10/5/06

ELE 488 F06

ELE 488 Fall 2006

Image Processing and Transmission


(10
-
12
-
06)





Re
-
sampling and Re
-
sizing



1D


2D, sinc interpolation


nearest neighbor, bilinear, bicubic, . . .


Geometric Transformation



translation


rotation


scaling


Affine




10/12/06

ELE 488 F06

ELE 488 Fall 2006

Image Processing and Transmission


(10
-
17
-
06)





Geometric Transformation



translation


rotation


scaling


Affine


Image Registration



Goodness of fit




10/17/06

ELE 488 F06

ELE 488 Fall 2006

Image Processing and Transmission


(10
-
19
-
06)





Image Compression


Review of Basics


Huffman coding


run length coding


Quantization


independent samples


uniform and optimum


correlated samples


vector quantization









10/19/06

ELE 488 F06

ELE 488 Fall 2006

Image Processing and Transmission


(10
-
24
-
06)




Image Compression


Quantization


independent samples


uniform and optimum


correlated samples


vector quantization


JPEG


block based


transform coding


. . . .

10/24/06

ELE 488 F06

ELE 488 Fall 2006

Image Processing and Transmission


(10
-
26
-
06)




Image Compression


JPEG


block based


transform coding


. . . .

10/26/06

ELE 488 F06

ELE 488 Fall 2006

Image Processing and Transmission


(11
-
10
-
06)




JPEG


block based


transform coding


. . . .


Why DCT for Image transform?


DFT


DCT


Wavelet


11/10

ELE 488 F06

ELE 488 Fall 2006

Image Processing and Transmission


(11
-
16
-
06)




JPEG


block based


transform coding


Is DCT best?


decorrelation, energy compaction


Karhunen
-
Loeve (K
-
L) transform


Spatial correlation


Subband decomposition & coding


Wavelet transform


Zero tree


11/16

ELE 488 F06

ELE 488 Fall 2006

Image Processing and Transmission


(11
-
20
-
06)




Lossy wavelet encoding


Subband decomposition & coding



Wavelet transform



Embedded zero tree


Successive approximation quantization


Digital Video


11/20

ELE 488 F06

ELE 488 Fall 2006

Image Processing and Transmission


(11
-
28
-
06)





Digital Video



Motion Pictures



Broadcast Television



Digital Video



11/28

ELE 488 F06

ELE 488 Fall 2006

Image Processing and Transmission


(12


5


06)





Reconstruction from Projection



Radon Transform



Reconstruction



Computation



12/5

ELE 488 F06

ELE 488 Fall 2006

Image Processing and Transmission


(12


7


06)





Digital Watermarking



What?


Why?


How?


Attacks, Legal issues


12/7