Copyright Protection Using Spread Weighted Sum Watermarking

8/4/2019 Copyright Protection Using Spread Weighted Sum Watermarking

1/12

COPYRIGHT PROTECTION USING SPREAD WEIGHTED

SUM WATERMARKING

ABSTRACT:

The copyright in the work of authorship immediately becomes the property of the author who created the work. Only the author or those deriving their rights through theauthor can rightfully claim copyright. In the case of works made for hire, the employer andnot the employee is considered to be the author. In this paper we introduce a techniquecalled SPREAD WEIGHTED SUM (SWS) watermarking which is used to protect the copyright of

the author. We also suggest a technique to protect the copyright of videos by means of a newtechnique called MOTION ENTROPY using H.264 STANDARD. We also compare theperformance of LEAST SIGNIFICANT BIT (LSB) with that of SPREAD WEIGHTED SUM (SWS)WATERMARKING for IMAGES. We also perform certain performance analysis such as SIGNALTO NOISE RATIO (SNR).

NECESSITY OF OUR PROJECT:

In this time and age, images and videos have a direct on thedaily life of average individual. Due to the ease of sharing digital images and videos,especially after the internet revolution, its important to protect ones authentication. Eventhough many techniques are available to protect the authentication and provide copyrightprotection to the owner of the image or video, nothing is fully able to provide full protectionto the owner of the image. Thus we develop a new technology to provide authentication andprovide copyright protection to the owner of the image. WATERMARKING plays an importantrole in providing authentication to the owners o f the image.

WATERMARKING:

A watermark is a digital code permanently embedded in cover content,in case of thesis, into a video sequence. A watermark can carry information you can imaginebut the amount of the information is unlimited. The more information the watermark carriesthe more vulnerable the information is. There are different types of watermarking available,they are


2/12

Visible watermarking; Invisible watermarking ;

VISIBLE WATERMARKING:

The data that is embedded into the image that is visible tothe user, that type of watermarking is usually, is referred to as VISIBLE WATERMARKING.


3/12

INVISIBLE WATERMARKING:

The data embedded into the image that is invisible to the

user is referred to as INVISIBLE WATERMARKING .

INVISIBLE WATERMARKING:There are two types available in invisible

watermarking. They are

1. LEAST SIGNIFICANT BIT WATERMARKING;

2. SPREAD WEIGHTED SUM WATERMARKING;

LEAST SIGNIFICANT BIT WATERMARKING:

The least significant bit (LSB) method embeds a digitalmessage inside the LSB plane of a truncated grayscale image. We should read the truncatedgray scale image from the file and we should reset the LSB plane of this image to zero. Thenthe message to be written into the image should be converted into a string of bits. Finally, weshould ad the string of bits to the truncated gray scale image and we should write the grayscale image as a .bmp file.


4/12

SPREAD WEIGHTED SUM WATERMARKING:

THE SPREAD WEIGHTED SUM (SWS) method embeds a digitalmessage inside a grayscale image by manipulating the value of its pixels. The value of a pixelp1 in the watermarked image is obtained based on the original value p of the pixel and thevalue b of the bit to be embedded in it as follows,

P1= p+, b=0

P-, b=1

After implementing this method, the gray scale imageis readed from the file. The message to be embedded is converted into a bit stream. Since

message to be embedded is very small compared to the size of image, we can affordrepeating each bit r times before embedding it in the image. The maximum value of r is theratio between the image size and the length of the string of bits.

IMAGE EXTRACTION:

Extracting the embedded message needs to read the gray scale imageand watermarked image from file. Subtract the watermarked image and the gray scale image

to isolate the matrix where our message is embedded. Compare the sum of r repeated bitsagainst zero and decide whether the message bit is a zero or one. Group each seven bitstogether and find their ASCII code.


5/12

SWS WATERMARKING EXAMPLE

ORIGINALINFORMATION

HELLO EMBED

72 101 108 108 111

245 66 62

140 214 237

35 65 89

INTEGER FORMAT

1001000 1100101 1101100

1101100 1101111BINARY FORMAT NOT NEEDED


6/12

SWS WATERMARKING EXAMPLE:

9 bits can fit in this segment

a = 5

1001000 1100101 1101100 245 66 62 11011001101111 140 214 237

35 65 89

1 0 0 1 0 0 0 1 1 245+5 66-5 62-5

140+5 214-5 237-5

35-5 65+5 89+5


7/12

PHASES INVOLVED IN OUR PROJECT:

PHASE1 APPLYING LSB TECHNOLOGY FOR IMAGES AND OBTAIN THE

RESULTS APPLYING SWS TECHNOLOGY FOR IMAGES AND OBTAIN THE

RESULTS.

PHASE2 PERFORM CODING FOR WATERMARKING USING MOTION

ENTROPY FOR VIDEOS. EVALUATION OF RESULTS OF VIDEO WATERMARKING WITH

CURRENT STANDARDS

PHASE3 COMPARE LSB AND SWS TECHNOLOGIES AND SPOT OUT THE

ADVANTAGES EVALUATION OF PERFORMANCE FEATURES LIKE SIGNAL TO

NOISE RATIO.


8/12

SOFTWARE USED:

IMAGE PROCESSING USING MATLAB.

TOOLBOXES USED:

Image acquisition toolbox5; Image processing toolbox42;

DESIGN PARAMETERS:LOW REDUNDANCY AND LOW:

LOW IMMUNITY TO NOISE AND COMPRESSION. GOODTRANSPARENCY.

ADVANTAGES:

Watermark can not be tempered with (only someone with the originalimage will be able to modify the watermark)

Adjustable level of visible damage to the watermarked image. Easy to embed data and then recover It.

APPLICATIONS:

Owners signatures (Reducing Multimedia piracy) Alert Line on TV broadcasts (Emergency public address) Time stamping of digital images Geo tagging of digital images Saving patients data on biomedical images .


9/12

COPYRIGHT PROTECTION FOR VIDEOS:

Video watermarking involves embedding cryptographicinformation derived from frames of digital video into the video itself. The majority of compression algorithms used in video formats are based on a frequency transform, thuswatermarking in frequency domain can be directly to the coefficients of that transform.

Compressed image is partially decoded to obtain thosetransform coefficients, watermarked and encoded back again.

We are going to present a technique called MOTIONENTROPY for the video streams of H.264 STANDARD.

MOTION ENTROPY:

Technique for embedding watermark information and to detect thepresence of watermark. Its the latest technique for watermarking videos streams.

H.264 STANDARD:

The H.264 standard represents an evolution of the exixting video codingstandards. It has been developed by the ITU-T video coding experts group and ISO/IEC

moving pictures experts group in response to the growing need for higher compression of moving pictures.

This standard has been chosen because it is the latest video compressionstandard and offers significant efficiency improvement over the precious standards.

It also provides better bit-rate to distortion rate.

PARAMETERS TO BE CONSIDERED:

NAL------NETWORK ABSTRACTION LAYER.

SPS -------SEQUENCE PARAMETER SYSYTEM------INFO ABT COLOR AND RESOLUTION.

PPS--------PICTURE PARAMETER SYSTEM--------PICTURE CODING, ENTROPY CODING.


10/12

IMPLEMENTATION OF SOFTWARE FRAMEWORK FOR WATERMARKING VIDEOS:

H.264 CODEC SUPPORTS:

SPS AND PPS;

PARTITOONING OF PICTURE INTO SLICES; ALL SLICE TYPES:I,P,B;

THE WATERMARKING FRAMEWORK:

This framework helps us in implementing any watermarking schemefor videos either in frequency and spatial domain.

GSTREAMER MULTIMEDIA FRAMEWORK:

Generally, a H.264 video stream is usually enveloped together withthe audio stream into a multimedia container format such as AUDIO VIDEO INTERLEAVE.

In order to avoid implementing of unpacking various containerformats and separating out the video stream, the watermarking framework is implementedas a plugin in the open source multimedia framework called GSTREAMER.

GSTREAMER - A library that allows the construction of graphs media handling components,ranging from simple audio playback to complex audio and video processing.

Generic audio-video player pipeline

Filesource

Input

file

Video

decoder

Image

sink

Audiodecoder

Audiosink

Demuxer


11/12

A Gstreamer part implements the gstreamer interface which consists of source program in c.

the incoming data blocks are sliced down and decoded in the gstreamer part itself .

PROCESS OF VIDEO WATERMARKING

EMBEDDING MODE:

INPUT: H.264 STREAM TO PLUGIN.

OUTPUT: WATERMARKED H.264 STREAM.

INPUT TO THE EMBEDDING PROCESS:

AN INTER OR INTRACODED SLICE;

CONTENT ID-IDENTIFIER OF THE COVER CONTENT;COPYID-THE PARTICULAR IDENTIFIER OF PARTICULAR COVER

CONTENT COPY;

WEIGHT-WEIGHT FACTOR SPECIFYING THE WATERMARKSTRENGTH.

EMBEDDING OF

DATA

GTREAMER

INTERFACE

DETECTION OF

DATA

DEMUX

ER

INPUT

FILEFILE

SOURCE

FILE

SINK

OUTPU

T FILE

MUXER

WATERMARK


12/12

CURRENT STATUS OF OUR PROJECT:

o Implemented LSB watermarking in images and compared the result with thatof proposed.

o Implemented SWS watermarking in images and analyzed its performance withthat of proposed work.

Copyright Protection Using Spread Weighted Sum Watermarking

Documents