This document discusses different pattern recognition algorithms that could be implemented in real-time data sets. It begins by defining pattern recognition and providing examples. It then discusses why pattern recognition is important and lists several applications. The document goes on to describe three main approaches to pattern recognition - statistical, syntactic, and neural pattern recognition - and provides examples for each. It then provides more detailed descriptions and pseudocode for several specific algorithms, including KMP, Boyer-Moore, Rabin-Karp, naive string matching, and brute-force string matching. It concludes by discussing future work improving algorithm complexity and potential applications in biometric identification.

1. PROJECT ON
IMPLEMENTATION OF DIFFERENT PATTERN
RECOGNITION ALGORITHM IN REAL TIME DATA
SET
Presented By:
Sudipta Pan (M.Tech -CSE ,3rd semester)
Roll No.:10911213014
NETAJI SUBHASH ENGINEERING COLLEGE
2014

2. WHAT IS PATTERN RECOGNISITION
Given a text string T[0..n-1] and a pattern P[0..m-1],
find all occurrences of the pattern within the text.
Example: T = 000010001010001 and P = 0001, the
occurrences are:
first occurrence starts at T[1]
second occurrence starts at T[5]
third occurrence starts at T[11]

3. WHY I CHOOSE PATTERN RECOGNISITION
APPLICATION :
• Image preprocessing
· Computer vision
· Artificial intelligence
· Radar signal classification/analysis
· Speech recognition/understanding
· Fingerprint identification
· Character (letter or number) recognition
· Handwriting analysis
· Electro-cardiographic signal analysis/understanding
· Medical diagnosis
· Data mining/reduction

4. PATTERN RECOGNISITION APPROACH
Statistical Pattern Recognition
 Syntactic Pattern Recognition
Neural Pattern Recognition

5. Statistical pattern recognition
approach
Statistical pattern recognition attempts to
classify patterns based on a set of extracted
features and an underling statistical model for the
generation of these patterns. It assumes a
statistical basis for classification of algorithms.
For e.g:Speech Recognition

6. Syntactic pattern recognition
approach
The principle behind this approach is that many
times the interrelationships or interconnections of
features yield important structural information,
which facilitates structural description or
classification.
Therefore, in these approaches we must be able
to quantify and extract structural information and to
assess structural similarity of patterns.
For e.g. Recognizing areas such as
highways,rivers,and bridges in satellite pictures.

7. Neural pattern recognition
approach
This approach makes use of the knowledge of how
biological neural systems store and manipulate
information. They are particularly well suited for pattern
association applications.
Artificial neural networks (ANNs) provide an emerging
paradigm for pattern recognition implementation that
involves large interconnected networks of relatively
simple and typically nonlinear units so called neural-nets.
For e.g. Back propagation ,high-order nets,time –delay
neural networks and recurrent nets.

8. ALGORITHM DESCRIPTION
The Knuth-Morris-Pratt (KMP) algorithm:
It was published by Donald E. Knuth, James H.Morris and
Vaughan R. Pratt, 1977 in: “Fast Pattern Matching in
Strings.“
To illustrate the ideas of the algorithm, consider the
following example:T = xyxxyxyxyyxyxyxyyxyxyxxy
And P = xyxyyxyxyxx
it considers shifts in order from 1 to n-m, and determines
if the pattern matches at that shift. The difference is that
the KMP algorithm uses information gleaned from partial
matches of the pattern and text to skip over shifts that
are guaranteed not to result in a match.

9. KMP Algorithm cont…..
The text and pattern are included in Figure 1, with
numbering, to make it easier to follow.
1.Consider the situation when P[1……3] is successfully
matched with T[1……..3]. We then find a mismatch: P[4]
= T[4]. Based on our knowledge that P[1…… 3] =T[1……
3], and ignoring symbols of the pattern and text after
position 3, what can we deduce about where a potential
match might be? In this case, the algorithm slides the
pattern 2 positions to the right so that P[1] is lined up
with T[3]. The next comparison is between P[2] and T[4].

10. P: x y x y y x y x y x x
q: 1 2 3 4 5 6 7 8 9 10 11
_(q): 0 0 1 2 0 3
Table 1: Table of values for pattern P.
Time Complexity :
The call to compute prefix is O(m)
using q as the value of the potential function, we argue
in the same manner as above to show the loop is O(n)
Therefore the overall complexity is O(m + n)

11. Boyer-Moore algorithm:
It was developed by Bob Boyer and J Strother
Moore in 1977. The algorithm preprocesses the
pattern string that is being searched in text
string.

12. String
pattern matching - Boyer-Moore
This algorithm uses fail-functions to shift the pattern
efficiently. Boyer-Moore starts however at the end of
the pattern, which can result in larger shifts.Two
heuristics are used:1: if we encounter a mismatch at
character c in Q, we can shift to the first occurrence
of c in P from the right:
Q a b c a b c d g a b c e a b c d a c e d
P a b c e b c d
a b c e b c d
(restart here)

13. Time Complexity:
•performs the comparisons from right to left;
•preprocessing phase in O(m+ ) time and space
complexity;
•searching phase in O(mn) time complexity;
•O(n / m) best performance

14. Rabin-Karp ALGORITHM
The Rabin –Karp algorithm searches for a pattern
in a text by hashing. So we preprocess p by
computing its hashcode, then compare that hash
code to the hash code of each substring in t. If we
find a match in the hash codes, we go ahead and
check to make sure the strings actually match (in
case of collisions).

15. Concept of Rabin Karp Algorithm
The Rabin-Karp string searching algorithm
calculates a hash value for the pattern, and for each
M-character subsequence of text to be compared.
If the hash values are unequal, the algorithm will
calculate the hash value for next M-character sequence.
 If the hash values are equal, the algorithm will
compare the pattern and the M-character sequence.
In this way, there is only one comparison per text
subsequence, and character matching is only needed
when hash values match.
Time Complexity:
Running time for Rabin Karp algorithm is O(mn)in the
worst case, since the Rabin Karp algorithm explicitly verifies every
valid shift.

16. The NAÏVE STRING MATCHING ALGORITHM
The naive algorithm finds all valid shifts using a loop
that checks the condition P[1……m]=T[s+1….s+m] for
each of the n-m+1 possible values of s.
NAÏVE-STRING-matCher(T,P)
1.n=T.length
2.m=P.length
3.for s=0 to n-m
4. if P[1……m]==T[s+1….s+m]
5. print “Pattern occurs with shift” s

17. NAÏVE STRING MATCHING ALGORITHM CONT…..
Below figure portrays the naive string-matching procedure as “template”
containing the pattern over the text,noting for which shifts all of the characters
on the template equal the corresponding characters in the text. The for loop of
lines 3-5 considers each possible shift explicitly.The text in line 4 determines
whether the current shift is valid; this test implicitly loops to check
corresponding character positions until all positions match successfully or a
mismatch is found.Line 5 prints out each valid shift s.
Time Complexity:
1. The overall complexity is O(mn)

18. Brute-Force String Matching
A brute force algorithm for string matching problem has
two inputs to be considered: pattern (a string of m
characters to search for), and text (a long string of n
characters to search in).
Algorithm starts with aligning a pattern at the beginning of
text. Then each character of a pattern is compared to the
corresponding character, moving from left to right, until all
characters are found to match, or a mismatch is detected.
While the pattern is not found and the text is not yet
exhausted, a pattern is realigned to one position to the right
and again compared to the corresponding character, moving
from left to right.

19. Check each position in the text T to see if the
pattern P starts in that position
a n d r e w
r e wP:
a n d r e wT:
r e wP:
P moves 1 char at a time through T

20. Brute Force Pseudo-Code:
do
if(text letter==pattern letter)
compare next letter of pattern to next letter of text
else
move pattern down text by one letter
while(entire pattern found or end of text)
Analysis
Brute force pattern matching runs in time O(mn) in the worst
case.
But most searches of ordinary text take
O(m+n), which is very quick.

21. SCREEN SHOTS:

22. FUTURE SCOPE
I have try to improve the complexity of different
algorithms.From above five algorithms are consider and I will
try to develop a new algorithm.
I want to research in biometrics identification using a
fingerprint or face image or a recorded voice and others.
Using this features can be identify a person, first a feature set
are extracted from the image or the audio, then are compare
with an stored feature set, some algorithms and process are
well known to made this task

23. REFERENCES
1 .Fu, K. S. (King Sun), 1930- “Syntactic pattern recognitionand applications”
Englewood Cliffs, N.J. : Prentice-Hall,c1982
2.Schalkoff, Robert J, “Pattern recognition : statistical,structural, and neural
approaches” New York : J. Wiley,c1992
Journals:
Journal of Pattern Recognition Society.
IEEE transactions on Neural Networks.
Pattern Recognition and Machine Learning.
Books:
Duda, Heart: Pattern Classification and Scene Analysis. J. Wiley & Sons, New York,
1982. (2nd edition 2000).
3.H. F. Ebel, C. Bliefert, and W. E. Russey, The Art of Scientific Writing: From Student
Reports to Professional Publications in Chemistry and Related Fields. Weinheim:
Wiley-VCH Verlag GmbH &Co. KGaA, 2004.
4.F. M. Lastname. (2006, January). Formatting papers for Journal of
Pattern Recognition Research.
Journal of Pattern Recognition Research [Online]. 1(1). pp. 1-3.
Available: http://www.jprr.org

24. THANK YOU