IRJET-Debarred Objects Recognition by PFL Operator

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 05 Issue: 06 | June 2018 www.irjet.net p-ISSN: 2395-0072
© 2018, IRJET | Impact Factor value: 6.171 | ISO 9001:2008 Certified Journal | Page 670
DEBARRED OBJECTS RECOGNITION BY PFL OPERATOR
Manish Kumar Srivastava1 ,Madan Kushwah2 , Abhishek kumar 3
1 Assistant Professor, Department of CSE, Lal Bahadur Shastri Group of Institutions Lucknow , Uttar Pradesh, India
2,3 Assistant Professor, Department of CSE, Bansal Institute of Engineering & Technology, Uttar Pradesh, India
------------------------------------------------------------------------***-------------------------------------------------------------------
Abstract - International security especially airport
security pose serious concern and have to be talked on
priority bases, Security has become one of the foremost
issues of apprehension that needs to be methodically
talked by every nation, in particular the developed
nations, which are playing an dynamic role in counter
terrorism. The planned system uses suitable preprocessed
X-ray images of passenger’s luggage and design to detect
the banned objects like Pistol, Knife, Explosive resources ,
scissors, and handguns of different size and orientation etc
The X-ray imaging is an important technology in many
fields, from non-intrusive assessment of elusive objects, to
weapons recognition at security checkpoints. In this work
we will detect the object by partial fuzzy logic
method(PFL) PFL operator delivers a parameterized
family of combination operators, include well-known
operators such as maximum, minimum, arithmetic mean,
k-order statistics and median. Sometimes, exact “and-
ness” is essential for multi-criteria decision making, which
offers minimum value and sometimes exact “or-ness”
which provides maximum value. The PFL aggregation
operator lies between the two extremes of and-ness and
or-ness. Two extremes are limited to mutually exclusive
probabilities for multiplication (like AND gate) and
summation (like OR gate). PFL operator is used to
estimate the degree of likeness of knives, scissor and
handguns.
Key Words: Partial Fuzzy Logic, Security, and
Prohibited Items like Pistol, Knife, and
Handguns.
1. INTRODUCTION
PROBABILITY STUDY
Better security in the aftermath of the 9/11 attack in the
United States of America has lead to added congestion in
airport terminals, interruptions, hassle, more boundaries
on carry-on luggage, a sense of anxiety, and sometimes a
breach of retreat between the public. All these simply add
cost to air-travel and thus have an effect on socio financial
factors. It has almost become an standard norm that
hundreds of flights have been recalled to terminals after
being air-born, plentiful events of relocation, passengers
rechecked, or even asked to take your clothes off.
The X-ray imaging is an important technology in many
areas, from non-intrusive inspection of delicate objects, to
weapons detection at security checkpoints.
1.1 REQUISITE FOR FUZZY LOGIC IN OBJECT
RECOGNITION SYSTEM
With the above scenario, the entire world must be looking
forward for a fuzzy object recognition system, which
responds to awareness based query in natural language in
an effective style. However, some of the vital task that
needs to be followed prior to object recognition is as
follows:
i. Assessment of fuzzy validity of hand drawn fuzzy
shapes.
ii. Assessment of fuzzy similarity among such family of
fuzzy shapes.
1.2 OBJECT RECOGNITION TECHNIQUES
In the recent past, the world faces the most hazardous
crimes in general. Mainly, the terrorism has panicked
people since a decade. The detection of threat objects
using X-ray luggage scan images has become an important
means of security. Most Computer Aided Screening (CAS)
is still base on the manual recognition of potential threat
objects by human expert's where probabilities of human
error is relatively high as thousands of bags need to be
scanned every day.
1.3 IMAGE SEGMENTATION
Image segmentation or division goes to separate an
image into its object classes. Clustering methods, edge
based methods, histogram-based methods, and region
growing methods offer different benefits and drawbacks.
The use of a Gaussian mixture expectation maximization
(EM) method has been investigated to realize
segmentation specifically for x-ray luggage scans.

Figure 1. Shows an input x-ray image and
examples of objects found by segmentation.
2. Fuzzy Logic
In this section, the conditions and the desperate
requirement of Fuzzy Logic, where objects for
computation are perception based verbal information,
instead of crisp measurements defined in terms of
numbers. Mostly, the problems of resolving such
perceptions in linguistics are carried forward by Zadeh for
a long time. In the beginning the attitude of implementing
linguistics is initiated in his serial papers.[1]
The fuzzy rule-based classification system creates too
many rules for high dimension problems. It is often
assumed that the numeral of fuzzy if-then rules
exponentially increases as the number of structures
increases. [2]For this purpose, only a small number of
features are selected for constructing a fuzzy classifier,
which decreases its accuracy. To solve this problem, we
present a multi-level fuzzy classifier consists of several
small fuzzy classifiers with a small number of structures,
which not only improve the performance of fuzzy classifier
but also solve the problem of high dimension.
2.1 CORRELATION FUNCTIONS
The person brain interprets the incomplete and partial
information delivered by the sensory organs. The fuzzy
logic delivers a systematic way for valuing this perception
or natural language based information. The fuzzy logic
used some arithmetical calculation on the basis of
linguistic qualifier used in the partial information. The
fuzzy inference system (if – then rules) or membership
functions are used convert the inaccurate in order into a
specific facts.
A fuzzy if-then rule assumes the form
If x is A then y is B,
Where A and B are linguistic values defined by fuzzy sets
on universes of discourse X and Y, respectively. “x is A” is
called antecedent and “y is B” is known as conclusion. For
example
If pressure is high, then volume is small. In fuzzy logic
membership function is used to map imprecise vague
information into a precise or crisp value. The membership
is the degree of belongingness of a particular value to
certain characteristics. For example if the temperature of
water is 20o then its membership value is closer to the
degree of coldness than the degree of hotness of water.
Fig 2. Membership function for valuing of degree of
belongingness of water with temperature.
3. PARTIAL FUZZY LOGIC METHODS (PFL)
Partial Fuzzy Logic (PFL) is the crucial concept of
information aggregation, was originally presented by
Yager.[3] PFL helps the means of aggregation in solving
problems that arises in multi criterion decision making.
Furthermore, PFL operator offers a parameterized family
of aggregation operators, with well-known operators such
as maximum, minimum, arithmetic mean, k-order statistics
and median. Sometimes, exact “and-ness” is essential for
multi-criteria decision making, which deals minimum
value and sometimes exact “or-ness” which offers
maximum value. The PFL aggregation operator lies
between the two extremes of and-ness and or-ness. Two
extremes are limited to equally exclusive probabilities for
multiplication (like AND gate) and summation (like OR
gate). Subsequent part discloses a brief account of PFL
operators, a detailed discussion about the behavior of
operators. The PFL operation involves three following
steps - 1) Reordering of inputs, 2) Weight determination
related with PFL operators, and 3) Aggregation process.
Original image
Segmentation Image
Original image
An object found by
segmentation

3.1 EXPLANATION OF PFL
Mapping the PFL operator R from R m R, (where R = [0,
1]), with dimension m, has weighting vector w= (w1, w2,
w3,… wm)T , where wj ∈ [0, 1] and ∑ wj = 1 , the summation
of individual weights will always found to be one. Thus, for
the multi-criteria of size m, the input parameter (x1, x2,
x3……xm), the PFL determines the f-validity in f-geometric
figures as follows:
where yj is the jth largest number in the vector(x1, x2,
x3,…xn), and y1 ≥ y2 ≥ y3 ≥ …≥ ym. However, the weights wj of
the operator R are not related with any exact value of xj ,
instead they are related with the ordinal position of yj.
The minimum and maximum range of values can be
decided based upon the concept of or-ness (β).
3.2 MANIPULATIVE PFL WEIGHTS
One of the vital tasks is to compute the weights. We use
the linguistic quantifier denoted as Q(r), to generate the
weights wj. Q(r) satisfies two properties: i) Q r
∈ [0, 1], such that Q(r) = 1. Furthermore, Q(r) is non-
decreasing if possesses the following property:
 1,0,  ba ,
when a > b then Q (a) ≥ Q (b).
The membership function of a relative quantifier can be
characterized as:
 













brif
arbif
ab
ar
arif
rQ
1
0
where a,b,r ∈ [0,1].
In Yager calculates the weights wj of the PFL aggregation
from the function Q describing the quantifier, with m
number of criteria.





 







m
j
Q
m
j
Qw j
1
The following figures are atmost,atleast half and as many
as possible.
Figure 3. Atmost , Atleast half and as many as possible.
4. Experiments and outcomes
The experiment are made on some sample images after
preprocessing. Some images of knives, scissors and
handguns have been presented in figure respectively. Each
image is of 96x96 pixel per inches and height and width of
image scale for 1x1 inch. Moreover Tables comprises of
mutual membership values of all the sample images of
knives, scissors and hand guns respectively.
Figure 4. Sample Images Of Knives Taking As Inputs
k1 k2 k3 k4 k5
k6 k7 k8 k9 k10
j
m
j
j yw1
m321 =)x……x,x,(xPFL
 1
1-m
1
=
1

mw
m
j
j

Table 1. Membership values of Knives
k1 k2 k3 k4 k5 k6 k7 k8 k9
k1
0
k
1 1
0.
57
4
0.
51
17
0.
50
98
0.
38
38
0.
58
46
0.
45
5
0.
84
46
0.
53
74
0.
59
16
k
2
0.
93
95 1
0.
59
09
0.
59
08
0.
42
38
0.
80
31
0.
81
35
0.
51
45
0.
66
08
0.
66
08
k
3
0.
97
41
0.
42
86 1
0.
96
13
0.
35
9
0.
49
62
0.
47
46
0.
82
01
0.
48
45
0.
62
12
k
4
0.
97
5
0.
42
86
0.
92
74 1
0.
35
99
0.
48
95
0.
45
95
0.
80
19
0.
47
6
0.
60
6
k
5
0.
89
63
0.
58
45
0.
60
83
0.
60
74 1
0.
64
06
0.
54
79
0.
83
01
0.
61
2
0.
62
12
k
6
0.
94
15
0.
65
35
0.
51
96
0.
52
4
0.
34
84 1
0.
47
09
0.
75
58
0.
48
52
0.
60
88
k
7
0.
82
32
0.
51
9
0.
52
96
0.
54
21
0.
44
93
0.
54
98 1
0.
75
25
0.
55
69
0.
60
69
k
8
0.
99
49
0.
19
47
0.
19
6
0.
20
7
0.
15
68
0.
25
28
0.
24
08 1
0.
29
16
0.
27
72
k
9
0.
90
38
0.
49
45
0.
55
57
0.
56
1
0.
36
55
0.
53
16
0.
44
21
0.
71
95 1
0.
62
k
1
0
0.
96
71
0.
34
69
0.
44
31
0.
45
62
0.
28
71
0.
39
74
0.
38
29
0.
72
01
0.
44
17 1
We have taken only one example from the above
mentioned items.
4.1 RESULTS FOR KNIVES(VSM)
An image of object is a group of pixel values. This image
can be measured as a 2-dimensional vector space (matrix)
each subscripts of vector comprises of pixel value. Lets
two matrixes A and B of unlike image are subtract for
finding the parallel. [5]The resultant vector is a matrix C.
The resultant matrix comprises of subscripts value zero
due to the similar pixel values. These zero values are taken
as count of similarity measure. The high number of the
zero counts in each column leads to the higher similarity.
The above said concept is basic of the VECTOR SPACE
MODEL. Each object is characterized as vector. The size of
the vector is number of column. Each element of vector
has sum of matching count i.e. zeros of that specific
column.
The ‘Figure 4.1 to 4.2’ compared the results for
similarity for different images of knives taken by
the PFL Method.
4.2 RESULTS FOR KNIVES(FSM)
In the environment where judgment making is very acute
for time saving as well as security, for example air port or
railway station there is a long line of passengers. And
safety personnel have to mark decision on the basis of
their perception at the moment very fast. Either they have
to stop some body for security check or permit him or her
to passes through. Both the thing detain the innocent
person very painful as well as slipping of illegal object may
be very dangerous. The exact explanation of objects inside
the heavy and opaque baggage is very costly. In this case
partial fuzzy logic can be used for estimating the shape of
purpose. The ending decision can be occupied by using
Partial Fuzzy Logic method. [4]

Figure 4.3 Comparison between PFL and FSM.
5. CONCLUSION
The most common technique for screening luggage at
airports is with the use of X-ray technology. There are a
number of reasons why it is commonly accepted including
safety factors and the fact that the technology is well
agreed and relatively inexpensive. As the digital X-ray
technology suits more noticeable and based on the current
state-of-the-art in image processing, feature extraction
and classification technology. The role of computers in
screening luggage will increase in order to enhance
manual screening processes. A new method is suggested
to determine the optimal number of clusters when
segmenting X-ray images and to estimate the results
acquired by different segmentation methods Compared
with the statistical validity index method; our method
deliberates both the spatial and statistical information of
the image. [6]Preliminary experimental results indicate
that our method yields results reliable with the human
assessment. Another advantage of our method is that it is
computationally well-organized. Our procedure only
computes the Euclidian distance and Key Points.
ACKNOWLEDGEMENTS
The authors would like to thank MR. Madan Kushwaha,
Assistant Professor ,BIET, Lucknow; and Mr. Abhishek
Kumar,Assistant Professor, Bansal Institute of Engineering
& Technology, Lucknow, for his valuable support, guidance
and encouragement.
REFERENCES
[1] L.A.Zadeh, From Computing with Numbers to
Computing with Words – From Manipulation of
Measurements to Manipulation of Perceptions, IEEE
Transactions on Circuits and Systems-I, vol. 45, Pg:
105-119, (1999).
[2] B.M.Imran, M.M.S.Beg, Image Retrieval by
Mechanization and f-Principle, Proc. II International
Conference on Data Management (ICDM’2009),
February 10-11, IMT Ghaziabad, India, McMillan
Advanced Research Series, ISBN 023-063-762-0, Pg:
157-165, 2009.
[3] L.A.Zadeh, Toward Extended Fuzzy Logic - A First
Step, Fuzzy Sets and Systems, Information Sciences,
Pg: 3175-3181, (2009).
157-165, 2009.
[5] L.A.Zadeh, From Computing with Numbers to
Computing with Words – From Manipulation of
Measurements to Manipulation of Perceptions, IEEE
Transactions on Circuits and Systems-I, vol. 45, Pg:
105-119, (1999).
157-165, 2009.
[7] H.R.Tizhoosh, Fuzzy Image Processing: Potentials and
State of the Art, IIZUKA'98, 5th International
Conference on Soft Computing, Iizuka, Japan, October
16-20, Vol. 1, Pg: 321-324, (1998L.A.Zadeh, A Note on
Web Intelligence, World Knowledge and Fuzzy Logic,
Journal of Data & Knowledge Engineering, Pg.291–
304,(2004).
[8] H.R.Tizhoosh, Fuzzy Image Processing: Potentials and
State of the Art, IIZUKA'98, 5th International
Conference on Soft Computing, Iizuka, Japan, October
16-20, Vol. 1, Pg: 321-324, (1998).
[9] M.Thint, M.M.S.Beg, Z.Qin, Precisiating Natural
Language for a Question Answering System, Proc. 11th
World Multi-conference on Systemics, Cybernetics and
Informatics (WMSCI 2007), Orlando, USA, July 8-11,
Pg:165-170 ,(2007).

IRJET-Debarred Objects Recognition by PFL Operator

More Related Content

What's hot (18)

Similar to IRJET-Debarred Objects Recognition by PFL Operator (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET-Debarred Objects Recognition by PFL Operator