Binus University Repository

CANDRA, SEVENPRI

English

Detecting Online Child Grooming Conversation

Empirical Relation Between Speech Signals and Heart Rate

Anggraeni, Adilla

Essence of Luxury
by Adilla Anggraeni
Submission date: 23-May-2018 07:10AM (UTC+0700)
Submission ID: 967384167
File name: Essence_of _Luxury_Consumption_and_Counterf eits.pdf  (491.19K)
Word count: 3949
Character count: 22345











7%
SIMILARITY INDEX
5%
INTERNET SOURCES
3%
PUBLICATIONS
1%
STUDENT PAPERS
1 1%
2 1%
3 1%
4 1%
5 1%
6 <1%
7 <1%
8 <1%
9 <1%
10 <1%
Essence of Luxury
ORIGINALITY REPORT
PRIMARY SOURCES
www.emeraldinsight.com
Internet  Source
Submitted to University of Glasgow
Student  Paper
Anne E. Wilcock, Kathryn A. Boys. "Reduce
product counterfeiting: An integrated
approach", Business Horizons, 2014
Publicat ion
emrbi.org
Internet  Source
Submitted to Nottingham Trent University
Student  Paper
espace.curtin.edu.au
Internet  Source
journals.cluteonline.com
Internet  Source
bear.warrington.ufl.edu
Internet  Source
www.andrew.cmu.edu
Internet  Source
Yixing (Lisa) Gao, Anna S. Mattila, Seoki Lee.
"A meta-analysis of behavioral intentions for
environment-friendly initiatives in hospitality
research", International Journal of Hospitality
Management, 2016
Publicat ion
11 <1%
12 <1%
13 <1%
Exclude quotes On
Exclude bibliography On
Exclude matches < 10 words
lib.dr.iastate.edu
Internet  Source
www.dekort-partners.eu
Internet  Source
ecoman.ktu.lt
Internet  Source


Similarity "Consumption and Counterfeits:The Role of Brand Familiarity, Perceived Quality and Anticipated Regrets"

Heriyati, Pantri

AMITY
by Pantri Heriyati
Submission date: 07-Jun-2018 02:16PM (UTC+0700)
Submission ID: 973293144
File name: AMITY_journal_Vol_1_No_2_2016.pdf  (304.35K)
Word count: 4873
Character count: 27880












14%
SIMILARITY INDEX
8%
INTERNET SOURCES
12%
PUBLICATIONS
7%
STUDENT PAPERS
1 2%
2 2%
3 1%
4 1%
5 1%
6 1%
AMITY
ORIGINALITY REPORT
PRIMARY SOURCES
www.deepdyve.com
Internet  Source
Richard Speed. "Maximizing the potential of
strategic typologies for marketing strategy
research", Journal of Strategic Marketing,
9/1993
Publicat ion
www2.warwick.ac.uk
Internet  Source
Simone Didonet, Geoff Simmons, Guillermo
Díaz‐Villavicencio, Mark Palmer. "The
relationship between small business market
orientation and environmental uncertainty",
Marketing Intelligence & Planning, 2012
Publicat ion
www.scribd.com
Internet  Source
publicgoods.eu
Internet  Source
7 1%
8 1%
9 1%
10 1%
11 1%
12 1%
13 <1%
14 <1%
Submitted to University of Leicester
Student  Paper
faculty.mu.edu.sa
Internet  Source
Roopali Batra, Satish Verma. "Non-financial
criteria in project appraisal methodologies:
empirical evidence from Indian companies",
International Journal of Accounting and
Finance, 2018
Publicat ion
Submitted to Gordon Institute of Business
Science
Student  Paper
Saara Hyvönen, Matti Tuominen, Leena
Erälinna. "Effects of Market-Related Assets on
Innovativeness and Operational Performance in
SMEs", The International Journal of
Entrepreneurship and Innovation, 2016
Publicat ion
Submitted to Macquarie University
Student  Paper
Submitted to Trident University International
Student  Paper
Abdelkader Daghfous, Nicholas Jeremy Ashill,
Michel Roger Rod. "Transferring knowledge for
15 <1%
16 <1%
17 <1%
18 <1%
19 <1%
20 <1%
21 <1%
22 <1%
organisational customers by knowledge
intensive business service marketing f irms",
Marketing Intelligence & Planning, 2013
Publicat ion
"Marketing Dynamism & Sustainability: Things
Change, Things Stay the Same…", Springer
Nature, 2015
Publicat ion
Asia Pacif ic Journal of Marketing and Logistics,
Volume 27, Issue 5 (2015)
Publicat ion
www.studyblue.com
Internet  Source
www.emeraldinsight.com
Internet  Source
m3.ithq.qc.ca
Internet  Source
Submitted to University of Newcastle
Student  Paper
Venkatraman, N., and John C. Camillus.
"Exploring the Concept of "Fit" in Strategic
Management", The Academy of Management
Review, 1984.
Publicat ion
Alan C.B. Tse, Leo Y.M. Sin, Oliver H.M. Yau,
Jenny S.Y. Lee, Raymond Chow. "A f irm's role
Exclude quotes On
Exclude bibliography On
Exclude matches < 10 words
in the marketplace and the relative importance
of market orientation and relationship
marketing orientation", European Journal of
Marketing, 2004
Publicat ion


Similarity "Assessment of Companies’ Peripheral Vision Capability - The Case of the Indonesian Manufacturing Industry"

Sasmoko, Sasmoko

Form Reviewer "Corruption Early Prevention: Decision Support System for President of the Republic of Indonesia"

Girsang, Abba Suganda

Form Reviewer "The Use of Data Mining For Prediction of Customer Loyalty"

Form Reviewer "Knowledge capability and key stakeholders' readiness in ERP implementation success"

Form Reviewer "Improvement Of Production System Efficiency and Production Capacity Using Line Balancing Method"

An Adaptive Cat Swarm Optimization Based On Particle Swarm Optimization Approach (ACPSO) For Clustering

Image Hiding Optimization Using 
Ant Colony Optimization Algorithm 
Abba Suganda Girsang 
Master in Computer Science, Bina Nusantara University, 
JI Kebon Jeruk Raya No 27, Jakarta Indonesia 
Email : agirsang@binus.edu 
Abstract- Deterioration in the quality on host media (such as 
image) usually occurs when the information (such as text or 
image) is hidden or embedded in the host. The smaller 
deterioration on host media after embedding the information, the 
system hiding is more successful because it shows there is no 
significant different between the original host image and the host 
image after embedded. This study proposes hiding information 
into a host image so that the deterioration is minimal using least 
significant bit (LSB) substitution and modified ant algorithm. 
The minimal deterioration is achieved when each pixels on the 
information is appropriate hidden on pixels on host media. The 
minimal deterioration as objective function is get by chasing the 
maximal peak signal-to-noise ratio (PSNR). This proposed 
method is implemented on some hosts image with two image 
information. The results show the host embedded image has no 
the different significant with the original host image with PSNR 
between about 63-73 
Keywords- Information hiding; Deterioration; Optimal; Ant 
colony optimization 
1. INTRODUCTION 
Currently, data hiding using digital media such as image, 
video, audio, and text becomes interesting topic. Watermark 
and steganography are two topics which relate to hiding 
information .. Several methods of them have been developed 
which can be implemented such as LSB (least significant bit) 
substitution [1], DWT and Huffman encoding [2], quantization 
table modification, and other technique. Research using LSB 
substitution method has been done as described by Blum et al 
[3]. LSB is a common technique used in the encryption and 
decryption of secret information. This method transforms the 
data into bit value, then replaces the lowest significant bit with 
new value. 
The LSB purposes transmitting confidential information to 
a receiver without being noticed by others when the message 
is sent [3]. In this work, LSB substitution replaces the cover 
image's bit with the all bits of the secret message. After 
substitution process, the new image constructed is called stego 
image. It has many possibilities mapping the secret image into 
the LSB of cover image. This mapping will determine the 
quality of stego image. The quality represents the similarity 
between the stego image with the cover (host) image. The 
stego image is more similar with the cover image, the mapping 
is identified more success or optimal. Their different is 
Fauzi Pujanandi Utama 
Master in Computer Science, Bina Nusantara University, 
JI Kebon Jeruk Raya No 27, Jakarta Indonesia 
Email : Pujanadi90@gmail.com 
represented as a stego image is a deterioration of host image 
and will be avowed by peak signal-to-noise ratio (PSNR). 
The optimization using LSB substitution has been 
developed in some various method, such as genetic algorithm 
(GA) [4], ant colony optimization (ACO) [5], and cat swarm 
optimization algorithm [2]. These studies aim to optimize the 
LSB substitution so that the deterioration of host image can be 
minimal. 
ACO algorithm is widely used to solve combinatorial 
optimization problem such as the traveling salesman problem 
(TSP) [6] [7], scheduling [8], repairing inconsistent matrix 
AHP [9] [10] [11], and so forth. ACO is an algorithm based on 
the habits of ants which was introduced Dorigo et al (1996) to 
solve TSP [6]. Refers to the performance of ACO, this method 
is tried to use ACO to get the optimal insertion for hiding 
information. By using ACO, the secret information is 
embedded into cover image with the LSB substitution based 
IT. RELATED WORK 
To set a cornerstone for a common understanding before 
constructing the research model, some related research 
perspectives based on the result of the literature analysis are 
briefly described in this section 
A. Ant System 
Ant system is the basis for the application of the Ant 
Colony Optimization (ACO), which was first introduced by 
Marco Dorigo and his colleagues in the early 1990s. The 
algorithm development inspired by the observation of the ant 
colony. Ants are social insects, where their actions are 
governed by the purpose colonies to survive. Ants in search of 
food can find the shortest path from the nest by using chemical 
substances produced are pheromones. Behavior that inspired 
the development of ant colony algorithm [12]. Figure 1 above 
shows how ants gets the shortest path. Ant start from point A 
to point E with shortest distance shows in Figure 1 a. Then an 
obstacle causes two alternative path with different distance 
can be taken of each ant in point B, which point C is shorter 
distance than point H. At the start both point C and H have 
equal probability to be chosen by ants at point B. Then some 
ants of the first group choose point C while the others choose 
point H. During the trip ants leave a certain amount of 
978-1-5090-5671-2/16/$31.00 ©2016 IEEE 
pheromone as trail, later the concentration of the pheromone 
count as part of probability for the second ants group in point 
B. As the sortest path requires a shorter time, then the level of 
pheromone of the trail BCD are higer than BHD. It impact 
BCD probability higer than BHD to the ants next group at 
point B. Ants recognize the pheromone on trails at the point B 
before they choose their route, the more ants go through BCD 
the stronger pheromone level it has, then pheromone in BHD 
will disappear due to evaporation. 
E 
A 
a) 
E 
al fl 
ih� 
.P"D� 
I"H".- lI!o !!lb s·!all!lclcIlllE 
1>:&<$ 
at 
A 
b) 
Fig.l An example with real ants [6] 
(a) Ants follow the path between A and E 
A 
c) 
(b) An obstacle makes ants have two optional path. 
(e) On the shorter path more pheromone is laid down 
From paragraph above, it has two important parts of ant to 
decide next node to be visited, distance and pheromone 
influence the probability next hop. Dorigo [6] has formulated 
the probability Eq. (1) as follows: 
(1) 
Probability (Probij) of next hop will be taken by ants get 
from pheromone on trail node i and node j (Tij) power a then 
multiple by distance power fJ divide sum of node i to other 
nodes j. When ant move to another node it changes the 
pheromone concentration using Eq. (2) below. 
(2) 
TlJ IS pheromone at trail ij, where t are previous time and n IS 
incremental from ant, and p are constant, so to update new 
value of pheromone p multiple by pheromone on trail add by 
sum of pheromone one trail ij 
B. Least Significant Bit (LSB) Substitution 
LSB is a common technique used in the encryption and 
decryption of secret information. LSBs are an ideal space to 
hide the secret data. Because the changing value of LSB does 
not change the significant of value of the host. The purpose of 
LSB is to transmit confidential information to a receiver 
without being noticed by others when the message is sent [3]. 
Fig. 2 shows the mechanism of LSB method on an 8 bit image 
by utilizing the 4-bit LSB. The value of cover or host image is 
stated as the bit value "11111010 ". Four bits LSB "1010 " are 
ideal bits changed by the bits of secret data (1101). Therefore , 
the stego image (cover+message) is identified as 11111101. 
Fig. 2. LSB Substitution 
m. THE PROPOSED METHOD 
A. The Concept 
The method of this study is to map LSB of secret image 
into cover/host image using ant colony optimization. This 
mapping occurs the deterioration in the quality of the 
original/coverlhost image. However the good mapping causes 
the difference between before and after embedding the secret 
image into cover image is not significant. To identify the 
similarity of two images (host and secret), each pixels of 
image is extracted into bits value. In a RGB color model , each 
pixels has three color; red, green, and blue. One color is 1 byte 
which consists 8 bits. Therefore, a pixel image is represented 
totally 24 bits with 8 bits red, 8 bits red, and 8 bits blue as 
shown a sample on Figure 3. 
10111111 11010101 01010110 
R G B 
Fig 3. One pixel is represented 24 bits 
However, to alleviate the deterioration of the result (stego) 
image, only bits in LSB position will be swapped by the bits 
of secret images. The bits on MSB position is retained. The 
concept of creating the stego image which is represented 
secret image on cover image is shown on Fig. 4. 
Each pixels of images is divided into some blocks. One 
block consists 4 bits. Thus one pixel is composed 6 blocks. In 
Fig.5 blocks of host image are represented 1,2 ,3,4,5, and 6 
while blocks of secret image are represented a, b, c, d, e, and f. 
Each LSB block position on host image is represented 2, 4 and 
6. 
Fig. 4 The concept of stego image. 
These blocks are the spaces where all blocks of secret image 
will be kept. So for this case, the number pixel of host image 
must be at least twice number pixel of secret image. 
R G B 
Host image lQll llll llQl 0101 0101 0110 
1 2 3 4 5 6 
R G B 
Secret image 0010 1001 1001 0101 1101 1011 
a b c d e f 
Fig.5 Representation block host and secret image 
Suppose a host image has 2 pixel with blocks 1,2,3, .... , 
12 as shown Fig 6.(a), and secret image has 1 pixel with 
blocks a,b,c,d,e, and f, as shown Fig. 6 (b). Since 1,3,5,7,9 and 
12 are the MSB block, the bits of them are retained. Yet the 
bits of LSB block (2,4,6,8,10, and 12) are replaced by the bits 
of the block of secret images. Therefore, one of solution 
hiding might be as shown Fig.6 (c) 
I� 1 3 1 + 1 5 Ie 1 7 1 &  1 9 1 .w l ll l � 1 
(a) Blocks of host image 
l a l b l c l d l e l f l 
(b) Blocks of secret image 
f 1 3 1 b 5 a 7 d 9 e 11 c 
(c) Blocks of Stego image as mapping secret into host image 
Fig. 6 Sample of a process image hiding 
Tn this study the pixels image are read flow vertically 
as shown Figure 7. 
pi 1'5 1") 1'13 
1'2 1'6 1'10 1'14 
pJ 1'7 I'll 1'15 
p4 1'8 1'12 1'16 
Fig. 7. Reading image pixel flow 
B. The Proposed Algorithm 
Implementation of ACO in this mapping is done as follows. 
Each blocks LSB identified as nodes that must be traveled by 
ants. While visiting each blocks LSB, each of blocks of secret 
images is mapped. While one block secret image is mapped 
into block LSB, the block LSB can not be used in next node. It 
can be illustrated as shown on Fig. 7. Two important variables, 
T and 11 in ACO are defined as follows. T is pheromone in 
edge between one block to candidate block of secret image. 11 
is used to represent the heuristic distance. Suppose x is 
represented one block of LSB host image and y is one block of 
one block secret imag, 1Jxy can be determined as Eg.(3) 
17xy = 1 + I Ex-Ey I (3) 
The process moving one block to the next block is used Eq.(I) 
When ant moves to another node it changes the pheromone 
concentration using Eg. (2). 
Block6 
BlockS 
Blockl0 8 
BIOCk12 8 
----. Alternative-l 
---. Alternative-2 
Fig. 7 Mapping blocks of secret images to block of LSB host 
image using ACO 
After traveling all nodes, the quality of ants are evaluated by 
the peak signal noise rate (PSNR) as shown Eg. (4), and Eq. 
(5). 
(4) 
The MSE between the cover image Z and the stego image H is 
defmed as follows: 
(5) 
Fitness function is used to evaluate each proccess. In this 
research, we use PNSR in the measurement of the fitness as 
follows: 
Therefore the fitness function can be defmed as Eq. (6) 
---- ) 
(6) 
The proposed algorithm can be summarized as shown 
Algorithm. 1 
Initialization and setting parameters 
Extracted the cover image into bits 
Categorized bits into two blocks. The tirst four bits are most 
significant bit (MSB), the rest are least significant bit (LSB) 
Extracted the secret image into bits 
F or each iterations 
F or each ants 
Constmcting the path 
Evaluate the fitness value using PNSR 
Update pheromone relates its performance 
Keeping the best ant tour 
End ants 
Comparing the quality tour with the previous tour 
Storing the best tour 
Algorithm I. The proposed algorithm 
IV. RESULT ANALYSIS 
A. The datasets and setting paramater 
In this work, six images are used as host images. These images 
comes from Signal and Image Processing (SIPI) University of 
Southern California datasets as shown Fig. 8. 
a) Girl b) Baboon c) Pappers 
d) Airplane e) JeUy beans f) Sailboat on lake 
Fig 8. The dataset of cover images 
These cover images are converted into 32 bit PNG with 
adjusting 100xi00 pixels. 
Two secret images ("BINUS ") are used in this research as 
shown Fig. 9. One secret image contains several color (Fig. 
9(a) and the other contains only black-white color (fig. 9(b)). 
Secret image use 32 bit PNG format with 50x21 pixel 
�s UNIVERSITY 
a) Logo b) Text 
Fig 8. Two secret images 
The parameters in this algorithm are set as shown Table. 1 
Table 1. Setting Parameters 
Variable 
Ants 
Iteration (NC) 
Initiate Pheromone ('r) 
Evaporation (p) 
a 
j3 
Q 
B. Performance Analysis 
Value 
12 
100 
1 
fWJ� 
0,5 
1 
1 
1 
The experiments are conducted by mapping each secret 
images into each host images 10 times. The result of 
experiments are shown by Table 2. It shows that the PSNR is 
about 67-73. 
T bl 2 E a e xpenment R esu ts 
Experiments Host+Secret WorstPNSR BestPNSR 
Experiment 1 Girl +Logo 68.07 72.15 
Experiment 2 Girl + Text 69.12 73.41 
Experiment 3 Baboon +Logo 67.86 72.22 
Experiment 4 Baboon + Text 69.62 73.79 
Experiment 5 Pappers +Logo 68.11 72.51 
Experiment 6 Pappers + Text 69.57 73.94 
Experiment 7 Airplane +Logo 67.89 72.07 
Experiment 8 Airplane + Text 69.35 73.49 
Experiment 9 Jelly Beans+Logo 67.29 71.64 
Experiment 10 Jelly Beans I + Text 69.15 73.31 
Experiment 11 Sailboat on Lake +Logo 68.01 72.23 
Experiment 12 Sailboat on Lake + Text 69.62 73.84 
The best and worst hldmg of expenments 1,2, 5 and 6 as 
samples experiment are shown on Table 2. 
V. CONCLUSIONS 
This experiment results show ACO can optimizing hiding 
the information (image) on image by using LSB concept. The 
result shows the different between the original image and the 
stego-image is not so significant. It looks the value PSNR 
more than 70. The modified ACO or hybrid with another 
algorithm might be implemented to optimize the image hiding. 
Table 2: Results 
�INUS UNIVERSITY 
�5 UNIVERSITY 
Worst Mapping Best Mapping 
REFERENCES 
[I] W. Awad, "Information Hiding Using Ant Colony 
Optimization Algorithm," DifJus. Adopt. Glob. 
Complexity, Glob. vol. 2, no. March, pp. 16-18,2013. 
[2] Z.-H. Wang, c.-c. Chang, and M.-C. Li, "Optimizing 
least-significant-bit substitution using cat swarm 
optimization strategy," in! Sci. (Ny)., vol. 192, pp. 98-
108,2012. 
[3] C. Blum and A. Roli, "Metaheuristics in combinatorial 
optimization: overview and conceptual comparison," 
ACM Comput. Surv., vol. 35, no. October 2015, pp. 189-
213,2003. 
[4] R. Z. Wang, C. F. Lin, and J. C. Lin, "Image hiding by 
optimal LSB substitution and genetic algorithm," Pattern 
Recognit., vol. 34, no. 3, pp. 671--683, 2001. 
[5] c.-S. H. c.-S. Hsu and S.-F. T. S.-F. Tu, "Finding 
Optimal LSB Substitution Using Ant Colony 
Optimization Algorithm," Commun. So/two Networks, 
2010.iCCSN '10. Second into Con!, pp. 0--4, 2010. 
[6] M. Dorigo, V. Maniezzo, and A. Colorni, "Ant system: 
optimization by a colony of cooperating agents," Syst. 
Man, Cybern. Part B Cybern. IEEE Trans., vol. 26, no. 
I, pp. 29--41, 1996. 
[7] M. Dorigo and L. M. Gambardella, "Ant colony system: 
a cooperative learning approach to the traveling salesman 
problem," IEEE Trans. Evol. Comput., vol. I, no. I, pp. 
53-66,1997. 
[8] C. Blum and M. Sampels, "An ant colony optimization 
algorithm for shop scheduling problems," J. Math. 
Model. Algorithms, vol. 3, no. 3, pp. 285-308, 2004. 
[9] A. S. Girsang, C.-W. Tsai, and C.-S. Yang, "Rectifying 
the Inconsistent Fuzzy Preference Matrix in AHP Using 
a Multi-Objective BicriterionAnt," Neural Process. Lett., 
vol. 44,no. 2,pp. 519-538,2016. 
[10] A. S. Girsang, C.-W. Tsai, and c.-S. Yang, "Ant Colony 
Optimization for Reducing the Consistency Ratio in 
Comparison Matrix. " 
[II] A. S. Girsang, c.-W. Tsai, and c.-S. Yang, "Ant 
algorithm for modifying an inconsistent pairwise 
weighting matrix in an analytic hierarchy process," 
Neural Comput. Appl., vol. 26, no. Issue 2, p. pp--313, 
2014. 
[12] c. Blum, "Ant colony optimization: Introduction and 
recent trends," Phys. Life Rev., vol. 2, no. 4, pp. 353-
373,2005. 


Image hiding optimization using ant colony optimization algorithm

Binus University

Not a member yet