Top Banner
DIGITAL WATERMARKING Ngô Huy Phúc 50701831 Trần Kim Lân 50701259 Phạm Quốc Hiệp 50700812
51

DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

Dec 29, 2015

Download

Documents

Edgar Dennis
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

DIGITAL WATERMARKING

Ngô Huy Phúc 50701831Trần Kim Lân 50701259Phạm Quốc Hiệp 50700812

Page 2: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

PART 1

INTRODUCTION

Page 3: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

STEGANOGRAPHY

• Steganography (art of hidden writing)

– The art and science of writing hidden messages in such a way that no one apart from the intended recipient knows of the existence of the message.

– The existence of information is secret.

Page 4: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

• Histaeus used his slaves (information tattooed on a slave’s shaved head)

Initial Applications of information hiding

Passing Secret messages

STEGANOGRAPHY

Page 5: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

STEGANOGRAPHY

• Physical steganography

Page 6: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

STEGANOGRAPHY

• Digital steganography

• Network steganography

Page 7: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

DEFINITION

• The process of embedding information into a digital signal in a way that is difficult to remove.

• The signal may be text, images, audio, video.

• The information is also carried in the copy if the signal is copied.

Page 8: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

DEFINITIONExample:

Page 9: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

GENERAL APPLICATIONS

Copyright Protecton • To prove the ownership of digital media.

Page 10: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

Tamper proofing• To find out if data was tampered.

GENERAL APPLICATIONS

Page 11: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

Quality Assessment• Degradation of Visual Quality

Loss of Visual Quality

GENERAL APPLICATIONS

Page 12: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

LIFE-CYCLE PHASES

Attemp to extract

watermark from signal

Attemp to extract

watermark from signal

The marked signal is

modified

The marked signal is

modified

Produce watermarke

d signal

Produce watermarke

d signal

Page 13: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

CLASSIFICATION

Digital watermarking techniques can be classified in many ways :

•Visibility•Robustness•Perceptibility•Capacity•Embedding method

Page 14: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

VISIBILITY

• Visible– Text or a logo which identifies the owner of the

media.

• Invisible– Information is added as digital data to audio,

picture or video, but it cannot be perceived.– May be a form of Steganography.

Page 15: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

ROBUSTNESS

• Robust – Resisted a designated a class of transformations.– Against adversary based attack.

(e.g. noise addition to images)– Used in copy protection application.

Example: Robust Private Spatial Watermarks

Page 16: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

ROBUSTNESS

• Fragile– Fail to be detected after the slightest

modification.– Used for tamper detection.

Example: Blind Fragile DCT based Watermarks

Page 17: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

ROBUSTNESS

• Semi-fragile– Resist benign tramsformations but fails detection

after malignant transformations.– Robust against user-level operation.

(e.g. image compression)– Used for detect malignant transformation.

Example: Blind Semi-fragile Spatial Watermarks

Page 18: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

PERCEPTIBILITY

• Perceptible– Its presence in the marked signal is noticable, but

non-intrusive.

• Imperceptible– Original cover signal and the marked signal are

close to perceptually indistinguishable.

Page 19: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

PERCEPTIBILITY

Stanford Bunny 3D Model Visible Watermarks in Bunny Model Distortion

Watermarking

Stanford Bunny 3D Model

Watermarking

Invisible Watermarks in Bunny Model Minimal Distortion

Page 20: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

CAPACITY

• Depend on the length of the embedded message.

• Zero-bit long– Detect the presence or absence of the watermark.– A 1 denotes the presence. 0 denotes the absence.

• N-bit long– Modulated in the watermark.– Support multiple watermarks.

Page 21: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

EMBEDDING METHOD

• Spread-spectrum– The marked signal is ontained by an additive

modification.– Modestly robust.– Have a low information capacity.

Page 22: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

EMBEDDING METHOD

• Quantization type– The marked signal is ontained by quantization– Low robustness.– Have a high infoirmation capacity.

• Amplitude modulation– The marked signal is ontained by additive

modification similar to spread spectrum method.– Embedded in the spatial domain.

Page 23: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

• As much information (watermarks) as possible.

Capacity• Only be accessible by authorized parties.

Security• Resistance against hostile/user dependent

changes Robustness• Invisibility Imperceptibility

DESIGN REQUIREMENTS

Page 24: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

PART 2

SPECIFIC WATERMARKING TECHNIQUES ON IMAGES

Page 25: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

• A very simple yet widely used technique for watermarking images is to add a pattern on top of an existing image.

• Usually this pattern is an image itself - a logo or something similar.

SIMPLE WATERMARKING

Page 26: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

The LSB technique is the simplest technique of watermark insertion.

• Consider a still image : each pixel of the color image has three components — red, green and blue.

• Allocate 3 bytes for each pixel. Then, each colour has 1 byte, or 8 bits.

LSB : LEAST SIGNIFICANT BIT

Page 27: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

A pixel that is bright purple in colour can be showN as X0 = {R=255, G=0, B=255}

• Look at another pixel: X1 = {R=255, G=0, B=254}• Detecting a difference of 1 on a color scale of 256

is almost impossible for human eye.

Replace the color intensity information in the LSB with watermarking information, the image will still look the same to the naked eye.

LSB : LEAST SIGNIFICANT BIT

Page 28: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

• Use a secret key to choose a random set of bits.• The more bits used in the host image, the more it

deteriorates.• Increasing the number of bits used though

obviously has a beneficial reaction on the secret image increasing its clarity.

LSB : LEAST SIGNIFICANT BIT

Page 29: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

Host image is on the left, watermark image is on the right

LSB : LEAST SIGNIFICANT BIT

Page 30: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

• Watermarking in the frequency domain involves selecting the pixels to be modified based on the frequency of occurrence of that particular pixel.

• Transform an image into the frequency domain. • A block-based DCT watermarking approach is

implemented. • An image is first divided into blocks and DCT is

performed on each block. The watermark is then embedded by selectively modifying the middle-frequency DCT coefficients.

FREQUENCY-BASED TECHNIQUES

Page 31: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

What is DCT ?• Formally, the discrete cosine transform (DCT)

is a linear, invertible function F : RN -> RN (where R denotes the set of real numbers), or equivalently an invertible N × N square matrix

FREQUENCY-BASED TECHNIQUES

Page 32: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

FREQUENCY-BASED TECHNIQUES

Page 33: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

Discrete wavelet transform (DWT)• The image is separated into different resolution• The original image is high-pass filtered, yielding

the three large images, each describing local changes details in the original image

• It is then low-pass filtered and downscaled, yielding an approximation image.

• This image is high-pass filtered to produce the three smaller detail images.

• And low-pass filtered to produce the final approximation image in the upper-left.

WAVELET WATERMARKING TECHNIQUES

Page 34: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

WAVELET WATERMARKING TECHNIQUES

Page 35: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

Embedding the watermark• The host image and watermark are transformed into

wavelet domain.• The transformed watermark coefficients were

embedded into those of host image at each resolution level with a secret key.

WAVELET WATERMARKING TECHNIQUES

Page 36: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

WAVELET WATERMARKING TECHNIQUES

Page 37: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

• A Narrow-band signal is transmitted over a much larger bandwidth such that the signal energy presented in any signal frequency is undetectable

• A watermark is spread over many frequency bins so that the energy in one bin is very small and certainly undetectable.

SPREAD-SPECTRUM TECHNIQUES

Page 38: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

SPREAD-SPECTRUM TECHNIQUES

Page 39: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

• Because the watermark verification process knows the location and content of the watermark, it is possible to concentrate these weak signals into a single output with high SNR (Signal-to-noise ratio).

• Remark– To destroy such a watermark would require noise of

high amplitude to be added to all frequency bins.– The location of the watermark is not obvious.– Frequency regions should be selected that ensures

degradation of the original datafollowing any attack on the watermark.

SPREAD-SPECTRUM TECHNIQUES

Page 40: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

References• Techniques and Applications of Digital

Watermarking and Content Protection Michael Arnold, Martin Schmucker, Stephen D. Wolthusen

• Steganography And Digital WatermarkingJonathan Cummins, Patrick Diskin, Samuel Lau and Robert Parlett, School of Computer Science, The University of Birmingham.

• Real-Time Digital Image WatermarkingSubramaniam Ganesan, Professor of Oakland University, Michigan

Page 41: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

PART 3

ATTACKING METHODS

Page 42: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

Foundations of Attacking

• 3 effects make detection of watermarking useless:

– Watermark cannot be detected.– False watermarks are detected.– Unauthorized detection of watermark.

Page 43: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

Classification of Attacking

• Removal attacks• Geometrical attacks• Cryptographic attacks• Protocol attacks

Page 44: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

Classification of watermarking attacks

Page 45: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

Removal Attacks

• Most obvious method• Aim for complete removal of watermarking• Extreme form of this type is restore the

original object• Can happen unintentionally due to operations

in some certain applications.

Page 46: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

Geometrical Attacks

• Do not actually remove the embedded watermark

• Intend to distort the watermark detector synchronization with the embedded information

Page 47: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

Cryptographic Attacks

• Aim at cracking the security methods in watermarking schemes

• Finding a way to remove the embedded watermark information

• Embed misleading watermarks• High computational complexity

Page 48: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

Protocol Attacks

• Aim at attacking the entire concept of the watermarking application

• First proposed in framework of invertible watermark

• The attacker subtracts his own watermark from the watermarked data and claims to be the owner

• Another type is copy attack

Page 49: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

Some Methods

• Collusion Attack– Estimate the watermark from different works with

same watermark– The attackers can obtain an approximation of the

watermark by averaging the watermarked works

Page 50: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

Some Methods

• Remodulation AttackDamage watermark base on watermark estimation

Page 51: DIGITAL WATERMARKING Ngô Huy Phúc50701831 Trần Kim Lân50701259 Phạm Quốc Hiệp50700812.

Some Methods

• Copy AttackEstimate a watermark from watermarked data and copy it to some other data