A SEMINAR REPORT
Submitted by
MADHURENDRA KUMAR
of
BACHELOR OF TECHNOLOGY
in
SCHOOL OF ENGINEERING
KOCHI-682022
AUGUST 2008
DIVISION OF COMPUTER SCIENCE & ENGINEERING
SCHOOL OF ENGINEERING
KOCHI-682022
MADHURENDRA KUMAR
Date:
ACKNOWLEDGEMENT
MADHURENDRA KUMAR
ABSTRACT
Automatic video surveillance (AVS) systems are used for continuous and effective
monitoring of dangerous and remote sites. Video data acquired by the automatic
court law. But digital video data lacks legal validity due to the ease to manipulate
tampering can be detected and we can indicate the true origin of the data. There a
2. Watermarking-based authentication.
In this presentation an attempt is made to present the basic features of the image
authentication techniques.
TABLE OF CONTENTS
1 ABSTRACT iii
2 LIST OF FIGURES v
3 LIST OF SYMBOLS vi
4 INTRODUCTION 1
6. AUTHENTICATION TECHNIQUES 3
7. CRYPTOGRAPHY 4
AUTHENTICATION
9. WATERMARKING 10
AUTHENTICATION SCHEMS
VS DATA AUTHENTICATION
12. ADVANTATGES 20
13. DISADVANTAGES 21
14. CONCLUSION 22
15. REFERENCES 23
LIST OF FIGURES
SI NO IMAGES PAGE NO
1. INTRODUCTION
This paper explores the various techniques used to authenticate the visual data rec
by the automatic video surveillance system. Automatic video
surveillance systems are
used for continuous and effective monitoring and reliable control of remote and
dangerous sites. Som e practical issues m ust be taken in to account, in order to take
advantage of the potentiality of VS system. The validity of visual data acquired,
processed and possibly stored by the VS system , as a proof in front of a court of law
one of such issues. But visual data can be m odified using sophisticated processing t
w ithout leaving any visible trace of the m odification.
o digital
S or im age data have no
value as legal proof, since doubt would always exist that they had been intentionally
tam pered with to incrim inate or exculpate the defendant. Besides, the video data ca
created artificially by computerized techniques su
ch as morphing. Therefore the true
origin of the data m ust be indicated to use them as legal proof. By data authenticat
m ean here a procedure capable of ensuring that data have not been tam pered w ith
indicating their true origin.
A pool of digital cameras is in charge of frame the scene of interest and sent
corresponding video sequence to central unit. The central unit is in charge of analyz
the sequence and generating an alarm w henever a suspicious situation is detected.
unit also transm its the video sequences to
intervention
an centre such as security service
provider, the police departm ent or a security guard unit. Som ewhere in the system
video sequence or som e part of it m ay be stored and w hen needed the stored seque
can be used as a proof in front of court
of law . If the stored digital video sequences have
to be legally credible, som e m eans m ust be envisaged to detect content tam pering
reliably trace back to the data origin
3. AUTHENTICATION TECHNIQUES
Authentication techniques are perform ed on visual data to indicate that the data is
forgery; they should
not dam age visual quality of the video data. At the sam e tim e, th
techniques must indicate the malicious modifications include removal or insertion o
certain fram es, change of faces of individual, tim e and background etc. Only a prope
authenticatedvideo data has got the value as legal proof. There are two m ajor techni
for authenticating video data. They are as follows
Cameras calculate a digital summary (digest) of the video by means of hash functio
Then they encrypt the digest with their
private key, thus obtaining a signed digest which
is transm itted to the central unit together w ith acquired sequences. This digest is us
prove data integrity or to trace back to their origin. Signed digest can only read by u
public key of the camera.
4. CRYPTOGRAPHY
M ounting concern over the new threats to privacy and security has lead pread
to wide s
adoption of cryptography. Cryptography is the science of transform ing docum ents. I
mainly two functions
♦ Encryption
♦ Decryption
The purpose of encryption is to render a docum ent unreadable by all except those w
authorize to read it.
Cryptographers refer to the content of the original docum ent as p
text. Plain text is converted in to cipher form using an algorithm and a variable or ke
The key is a randomly selected string of numbers.
PLAIN
TEXT CIPHER TEXT
ENCRPTION DECRYPTION
ENCRYPTION ALGORITHEM KEY DECRYPTION
KEY ALGORITHEM
CIPHER
TEXT PLAIN TEXT
Only after decoding the cipher text using the key the content
the docum
of ent is revealed
to the common people. Encryption schemes are classified in to :
1. Symmetric encryption
In which the same key is used to both encode and decode the document.
It requires a pair of
keys: one for encrypting the plain text and the other for decrypting
the cipher text. A file encrypted w ith one key of a pair can be decrypted w ith other k
the same pair
To authenticate visual data each video cam era is assigned a different public or
keypriva
pair, with private key hardwired within the cameras. A document encrypted with the
private key of any particular camera can be decrypted with its own public key. This
property is used to provide center authentication that is to trace back to origin
the true
of
the data.
Before sending the video sequence to the central unit, cameras calculate a digital
sum m aryor digest of the video by m eans of a proper hash function. The digest is the
D
o
C
u
Engineering
6
Image Authentication Techniques
encrypted with their private key. Encryption is done by considering the digitized val
the brightness of each pixel. Digital signal is a sequence ofzeros and ones and it is
encrypted with the private key using a proper algorithm . The signed digest thus obt
is then transmitted to the central unit together with the acquired visual sequence.
Later the signal digest is used to prove data integr
ity or to trace back to their origin. The
signed digest is read using the public key of the cam era which produce the video an
check if it corresponds to the digest derived from the decrypted video content using
same hash function.
Any m anipulation
of the data w ill change the calculated im age digest derived from the
decrypted data. Any discrepancy between the decrypted digest and calculated imag
digest indicate that the data has been tam pered, w ith identical digest indicates that
data is genuine.
Value of the visual data can be added by tying each fram e to the particular label of
instant the fram e has been produced yet. This can be achieved by printing date and
of creation of each frame. Any modification of either the date time
or could be easily
revealed since it would change the locally calculated image digest.
Even though cryptographic data authentication is highly resistant to content tam per
suffers from few drawbacks. They are as follows:
1.Knowledge of private key
If the manipulatorknows the private key of the camera, he can change the digest to
involve the m odifications he had m ade on the actual sequence. But the possibility o
a thing is very small because the private key is hardwired within the camera.
4. Delay in transmission
Digest calculations and encryption introduces a delay in transmission of video
documents. This is harmful in system where the timely generation of alarm is critica
6. WATERMARKING
Digital w aterm ark is a signal which added to a docum ent to authenticate it and to p
the ow nership. A com m only encountered
digital w aterm ark is the logo m ost television
channels display on the top of the television screen. Not only does it advertise the c
but also provides the legal benefit of having a source signature persist during video
recording. Watermark task consists of two main steps
1. W aterm ark casting:-in w hich the signal represented by the w aterm ark is transm it
o v e r th e c h a n n e l, th a t is in w a te rm a rk c a s tin g a n e n c o d etakes
r fu n catiohost
n image
‘f’ and a watermark ‘w’ and generate a new image
Fw= (f, w)
2. Watermark detection:-in which the signal is received and extracted from possibly
corrupted image.
D
i
o
C
p
r
E
n
n
10
Image Authentication Techniques
Fragile waterm arking refers to the case w here w aterm ark inserted w ithin the data is
or altered as soon as host data undergoes any m odification. W aterm ark loss or alter
is taken as evidence that data has been tam pered w ith, whereas the inform ation co
within data used to demonstrate data origin
In case of robust w aterm arking a sum m ary of the candidate fram e or video sequenc
com puted and is inserted w ithin video
the sequence. Inform ation about the data origin i
also with the summary. To prove data integrity the information conveyed by the
w aterm ark is recovered and com pared w ith the actual content of the sequence. The
mismatch is taken as an evidence of data tampering. The capability to localize the
manipulation will depend on the summary of which is embedded in to the image.
WATERMARKING
FRAGILE ROBUST
WATREMARKING WATERMARKIN
G
Sem i fragile w aterm ark is m ore m ature than robust waterm arking. Tam per localizat
easier in fragile watermarking but it is difficult todistinguish between malicious and
innocuous manipulations.
Im age authentication by m eans of robust waterm arking is very prom ising w ith regar
the distinction betw een m alicious and innocuous m anipulations. The robustness of s
technique depends on the number of bits that can be hidden in to the image.
D
i
o
C
p
r
E
n
n
Image Authentication Techniques
By considering particular cases the requirem ent for the authentication of VS data ar
following.
The authentication technique m ust not deteriorate the visual quality of data. In this
case how ever this is not crucial issue since VS do not exhibit a com
quality
parable that of
visual data used in a m edia. VS data acquired by inexpensive, low quality devices. V
7. WATERMARKING ALGORITHM
In the proposed m ethod as indicating in the figure the waterm ark generation by usin
pseudo random number generator and appropriate thresholding.
The w aterm ark key used for waterm arking a specific fram e in the sequence is com p
of the cam era id and fram e num ber. Generating the w aterm ark key using fram e num
provides the advantage of producing frame–dependent watermarks.
In this case, fram e rem oval or fram e substitution can be easily detected as non auth
To perform authenticity check, the detection should first know the fram e num ber in
sequence which is always zero.
Watermark embedding is performed by altering all the pixels of the original frame
according to the following formula.
In the w aterm ark detection procedure, the detector generates first the water m ark f
fram e to be checked. To do so, the id num ber of cam era that produce the sequence
the frame number are needed.
If the first level decision test indicates that the im age is som ehow altered but authe
alternations m ade on the im age are concentrated in certain regions (Malicious tam p
or one spread on the image (innocuous alternations).
D
i
o
C
p
r
E
n
n
18
Image Authentication Techniques
8. OTHER APPLICATIONS
1. It canbe used for everything from sending e-m ail and storing m edical record
and legal contracts to conducting on-line transactions.
9. ADVANTATGES
10. DISADVANTAGES
2. The detector should know the fram e num ber in order to perform authenticity che
11. CONCLUSION
In these modern eras, visual surveillance system finds application in almost all fields
ranging from com m ercial to defense. The video data acquired
by VS system are form ing
vital evidence for several legal situations. So for such situations, the importance of
authenticating their content is very high. Cryptography and watermarking based
authenticating techniques are quite safe and efficient for
purpose
this and they are likely
to remain for quite for some while.
12. REFERENCES
2. “Digital Watermarking for protecting piracy”, Electronics for you, January 2003.
4. M.M.Yeung and F.Mintzer “A watermark for digital image” IEEE Spectrum, April
2002.
5. www.ctr.columbia.edu