Embed Size (px)
Citation preview
10TH INTERNATIONAL CONFERENCE ON IMAGE PROCESSING,
WAVELET AND APPLICATIONS IWW2019, 18-20 October 2019
ABSTRACT BOOKLET
2 3
E-Abstract Booklet
Copyright © by Kocaeli University with the co-operations of
İstanbul Aydın University,
EURAS Eurasian Universities Union and
Sharda University
All rights reserved.
No part of this booklet may be reproduced in any form or by any means,
without permission in writing from the publisher.
Published by
Nabi SARIBAŞ
C.Kenan ÖZKAN
Istanbul Aydın University Graphic Design Department
İnönü Cad. No: 38 Sefaköy - 34295 Küçükçekmece / İstanbul
Phone: 444 1 428 Fax: 0 212 425 57 59
www.aydin.edu.tr
All rights reserved.
This work is subject to copyright, no part so publication may be reproduced,
stored in aretrieval system or transmitted in any for more by any means,
electronic, mechanical photocopying, recording, and scanning without the
permission in writing by Kocaeli University, İstanbul Aydın University,
EURAS Eurasian Universities Union and Sharda University
The volume is solely the responsibility of the authors.
ABSTRACT BOOKLET
Editors: F.DOKMEN, Z.ASLAN, N.DURU, H.M. ERTUNÇ, H.OCAK Co-Editor: C.YAZICI
18-20 October, 2019 Kocaeli, Turkey
IWW 2019, 18-20 OCTOBER 2019
4 5
CONTENTS
A.H. Siddiqi,
P. Manchanda
APPROXIMATION BY WAVELETS
CONSTRUCTED THROUGH WALSH
FUNCTIONS WITH APPLICATIONS
23
Layth Hani
Al-Samarai,
Osman Nuri Uçan,
Oğuz Bayat
AN ANALYSIS OF ACCESSIBILITY IN
LEARNING MANAGEMENT SYSTEM IN
THE CONTEXT OF HIGHER EDUCATION
INSTITUTION
24
Mohammed Mahmood,
Osama Ashour,
Oğuz Bayat, Osman
Nuri Uçan
DESIGN AND IMPLEMENTATION OF WEB
BASED FOR INTERMEDIATE ONLINE SHOP
WITH LARAVEL FRAMEWORK
25
Naama Methab Athab,
Osman Nuri Uçan, Oğuz
Bayat and Saadaldeen
Ahmed
INTELLIGENT SYSTEM FOR PREDICTING
BEHAVIOR OF ELECTRICAL ENERGY
CONSUMPTION
26
Ali Saad Al-Naser,
Osman Nuri Uçan,
Oğuz Bayat
PATTERN-BASED BEHAVIORAL BIOMETRIC
AUTHENTICATION USING ARTIFICIAL
NEURAL NETWORKS
27
Lin Wang INFLUENCES OF THE NORTH ATLANTIC SEA
SURFACE TEMPERATURE ON THE EAST ASIAN
CLIMATE
28
29
Mert Dedeoğlu,
Levent Başyiğit
THE ESTIMATION OF WHEAT YIELD DURING
GROWTH STAGES USING VEGETATION
INDICES ON SENTINEL-2 IMAGES
30
Osman Nuri Uçan, Ziad M.
Abood, Ahmed R. Ibrahim
ASSESSMENT THE QUALITY MEDICAL
IMAGES ENHANCEMENT
31
İhsan Doğan, M.D. VIRTUAL AND AUGMENTED
MICRONEUROSURGERY
32
Tuba Kurban FUSION OF MULTISPECTRAL FACE IMAGES
USING TRANSFORM BASED METHODS
33
34
Kaplan Kaplan, Yasin
Cuşkun, H. Metin Ertunç
WAVELET DECOMPOSITION BASED FEATURE
EXTRACTION OF BRAIN MR IMAGES
35
Bülent Oruç, Seray
Bozkurt
DETERMINATION of SUBSURFACE POSITION
of BURIED MAGNETIC SOURCES USING
1-D COMPLEX CONTINUES WAVELET
TRANSFORM
36
Özlem Evrim Gündoğdu,
Nevcihan Duru
OLAY TABANLI METİN ÖZETLEME
YAKLAŞIMLARI
37
İsmail Malik Kundakcı,
Kaplan Kaplan, Hüseyin
Metin Ertunç
FAULT CLASSIFICATION OF ROLLING
BEARINGS WITH TIME-FREQUENCY
IMAGE ANALYSIS AND PRE-TRAINED CNN
ARCITECHTURE GOOGLENET
38
Ethem Kelekçi, Tuğce
Yaren, Selcuk Kizir
MODEL BASED PID CONTROLLER DESIGN OF
A DC MOTOR USING BLACK BOX SYSTEM
IDENTIFICATION METHOD
39
Ethem Kelekçi, Selcuk
Kizir
TOOL PATH PLANNING & CONTROL OF
THE 3-AXIS CNC MACHINING WITH HIGH
ACCURACY STEP MOTOR PULSE/DIRECTION
COMMAND
40
Suat Karakaya, Hasan
Ocak
DESIGN AND IMPLEMENTATION OF A
WHEELED MOBILE ROBOT PLATFORM
41
Amirmohammad Abhary HOW IMAGE PROCESSING CHANGED MINING
ENGINEERING?
42
Gürkan Küçükyıldız,
Hasan Ocak
BOTTLE COUNTING SYSTEM FOR A RE-
CYCLING FACILITY BASED ON PROCESSING
OF DEPTH IMAGES
44
Gürkan Küçükyıldız,
Hasan Ocak
IMAGE PROCESSING BASED PUPIL
TRACKING SYSTEM FOR A WHEEL CHAIR
SYSTEM
46
Bernd Richter MODEL OUTPUT STATISTICS (MOS) METHODS
OPERATIONAL APPLICATIONS AND VERIFICATION
47
Ergün Can, Ali Can
Karaca, Oğuzhan
Urhan, M. Kemal
Güllü
LOSSY COMPRESSION OF HYPERSPECTRAL
IMAGES USING WAVELET TRANSFORM,
PRINCIPAL COMPONENT ANALYSIS AND
TENSOR DECOMPOSITION
00
Ali Can Karaca, M.
Kemal Güllü
DETECTION OF BURNT REGIONS IN FOREST
FIRES FOR MULTISPECTRAL IMAGES USING
WAVELET TRANSFORM AND SUPERPIXEL
SEGMENTATION
48
IWW 2019, 18-20 OCTOBER 2019
6 7
Nurgül Özmen Süzme, Gür
Emre Güraksın
THRESHOLD DETERMINATION OF CANNY
EDGE DETECTION ALGORITHM BASED ON
PARTICLE SWARM OPTIMIZATION
50
Sheme Hamitaj, A. Burak
İnner
A SURVEY OF ROS SECURITY 51
Batuhan Meşeci,
Abdulkadir Yapıcı, Murat
Acar, Rıza Boçoğlu,
Meltem Kurt Pehlivanoğlu,
and Nevcihan Duru
DETECTING BOTS ON TWITTER USING
MACHINE LEARNING TECHNIQUES
52
Firdaws Farukh,
Nevcihan Duru
AUTOMATIC DETECTION OF ELDERLY ON
TOUCH SCREEN INTERFACES USING TOUCH
GESTURE MEASUREMENTS
53
Zafer Erdogan U-NET ARCHITECTURE OPTIMIZATION FOR
OPTIC DISC SEGMENTATION
54
Hatice Citakoglu, Yusuf
Özeren, Ömer Coşkun
DROUGHT ESTIMATION OF SAKARYA BASIN
STATION WITH WAVELET MODEL‒ADAPTIVE
NEURO‒FUZZY INFERENCE SYSTEM
55
56
Süleyman Çetinkaya,
Ali Demir, Hülya Kodal
Sevindir
DETECTION OF PARKINSON’S DISEASE VIA
J48 ALGORITHM
57
Süleyman Çetinkaya,
Ali Demir, Hülya Kodal
Sevindir
PERFORMANCE EVALUATION OF ADABOOST
AND NAIVE BAYES ALGORITHMS FOR THE
DETECTION OF HEART ATTACK
58
Süleyman Çetinkaya, Ali
Demir
DETERMINATION OF ATRIAL PREMATURE
BEAT BY USING ADABOOST ALGORITHM AND
FRACTIONAL DERIVATIVE
59
Süleyman Çetinkaya, Ali
Demir
THE ANALYTIC SOLUTION OF INITIAL
BOUNDARY VALUE PROBLEM INCLUDING
SPACE-TIME FRACTIONAL DIFFUSION
EQUATION
60
Cüneyt Yazıcı, Ali Fuat
Yeniçerioğlu
STABILITY IN RETARDED FUNCTIONAL
EQUATIONS
61
Ali Fuat Yeniçerioğlu,
Cüneyt Yazıcı
ON THE BEHAVIOR OF THE SOLUTIONS FOR
FIRST ORDERDELAY INTEGRO-DIFFERENTIAL
EQUATIONS
62
Rashmi Bhardwaj RIVER WATER QUALITY ESTIMATION USING
WAVELET DECOMPOSED CONJUNCTED
ARTIFICIAL INTELLIGENCE
64
Yeşim Ahi, Korkmaz
Bellitürk, Hüseyin T.
Gültaş
WATER PRODUCTIVITY ASSESSMENT OF
WHEAT AND SUNFLOWER IN RAINFED AREAS
USING FUZZY LOGIC APPROACH
66
Vahit Feryad, ˙Ismail
Hakkı Çavdar
AKILLI ELEKTRİK ŞEBEKELERİ İÇİN
TÜRKIYE ELEKTRİKLİ EV ALETLERİ VERİ
SETİ (TEEAVS) İLE DERİN ÖĞRENME
KULLANARAK CİHAZ SINIFLANDIRILMASI
67
Funda Dökmen, Zafer
Aslan
EVALUATION OF SALINITY FOR GROUND
WATER RESOURCES BY USING WAVELET
MODELING
68
Funda Dökmen,
Nevcihan Duru
MODELING OF WATER QUALITY FOR
SALINITY PARAMETERS BY USING FUZZY
LOGIC SYSTEM
69
Emre Parlak, Çağdaş Özer,
Mustafa Takaoğlu
INCREASING THE EFFICIENCY OF EMG
SIGNALS BY USING MACHINE LEARNING
ALGORITHMS
70
Kemalettin Yıldız RECONSTRUCTION OF MANDIBLE USING
VIRTUAL SURGICAL PLANNING
71
Tolga Turan Dündar THE IMAGE PROCESSING, WAVELET AND
APPLICATIONS IN NEUROSURGERY
72
Ş. Sibel Menteş , Zafer
Aslan, Duygu Akyıl
DIAGNOSTIC ANALYSES and WIND SPEED
ESTIMATION
74
Emrah Tuncay Özdemir,
Zafer Aslan
INVESTIGATION OF THE TIDES OF
NEWCASTLE CITY IN AUSTRALIA
75
Gülsen Çimen,
Orhan Akbulut
A COMPARATIVE STUDY
ON DECOLORIZATION METHODS
76
IWW 2019, 18-20 OCTOBER 2019
8 9
IWW'2019: International Conference on
Image Processing, Wavelet and Applications
TECHNICAL PROGRAMME
18.10.2019 - Friday Hours MAIN HALL (Derbent Hotel)
09:30 – 10:00 Registration
10:00 - 12:00
OPENING SESSION and WELCOME SPEECH
Assoc. Prof. Dr. Funda DÖKMEN, Conference Chair, Kocaeli University
Prof. Dr. Zafer ASLAN, Conference Chair, Istanbul Aydın University
Prof. Dr. Sadettin HÜLAGÜ, Rector, Kocaeli University
INVITED TALKS Wavelets Constructed by Walsh Functions (Prof. Abul Hasan Siddiqi)
The Image processing, Wavelet and Applications in Neurosurgery (Tolga Turan DÜNDAR, M.D.)
Influences of the North Atlantic Sea surface temperature on the East Asian climate (Prof. Lin WANG)
12:00 - 13:00 LUNCH
14:00 - 16:00
HALL A (MARMARA) HALL B (KARTEPE)
IMAGE PROCEESING I
Session Chair: Hasan OCAK
MEDICAL APPLICATIONS
Session Chair: H. Metin ERTUNÇ
Tuba Kurban
Fusion of Multispectral Face Images Using Transform Based Methods
Kaplan Kaplan, Yasin Cuşkun and H. Metin Ertunç
Wavelet Decomposition based Feature Extraction of Brain MR Images
Gürkan Küçükyıldız and Hasan Ocak
Bottle Counting System for a Re-cycling Facility Based on Processing of
Depth Images
Zafer Erdoğan
U-Net Architecture Optimization for Optic Disc Segmentation
Gürkan Küçükyıldız and Hasan Ocak
Image Processing Based Pupil Tracking System for a Wheel Chair System
Ahmed Rashid, Osman Nuri and Ziad Abood
Assessment the Quality Medical Images enhancement
Ergün Can, Ali Can Karaca, Oğuzhan Urhan and Mehmet Kemal Güllü
Lossy Compression of Hyperspectral Images Using Wavelet Transform,
Principal Component Analysis and Tensor Decomposition
Suleyman Cetın kaya, Ali Demir and Hulya Kodal Sevindir
Detection of Parkinson’s Disease via J48 Algorithm
Ali Can Karaca and Mehmet Kemal Güllü
Detection of Burnt Regions in Forest Fires for Multispectral Images using
Wavelet Transform and Super pixel Segmentation
Suleyman Cetın kaya, Ali Demir and Hulya Kodal Sevindir
Performance Evaluation of Adaboost and Naive Bayes Algorithms for the
Detection of Heart Attack
Gülsen Çimen and Orhan Akbulut
A Comparative Study on Decolorization Methods
Suleyman Cetın kaya and Ali Demir
Determination of Atrial Premature Beat by Using Adaboost Algorithm and
Fractional Derivative
16:00 - 16:30 COFFEE BREAK
16:30 - 18:00
ARTIFICIAL INTELLIGENCE I
Session Chair: Zafer ASLAN
ENGINEERING APPLICATIONS
Session Chair: Bülent ORUÇ
Ali Al-Naser, Osman N. Uçan and Oguz Bayat
Pattern-Based Behavioral Biometric Authentication
using Artificial Neural Networks
Süleyman Cetın kaya and Ali Demir
The Analytic Solution of Initial Boundary Value Problem Including Space-Time
Fractional Diffusion Equation
Naama M. Athab, Osman N. Uçan, Oğuz Bayat and Saadaldeen Ahmed
Intelligent system for predicting behavior of electrical energy consumption
Suat Karakaya and Hasan Ocak
Design and Implementation of a Wheeled Mobile Robot Platform
Mohammed Mahmood, Osama Ashour, Oğuz Bayat, Osman N.
Uçan Design and implementation of web based for intermediate
online shop with laravel framework
Bülent Oruç and Seray Bozkurt
Determination of Subsurface Position of Buried Magnetic Sources Using 1-D
Complex Continuous Wavelet Transform
Batuhan Meşeci, Abdulkadir Yapıcı, Murat Acar, Rıza Boçoğlu,
Meltem Kurt Pehlivanoğlu and Nevcihan Duru
Detecting Bots on Twitter using machine Learning Techniques
Ethem Kelekçi, Tuğce Yaren and Selcuk Kizir
Model Based PID Controller Design of a DC Motor using Black
Box System Identification Method
Emine Merve Öztürk, Ayhan Küçükmanisa and Oğuzhan Urhan
Drowsiness Detection System based on Machine Learning using Eye State
Ethem Kelekçi and Selcuk Kizir
Tool Path Planning & Control of the 3-Axis CNC Machining with
High Accuracy Step Motor Pulse/Direction Command
18.10.2019 - Friday
POSTER PRESENTATION
13:00 - 18:00 Ş. Sibel Menteş, Zafer Aslan and Duygu Akyil
Diagnostic Analyses And Wind Speed Estimation
18:30 – 20:00 GALA DINNER
19.10.2019 - Saturday
Hours
09:30 – 10:00 Registration
10:00 - 12:00
HALL A (MARMARA) HALL B (KARTEPE)
ENVIRONMENT APPLICATIONS
Session Chair: Selahattin İNCECİK
ENGINEERING APPLICATIONS II
Session Chair: Hami ÖZ
Rashmi Bhardwaj
River Water Quality Estimation using Wavelet Decomposed
Conjuncted Artificial Intelligence
Nurgül Özmen Süzme and Gür Emre Güraksın
Threshold Determination of Canny Edge Detection Algorithm
Based on Particle Swarm Optimization
Mert Dedeoğlu and Levent Başayiğit
The Estimation of Wheat Yield During Growth Stages Using
Vegetation Indices on Sentinel-2 Images
Sheme Hamitaj and Burak İnner
A Survey of ROS Security
Hatice Citakoglu, Yusuf Ozeren and Omer Coskun Drought Estimation of Sakarya Basin Station with
Wavelet Model‒Adaptive Neuro‒Fuzzy Inference System
Cüneyt Yazıcı and Ali Fuat Yeniçerioğlu
Stability in Retarded Functional Equations
Yeşim Ahi, Korkmaz Bellitürk and Hüseyin Tevfik Gültaş Water Productivity Assessment of Wheat and Sunflower in
Rainfed Areas Using Fuzzy Logic Approach
Ali Fuat Yeniçerioğlu and Cüneyt Yazıcı
On the Behavior of the Solutions for First Order Delay Integro-
Differential Equations
Funda Dökmen and Zafer Aslan
Evaluation of Salinity for Ground Water Resources
by using Wavelet Modeling
İsmail M. Kundakcı, Kaplan Kaplan and H. Metin Ertunç
Fault Classification of Rolling Bearings with Time-Frequency Image
Analysis and Pre-trained CNN Architecture GoogleNet
Funda Dökmen and Nevcihan Duru
Modeling of Water Quality for Salinity Parameters
by using Fuzzy Logic System
Hale Yenginer and Erkan Dursun Efficiency Analysis of Shadow Effect on Photovoltaic Panels
by using Image Processing Techniques
12:00 – 13:00 LUNCH
13:00 - 14:45
ARTIFICIAL INTELLIGENCE II
Session Chair: Osman N. UÇAN
OTHER APPLICATIONS Session
Chair: Nevcihan DURU
Emre Parlak, Çağdaş Özer and Mustafa Takaoğlu
Increasing The Efficiency Of Emg Signals By Using Machine Learning
Algorithms
Ihsan Dogan
Virtual and Augmented Microneurosurgery
Mehmet Can Yavuz
How to count people in Ottoman Empire?
Kemalettin Yıldız
Reconstruction of Mandible Using Virtual Surgical Planning
Firdaws Farukh and Nevcihan Duru
Automatic detection of elderly on touch screen interfaces using touch
gesture measurements
Ruowen Yang, Jie Cao and Wei Huang
Cross wavelet analysis of the relationship between total
solar irradiance and sunspot number
Layth Hani Al-Samarai, Osman Nuri Uçan, Oğuz Bayat
An Analysis of Accessibility in Learning Management System in
The Context of Higher Education Institution
Aparna Vyas
Wavelets and Its applications
Özlem Evrim Gündoğdu and Nevcihan Duru
Approaches for Event-based Text Summarizing
Amirmohammad Abhary
How Image Processing Changed Mining Engineering?
Vahit Feryad and İsmail Hakkı Çavdar
Device Classification with Turkish Home Electronics Database (Teevas) for
Smart Electrical Networks Using Deep Learning
Emrah Tuncay Özdemir and Zafer Aslan
Investigation of the Tides of Newcastle City in Australia
15:00–18:00 SOCIAL PROGRAMME
SPECIAL SESSION
18:00 – 19:00
Bernd Richter
Model Output Statistics (MOS) Methods, Operational Applications and Verification
Pammy Manchanda
Joint Research Workshop
IWW 2019, 18-20 OCTOBER 2019
10 11
INVITED SPEAKERS
Prof. Rashmi BHARDWAJ, INDIA
Guru Gobind Singh Indraprastha Universiy
Dr. Tolga Turan DÜNDAR, TURKEY
İstanbul Bezmialem Vakıf University
Prof. Fahima NEKKA, CANADA
The Faculty of Pharmacy, Université de Montréal
Prof. Abul Hasan SIDDIQI, INDIA
Sharda University
Prof. Lin WANG, CHINA
Institute of Atmospheric Physics, Chinese Academy of Sciences
20.10.2019 - Sunday
Hours
SOCIAL PROGRAMME
10:00 Departure from Derbent Hotel
Trip to Ormanya Natural Park https://www.ormanya.com/
Trip to Historical Corridor of Izmit http://www.izmit.bel.tr/tarih_koridoru/
17:00 Return to hotel
IWW 2019, 18-20 OCTOBER 2019
12 13
PREFACE
The 10th International Conference on Image Processing, Wavelet and
Applications (IWW2019) will be held in Kocaeli, Turkey between October
18 - 20, 2019. IWW2019 will be organized by Kocaeli University with the co-
operations of İstanbul Aydın University, EURAS EURASIAN UNIVERSITIES
UNION and Sharda University
Wavelet analysis can provide a unique decomposition of time series into
signal and noise components. From the overall spectrum, independent spectra
can be extracted allowing the separate detection and monitoring of the world
problems. The participants will discuss major issues of importance for integrated
wavelet implementation, including development experiences worldwide wavelet
applications.
The wavelet methods provide avenues for proper understanding of real
world problems. These scientific meetings are intended to provide opportunity
for interaction of scientists, engineers, planners, social scientists, agricultural
scientists, medical scientists who are applying wavelets in their research.
Mathematical studies are often aimed to analyze and visualize real world
problems. Discussion on this theme during ICIAM2003 in Sydney, mooted the
idea of holding symposium on wavelets by Prof. Dr. Abul H. SIDDIQI.
Keeping in view importance of wavelet tools, İstanbul Commerce
University (Turkey) had organized the First International Workshop on
Applications of Wavelets to Real World Problems during 17-18 July, 2005.
The “INTERNATIONAL WORKSHOPS II and III: MINI SYMPOSIUM ON
APPLICATIONS OF WAVELETS TO REAL WORLD PROBLEMS: IWW07
and IWW08” were organized by İstanbul Aydın University with the co-operation
of The Abdus Salam International Centre for Theoretical Physics-Office of
External Activities (ICTP-OEA, Italy). “The Fourth International Workshop on
Applications of Wavelets to Real World Problems (IWW2009)” was organized
by Kocaeli University with the co-operation of ICTP-OEA in June 2009. A fairly
good number of active researchers of the field participated in its deliberations,
and the papers presented were published in the Proceeding of the Workshop.
Encouraged by the success of the series of Workshops, The Fifth International
IWW 2019, 18-20 OCTOBER 2019
14 15
Symposium on Wavelet Applications to World Problems (IWW2010) was held
in Istanbul Aydın University with the co-operation of ICTP-OEA, between June
7-8, 2010. The sixth International Image processing & Wavelet on real World
applications conference (IWW 2012) was organized by Boğaziçi University
with the co-operation of Istanbul Aydın University in Istanbul. Universitat
Politècnica de València in Valencia, Spain organized the seventh IWW2013. The
8th International conference on Image Processing, Wavelet and Applications
(IWW2016) was held in Istanbul, Turkey between September 22 - 24, 2016. IWW
2016 was organized by Marmara University with the co-operations of Istanbul
Aydın University and Sharda University. The ninth IWW2017 has been organized
by Kafkas University, Kars between 5 - 8 November 2017. In the international
symposium/conference, it was provided that Turkish scientist interacted with
scientists from different countries by gathering specialist researchers in this
field and it created opportunity that Turkish scientist cooperated with foreign
scientists. Joint study subjects were developed and young scientist reunited with
internationally famous scientists by providing opportunity to get in contact with
scientists from foreign countries for new projects in the conference.
We hope you will spend good time in Turkey and benefit from the high-level
presentations, focused debate and exchange of information during the IWW2019
Conference.
Assoc. Prof. Dr. Funda DÖKMEN, Kocaeli University, Chair
Prof. Dr. Zafer ASLAN, İstanbul Aydın University, Chair
On the behalf of Organizing Committee
CONFERENCE TOPICS
All the research areas of Numerical Analysis and Computational Mathematics
• Image Processing and Wavelet Application
• Signal Processing and Wavelet Application
• Applications of Wavelet to Real problems
• Numerical ODEs
• Numerical PDEs (Inc. BVPs)
• Scientific Computing and Algorithms
• Stochastic Differential Equations
• Approximation
• Numerical Linear Algebra
• Numerical Integral Equations
• Error Analysis and Interval Analysis
• Difference Equations and Recurrence Relations
• Numerical problems in Dynamical Systems
• Applications to the Sciences (Computational Physics, Computational
Statistics, Computational Chemistry, Computational Engineering etc.)
• Differential Algebraic Equations
• Numerical methods in Fourier analysis
All the research areas of Applied Sciences and Mathematics
• Mathematical Physics and Wavelets
• Computers simulations and Wavelets
• Mathematical Physics
• Mathematical Chemistry
• Mathematical Biology and Mathematical Medicine
• Optimization and Operational Research
• Theoretical Mechanics
• Discrete Applied Mathematics
• Statistics
IWW 2019, 18-20 OCTOBER 2019
16 17
• Probability
• Dynamical Systems
• Algorithms
• Experimental Mathematics
• Theoretical Computer Science
• Applied Analysis
• Scientific Analysis
• Environmental Analysis
• Wavelets in Environmental Problems
• Wavelets in Agricultural Problems
• Wavelets in Engineering Problems
• Mathematical Modelling including but not limited to:
• Mathematical modelling of engineering processes, manufacturing, and
industrial systems, heat transfer, fluid mechanics, CFD, andtransport
phenomena environmental modelling
• Solid mechanics and mechanics of metals
• Electromagnets and MHD
• Reliability modelling and system optimization
• Decision sciences in an industrial and manufacturing context
• Civil engineering systems and structures
• Mineral and energy resources
• Relevant software engineering issues associated with CAD and CAE
• Materials and metallurgical engineering,
• Mathematical modelling of social, behavioral and other sciences
• Decomposition and Reconstruction Algorithms, Subdivision Algorithms
• Continuous and Discrete Wavelet Transform
• Wavelets in Finance(Economics)
• Wavelets in Geophysics
• Wavelets in Medicine
• Time-frequency Localization
• Phase-Space Analysis
• Subband Coding
• Image Compression
• Real-Time Filtering
• Radar and Sonar Applications
• Transient Analysis
• Medical Imaging
• Multigrid Methods
• Frames
• Bifurcation and Singularity Theory
• Deterministic Chaos and Fractals
• Solution and Coherent Phenomena
• Formation of Pattern
• Evolution
• Complexity Theory and Neural Networks
• Analytical Approaches and Simulations for more Accurate Descriptions
• Predictions
• Experimental Observations and Applications of Nonlinear Phenomena in
Science and Engineering
• Theoretical and Applied aspects of Computational Geometry
• Control Theory and Automation
• Applied Algebra
• Quality Theory of Differential Equations
• Neural Networks
• Artificial Intelligence Applications
• Artificial Intelligence Applied to Energy Process
• Miscellaneous
IWW 2019, 18-20 OCTOBER 2019
18 19
SCIENTIFIC COMMITTEE
Honoree Chair
Saadettin Hülagü, Rector, Kocaeli University, Turkey
Honoree Co-Chair(s)
• Zafer Aslan, İstanbul Aydın University, Turkey
• Abul Hasan Siddiqi, Sharda University, India
Honoree Committee
• Mustafa Aydın, İstanbul Aydın University, Turkey
• Yadigar İzmirli, İstanbul Aydın University, Turkey
Chairs
• Funda Dökmen, Kocaeli Üniversity, Turkey
• Zafer Aslan, İstanbul Aydın University, Turkey
Co-Chair
• Cüneyt Yazıcı, Kocaeli University, Turkey
Organizing Committee
• Zafer Aslan, İstanbul Aydın University, Turkey
• Funda Dökmen, Kocaeli University, Turkey
• Nevcihan Duru, Kocaeli University, Turkey
• H. Metin Ertunç, Kocaeli University, Turkey
• Hasan Ocak, Kocaeli University, Turkey
• Cüneyt Yazıcı, Kocaeli University, Turkey
International Advisory Board
• Zafer Aslan, Istanbul Aydın University, Turkey
• Kenneth E. Barner, University of Delaware, USA
• Rashmi Bhardwaj, Guru Gobind Singh Indraprastha Universiy, India
• Carlo Cattani, Tuscia University, Italy
• Funda Dökmen, Kocaeli University, Turkey
• Yeliz Karaca, Massachusetts University, Medical School, USA
• Kenneth A. Loparo, Case Western Reserve University, USA
• Majoz Moonis, University of Massachusetts Medical School, USA
• Hami Öz, Kafkas University, Turkey
• Luis M. Sánchez Ruiz, Universidad Politécnica de Valencia (UPV), Spain
• Abul Hasan Siddiqi, Sharda University, India
• Osman Ucan, Altınbaş University, Turkey
Scientific Committee
• Yeşim Ahi, Ankara University, Turkey
• Orhan Akbulut, Kocaeli University, Turkey
• Zafer Aslan, Istanbul Aydın University, Turkey
• Bülent Oktay Akkoyunlu, Marmara University, Turkey
• Zelha Altınkaya, Yalova University, Turkey
• Faruk Aras, Kocaeli University, Turkey
• Nizamettin Aydın, Yıldız Technical University, Turkey
• Burak Barutçu, İstanbul Technical University, Turkey
• Oğuz Bayat, Altınbaş University, Turkey
• Zafer Bingül, Kocaeli University, Turkey
• Carlo Cattani, Tuscia University, Italy
• Ali Deniz, İstanbul Technical University, Turkey
• Funda Dökmen, Kocaeli University, Turkey
• Nevcihan Duru, Kocaeli University, Turkey
• Tolga Turan Dündar, İstanbul Bezmialem Vakıf University, Turkey
• Şeref Naci Engin, Yıldız Technical University, Turkey
IWW 2019, 18-20 OCTOBER 2019
20 21
• Gökhan Erdemir, İstanbul Sabahattin Zaim University, Turkey
• Mustafa Nizamettin Erduran, İstanbul Sabahattin Zaim University, Turkey
• H. Metin Ertunç, Kocaeli University, Turkey
• Enrico Feoli, University of Trieste, Italy
• Filippo Giorgi, ICTP, Italy
• Ali Güneş, İstanbul Aydın University, Turkey
• Hasan Heperkan, Istanbul Aydın University, Turkey
• Selahattin İncecik, İstanbul Technical University, Turkey
• Burak İnner, Kocaeli University, Turkey
• Serdar İplikçi, Pamukkale University, Turkey
• Yeliz Karaca, Massachusetts University, Medical School, USA
• Kasım Koçak, Istanbul Technical University, Turkey
• Fred Kucharski, ICTP, Trieste, Italy
• İsmail Küçük, İstanbul Sabahattin Zaim University, Turkey
• Kerem Küçük, Kocaeli University, Turkey
• Serdar Küçük, Kocaeli University, Turkey
• Pammy Machanda, Guru Nanak Dev University, India
• Derya Maktav, İstanbul Technical University, Turkey
• Waqas Masood, Comsats University, Pakistan
• Sibel Menteş, İstanbul Technical University, Turkey
• Majoz Moonis, University of Massachusetts Medical School, USA
• Hasan Ocak, Kocaeli University, Turkey
• Oya Oğuz, Haliç University, Turkey
• Sevinç İlhan Omurca, Kocaeli University, Turkey
• A. Halim Orta, Namık Kemal University, Turkey
• Halit Hami Öz, Gedik University, Turkey
• Oktay Özcan, Istanbul Aydın University, Turkey
• Haluk Özener, Boğaziçi University, Turkey
• Sandro M. Radicella, ICTP, Italy
• Mahir Rasulov, Baku State University, Turkey
• Mukhayo Rasulova, Physical-Technical Institute, Tashkent, Uzbekistan
• Luis M. Sánchez Ruiz, Universidad Politécnica de Valencia (UPV), Spain
• Ahmet Sayar, Kocaeli University, Turkey
• Hülya Kodal Sevindir, Kocaeli University, Turkey
• Tahir Shah, ICTP, Italy
• Katepalli R. Sreenivasan, New York University, USA
• Filiz Sunar, İstanbul Technical University, Turkey
• Ahmet Duran Şahin, İstanbul Technical University, Turkey
• Serhat Şeker, İstanbul Technical University, Turkey
• Orhan Şen, İstanbul Technical University, Turkey
• Ali Tangel, Kocaeli University, Turkey
• Mete Tayanç, Marmara University, Turkey
• Sezai Tokat, Pamukkale University, Turkey
• Ahmet Tokgözlü, Süleyman Demirel University, Isparta
• Sema Topçu, İstanbul Technical University, Turkey
• Osman Nuri Ucan, Altınbaş University, Turkey
• Oğuzhan Urhan, Kocaeli University, Turkey
• Yurdanur Ünal, İstanbul Technical University, Turkey
• Satılmış Ürgün, Kocaeli University, Turkey
• Lin Wang, Institute of Atmospheric Physics,
Chinese Academy of Sciences, China
• Aydın Tarık Zengin, İstanbul Sabahattin Zaim University, Turkey
• Metin Zontul, Arel University, Turkey
IWW 2019, 18-20 OCTOBER 2019
22 23
Steering Committee
Fulya Akdeniz, Kocaeli University, Turkey
Samet Bayram, University of Delaware, USA
Ömür Can, Kocaeli University, Turkey
Süleyman Çetinkaya, Kocaeli University, Turkey
Deniz Demirhan, İstanbul Technical University, Turkey
Hüseyin Emre Güner, Kocaeli University, Turkey
Burak İnner, Kocaeli University, Turkey
Kaplan Kaplan, Kocaeli University, Turkey
Suat Karakaya, Kocaeli University, Turkey
Hakan Kaygusuz, Altınbaş University, Turkey
Seda Kul, Kocaeli University, Turkey
Deniz Okçu, Boğaziçi University, Turkey
Güven Özdemir, İstanbul Aydın University, Turkey
Tuncay Özdemir, İstanbul Technical University, Turkey
Bahattin Sinsoysal, Beykent University, Turkey
Adil Tek, Boğaziçi University, Turkey
Vildan Yazıcı, Kocaeli University, Turkey
APPROXIMATION BY WAVELETS CONSTRUCTED
THROUGH WALSH FUNCTIONS WITH APPLICATIONS
A.H. Siddiqi;
President ISIAM, Sharda University India
E-mail: [email protected]
P. Manchanda, Secretary ISIAM, GNDU, Amritsar India.
E-mail: [email protected]
ABSTRACT
The concept of wavelets constructed through Walsh functions and their
Applications have been discussed in a recent book by Farkov, Manchanda and
Siddiqi (Springer Nature 2019). Wavelets have been used in diverse fields such
as image processing, analysis and prediction of occurrence of epileptic seizure,
analysis of metrological data and prediction and data mining etc. Our future plan
is to examine effect of replacing wavelets by new class of wavelets. We are also
trying to apply wavelet and fractal methods for predicting earthquake.
Keywords: Walsh functions, epileptic seizure, wavelet and fractals.
IWW 2019, 18-20 OCTOBER 2019
24 25
AN ANALYSIS OF ACCESSIBILITY IN LEARNING
MANAGEMENT SYSTEM IN THE CONTEXT OF
HIGHER EDUCATION INSTITUTION
Layth Hani Al-Samarai1, Osman Nuri Uçan2, Oğuz Bayat3
1Altınbaş Üniversitesi, [email protected]
2Altınbaş Üniversitesi, [email protected] 3Altınbaş Üniversitesi, [email protected]
ABSTRACT
The main outcomes of the research to define new stander as guidelines
for disability usage for LMS in higher and further education institutions, this
results enable website designer and programmer to support system disability by
enhance the system and analysis it for student to develop improved system that
enable disability student to use eLearning CMS, Our main objective to design
methodology or stander to indicate if accessible LMC is support disability users
to facilitates full interaction by all usersin this research first, we will explore
main disability function that include in current online eLearning LMS in higher
for two main famous eLearning system as Chamilo and Moodle. It also questions
the legal and moral.
Keywords: Moodle, eLearning, Accessibility, Chamilo.
DESIGN AND IMPLEMENTATION OF WEB BASED
FOR INTERMEDIATE ONLINE SHOP WITH LARAVEL
FRAMEWORK
1Mohammed Mahmood, 2Osama Ashour, 3Oğuz Bayat ,
4Osman Nuri Uçan
1Altınbaş Üniversitesi, [email protected]
2Cankaya Üniversitesi, [email protected] 3Altınbaş Üniversitesi, [email protected]
4Altınbaş Üniversitesi, [email protected]
ABSTRACT
Working with traditional methods to develop a web applications causing
large limitations and a lot of time consuming and with number of unexpected
errors. For this reason a new technology like my pattern frameworks found
by some companies to deal with such issues. In this research we presented a
design and implementation for web based application for e-commercial shop
and third-part to buy products from online shops.
We have used laravel framework to do so. As result of this research we can
find out the development was standardized and non-business logic relationships
automatically processed, there was much scalability so this gives us more
efficiency through the implementations.
Keywords: E-Commercial , Mvc Patterns , Php Framework , Laravel , Php
Programming.
IWW 2019, 18-20 OCTOBER 2019
26 27
INTELLIGENT SYSTEM FOR PREDICTING BEHAVIOR
OF ELECTRICAL ENERGY CONSUMPTION
Naama Methab Athab1, Osman Nuri Uçan2, Oğuz Bayat3 and
Saadaldeen Ahmed
1Altınbaş University, [email protected]
2Altınbaş University, [email protected]
3Altınbaş University, [email protected] 3Altınbaş University, [email protected]
ABSTRACT
This paper will explore the intelligent system that could predict the usage and
saving of electricity and it plays a major role in the smart home era, since can
provide benefits with regard to comfort, safety and energy savings to electricity
consumers. Many authors have already explored residence monitoring and
prediction systems, however, very few approached the residence detection for
predicting the energy consumption and prediction by using smart meter data.
In this work, it can be achieved by using solely electricity consumption data
and integrating it into an intelligent system. Also, we address the problem of
generalizing a classification model, i.e., we analyze the possibility of using a
single classification model to monitor residence in multiple households. We found
that a residence detection accuracy and predict the usage was possible by using a
generic classification model. Regarding residence prediction, we showed that it
is possible to predict residence in multiple households, by using solely electricity
consumption data. In addition to a higher energy efficiency, residence monitoring
also provides more safety to the consumers. If a high electricity consumption is
verified in periods that are not supposed, residence monitoring systems can be
used as an intruder’s detector, by sending alarms in real time to the smartphones
of the occupants. If we analyze residence at the room-level, residence monitoring
systems can also be used for health monitoring applications.
Keywords: Occupancy prediction, residence detection, electricity
consumption, smart meter, opportunistic sensing, intelligent system
PATTERN-BASED BEHAVIORAL BIOMETRIC
AUTHENTICATION USING ARTIFICIAL
NEURAL NETWORKS
Ali Saad Al-Naser1, Osman Nuri Uçan2, Oguz Bayat3
1Department of Computer Engineering, Altınbaş University, Istanbul-Turkey
2School of Engineering and Natural Sciences, Altınbaş
University, Istanbul-Turkey [email protected]
3School of Engineering and Natural Sciences, Altınbaş
University, Istanbul-Turkey [email protected]
ABSTRACT
Smart devices with touch-sensitive screens are being widely used by users
for different applications. Many of these applications require storing sensitive
and personal information, which increases the importance of using secure
authentication system to only allow legitimate users to access such information.
Biometric authentication has shown good security performance, but users have
expressed sincere concerns about their privacy when physical biometric data are
collected from them. Moreover, studies show that most of the users rely on pattern
secrets to authenticate to their devices. Thus, a new method that uses artificial neural
networks is proposed in this study, which collects behavioral biometric features
during the authentication process using the pattern-based method. Convolutional,
Gated Recurrent and Long- Short-Term Memory neural networks are evaluated
in this study. The results show that the use of the Gated Recurrent neural network
has the lowest error rate. This neural network has 6.91% False Acceptance and
False Rejection rates at threshold of 8.81, of the distance between the template and
current attempt descriptors, which has achieved Equal Error Rate.
Keywords: Artificial Neural Networks; Biometric Authentication; Graphical
Authentication; Smartphones Security.
IWW 2019, 18-20 OCTOBER 2019
28 29
INFLUENCES OF THE NORTH ATLANTIC SEA
SURFACE TEMPERATURE ON THE EAST ASIAN CLIMATE
Lin Wang,
Center for Monsoon System Research, Institute of Atmospheric Physics, Chinese Academy of
Sciences, Beijing, China, Email: [email protected]
ABSTRACT
In this talk, the influences of the North Atlantic sea surface temperature (SST)
on the East Asian climate will be discussed. Cross-wavelet analysis indicates
a close relationship between the North Atlantic SST and East Asian climate in
boreal winter and summer. The diagnostic analysis suggests that the anomalous
warm North Atlantic SST in boreal winter could change the SST gradient along
the Gulf Stream and thereby the low-level baroclinicity in the atmosphere. As a
result, the activity of synoptic high-frequency atmospheric eddies is enhanced,
which excite an anomalous anticyclone above the warm SST via the vorticity
forcing of the eddies. The downstream dispersion of the anomalous anticyclone
forms a Rossby wave train that reaches East Asia, leading to cold winter over
East Asia. The warm North Atlantic SST anomalies could persist from winter to
summer, and it could lead to enhanced precipitation and prolonged rainy season
in the lower reaches of Yangtze River in summer. The mechanism also involves
a Rossby wave train emanating from the North Atlantic.
Keywords: North Atlantic, sea surface temperature, Rossby wave, transient
eddy, wavelet analysis
IWW 2019, 18-20 OCTOBER 2019
30 31
THE ESTIMATION OF WHEAT YIELD DURING
GROWTH STAGES USING VEGETATION INDICES ON
SENTINEL-2 IMAGES
Mert Dedeoglu1, Levent Başayigit2
1 Selçuk University, Agriculture Faculty, Department of Soil Science and Plant Nutrition,
Konya, Turkey, [email protected]
2 Applied Sciences University of Isparta, Agriculture Faculty, Department of Soil Science and
Plant Nutrition, Isparta, Turkey, [email protected]
ABSTRACT
Remotely sensed data has been identified as an important tool for crop yield
prediction especially for wheat which has a strategic importance. The easiest
method to monitor the crop yield during growth stages is vegetation indices
obtained from satellite images. The vegetation indices applications based on
chlorophyll content are useful spectral reflectance approaches. In this study,
NDVIred&RE and RE-OSAVI indices were used to estimate the wheat yield in
the growth stages. The research were conduct in Muğla–Dalaman State Farm.
Multitemporal images of SENTINEL-2 were used. Five years average product
yields from 2014 to 2018 were compared to index values calculated from the
satellite imagery of April- May- June 2018. For tillering periods, RE-OSAVI with
r2=0.681 (RMSE = 0.71) were made stronger predictions than NDVIred&RE.
For heading and ripening stage, NDVIred&RE were supplied higher accuracy
coefficients with r2 of 0.718 (RMSE = 0.64) and r2 of 0.745 (RMSE = 0.69)
respectively. As result, NDVIred&RE and RE-OSAVI indices calculated from
SENTINEL 2 satellite images in the early and late growth periods for wheat can
be used for the estimation of yield and the offering a suggestion in the planning
and management studies.
Keywords: Remote sensing, SENTINEL-2, vegetation index, wheat
ASSESSMENT THE QUALITY MEDICAL IMAGES
ENHANCEMENT
Osman Nuri Uçan1, Ziad M. Abood2.Ahmed R. Ibrahim3
1 Altınbaş University, School of Engineering and Natural Sciences, Istanbul, Turkey,
2 Mustansiriyah University, Baghdad- Iraq,
ABSTRACT
Algorithms are used to improve medical images (angina) and white and black
medical images in the Matlab method using image enhancement and waveform
image conversion. The goal of improving images is to improve the ability to
interpret and perceive information in images for human viewers, or to provide a
better introduction to other image processing technology. Image Enhancement (IE)
converts the image to provide better representation for high resolution details. Are
indispensable tools for the researcher of many groups in fields including medical
images, forensics and technical studies. Appropriate application: An appropriate
technology to solve one problem may be inappropriate for other problems. For
example, the use of images or videos is one way to solve low visibility and motion
problems and then medical imaging is used in a lot of contrast and sharpness.
In response to the growing demand for digital photography, companies have
released software, which is a commercial program for users who do not want to
edit the image and then improve it with the eyes.
Keywords: Contrast filter, Image enhancement , Type of noise , Type of
wavelet transform.
IWW 2019, 18-20 OCTOBER 2019
32 33
VIRTUAL AND AUGMENTED MICRONEUROSURGERY
İhsan Doğan, M.D.
Ankara University, School of medicine, Department of Neurosurgery
ABSTRACT
Neurosurgeryisanimportantanddelicatebranchofmedicinethatrequiresevery
detailed anatomy in the preoperative period. 3D-printings, augmented and virtual
reality technologies can easily be adapted to the neurosurgical practice. Here,
our aim is to adapt these 3D-printing models to neurosurgical practice for better
planning the individual neurosurgical treatment strategies. The thin slices
of MRI and CT angiography images of 50 patients were included in our
study. These images were processed using Osirix software. STL files of these
reconstructed images were created digitally. Additionally, these images were
uploaded to Sketchfab application in order to see detailed cerebral anatomy of
the patients. These digital models were processed and created, visualized without
any technical problem. This method was also free and practical. These models and
method allow neurosurgeons to understand the complex anatomy of the cerebral
structures. It is also possible to simulate the surgery. These models and methods
are the practical tools for understanding the cerebral anatomy And planning the
neurosurgical interventions.
Keywords: Virtual reality, augmented reality, microneurosurgery, cerebral
anatomy
FUSION OF MULTISPECTRAL FACE IMAGES USING
TRANSFORM BASED METHODS
Tuba Kurban
Erciyes University, Department of Geomatics Engineering, Kayseri, Turkey
ABSTRACT
Due to the developments in sensor technology, enhanced sensors have been
used in many areas. Inspection of an object (or scene) using more than one sensor
capturing images at the different wavelengths of the spectrum, results acquiring
much more valuable information from the object (or scene). Evaluation of the
data becomes more complex while the number of spectral bands are increased,
hence the idea of fusing images obtained at different wavelengths is emerged.
The aim of multispectral image fusion is the combination of the information
existed in different bands to enhance the complementary features. Fused image
obtained by combining the images captured at two or more bands, becomes more
useful for many applications such as face recognition.
In this paper, face images obtained from LHFD (long distance heterogeneous
face) database are fused with discrete wavelet transform (DWT), Laplacian
pyramid (LP), discrete cosine transform (DCT) and cross bilateral filter (CBF)
methods. Results are compared with edge quality (QE), spatial frequency (SF),
fusion factor (FF) and variance weighted structural similarity measure (Qy)
metrics.
Keywords: Image fusion, discrete wavelet transform, image quality metrics
IWW 2019, 18-20 OCTOBER 2019
34 35
WAVELET DECOMPOSITION BASED FEATURE
EXTRACTION OF BRAIN MR IMAGES
Kaplan Kaplan1, Yasin Cuşkun1, H. Metin Ertunç1
1Kocaeli University, Mechatronics Engineering, 413800, Turkey kaplan.kaplan@kocaeli.
edu.tr, [email protected], [email protected]
ABSTRACT
Early diagnosis of brain tumors plays an important role in improving the
treatment possibilities and increases the recovery rate of patients. Manual
classification of large amounts of MRI produced from brain tumors for cancer
diagnosis is a difficult and time-consuming task for radiologists. The automatic
classification of brain tumor images with the help of a model facilitates the work
of radiologists. Magnetic resonance imaging (MRI) is the most widely used
medical imaging technique for brain tumor diagnosis and treatment planning.
In this study, Wavelet transformation based feature extraction was performed in
order to classify the degree of brain tumor using MR images. A benchmark data
set, the BRATS’15 data set, was used in the study. The proposed method consists
of 2 steps, (1) wavelet decomposition, (2) classification sections. High - grade
tumors (HGT) and low - grade tumors (LGT) were classified successfully.
Keywords: Brain tumor detection, wavelet coefficient, feature extraction,
brain tumor grade, classification.
IWW 2019, 18-20 OCTOBER 2019
36 37
DETERMINATION OF SUBSURFACE POSITION OF
BURIED MAGNETIC SOURCES USING 1-D COMPLEX
CONTINUES WAVELET TRANSFORM
Bülent Oruç1, Seray Bozkurt1
1Kocaeli University Department of Geophysical Engineering, 41380, İzmit, Turkey
ABSTRACT
In this study, Complex Continuous Wavelet Transform (CCWT) is used to
estimate the horizontal location and depth to the top of anomalous sources by
transferring the magnetic data into an auxiliary space. Using CCWT to analyze
magnetic data allows us to estimate the subsurface position of buried anomalous
sources at different depth levels in scale space domain. The modulus maxima
of wavelet coefficients of magnetic anomalies exhibit cone like structures
for isolated and extended sources. The modulus maxima lines or ridge of the
CCWT coefficients are thus used to find location and mean depths to the top
of anomalous sources. The technique has been tested on theoretical examples
and applied to magnetic anomalies from Pishabo Lake, Ontario-Canada, Western
Gansu Province, China, and Bayburt-Turkey. The horizontal location and mean
depths of dike-like ore bodies have been estimated from the modulus maxima
lines of the CCWT coefficients and the efficiency of the technique is demonstrated
through comparisons with other methods.
Keywords: Complex Continuous Wavelet Transform
OLAY TABANLI METİN ÖZETLEME YAKLAŞIMLARI
Özlem Evrim Gündoğdu1, Nevcihan Duru
2
1Kocaeli Üniversitesi,Mühendislik Fak., Blgisayar Müh.Böl., [email protected]
2Kocaeli Üniversitesi, Mühendislik Fak., Müh.Böl., [email protected]
ABSTRACT
Gelişen bilgi teknolojileri çerçevesinde ortaya çıkan ihtiyaçlar insanoğlunu sürekli değişime,
bir kademe daha üstünü araştırmaya ve düşünmeye yönlendirmektedir. Hızlı bilgi akışı ve
güncellemeleri beraberinde donanım ve kısa zamanda çözüme ulaşma ihtiyacı gözönünde
bulundurulduğunda otomatik metin özetleme kavramı gelişmeye son derece açık bir konu olarak
karşımıza çıkmaktadır. Bu çalışmada, bahsettiğimiz akış içerisinde ortaya çıkan konulardan biri olan
otomatik metin özetleme ve olay tabanlı özetleme yönteminden bahsedilecektir. Metin içerisindeki
nedensellik bilgisini inceleyen olay tabanlı (event-based )metin özetleme de amaç metindeki
olayların bulunup; olayla ilgili olduğu tespit edilen terimlerin kümelenmesi ve kümelenmeden
uygun parçalar seçilerek özete eklenmesidir. Çalışmamızda olay tabanlı metin özetleme
yönteminden ayrıntılı olarak bahsedilecek olup; Türkçe metin özetlemede kullanılan yöntemler ve
uygulamaları, olay tabanlı metin özetleme yönteminin Türkçe metinlerde kullanılablirliği ve
istatistiksel veriler irdelenip ele alınmıştır.
Keywords: Doğal dil işleme, metin özetleme, olay tabanlı metin özetleme.
IWW 2019, 18-20 OCTOBER 2019
38 39
FAULT CLASSIFICATION OF ROLLING BEARINGS
WITH TIME-FREQUENCY IMAGE ANALYSIS AND
PRE-TRAINED CNN ARCITECHTURE GOOGLENET
İsmail Malik Kundakcı1, Kaplan Kaplan2, Hüseyin Metin Ertunç3
1Mechatronics Engineering Department, Kocaeli University, [email protected]
2 Mechatronics Engineering Department, Kocaeli University, [email protected]
3 Mechatronics Engineering Department, Kocaeli University, [email protected]
ABSTRACT
The machines used in automation systems, the drive is realised by a large
amount of rotation force. Bearings are machine elements that operate on rolling
basis, which are widely used in motor systems that perform this rotational motion.
This study aims to classify bearing errors as baseline, inner race and author
race. There are a lot of proposed methods in the literature for this topic. Unlike,
in this study we proposed GoogleNet Convolutional Neural Networks (CNN)
structure. GoogleNet is a pre-trained CNN structure. The winner of ILSVRC
2014 has been this network. Avarage pooling layers are used to significantly
reduce the number of parameters in the network.
CNN is a very powerful algorithm for image classification. The popular
applications like Instagram, Facebook etc. use this structure for face recognition.
In order to take advantage of this, the vibrational data has been translated into an
image called “Scalogram” by wavelet transformation.
Firstly, 100 healthy, 100 inner ring faults and 100 outer ring faulty bearings is
converted to image. 60% of these images are used for learning and the others are
test. As a result, accuracy is 97,33%.
Keywords: Rolling Bearings, CNN, Wavelet, Fault Classification
MODEL BASED PID CONTROLLER DESIGN OF
A DC MOTOR USING BLACK BOX SYSTEM
IDENTIFICATION METHOD
Ethem Kelekci1, Tugce Yaren1, Selcuk Kizir1
1Department of Mechatronics Engineering, Kocaeli University, Kocaeli, Turkey,
[email protected], [email protected], [email protected]
ABSTRACT
This study presents experimental results of a root locus based PID controller
design and implementation of a DC motor. Black box system identification
method is used to obtain mathematical model of the system. Hardware in the loop
setup is developed using STM32F4 discovery kit and rapid control prototyping
approach. The discrete time system model is analyzed with Root locus method,
and PID gains are determined under overshoot and settling time control goals.
Position control and trajectory tracking performance of the system were given
both on the simulation environment and real time experimental setup. Proposed
approaches in this study allows researchers to develop low cost control devices
and its easy implementation.
Keywords: Root Locus, Black Box, Rapid Control, System Identification, DC
Motor
IWW 2019, 18-20 OCTOBER 2019
40 41
TOOL PATH PLANNING & CONTROL OF THE 3-AXIS
CNC MACHINING WITH HIGH ACCURACY STEP
MOTOR PULSE/DIRECTION COMMAND
Ethem Kelekci1, Selcuk Kizir1
1Department of Mechatronics Engineering, Kocaeli University, Kocaeli, Turkey,
[email protected], [email protected]
ABSTRACT
Most common form of tool path in Computer Numerical Control (CNC)
machining are G01 (linear) and G02/G03 (circular) codes generated by Computer
Aided Manufacturing (CAM) system. In this study, tool path planning of the
G01, G02 and G03 codes and control of the 3-axis CNC machining with step
motor pulse/direction command is presented. Also, this study proposes a novel
method for generating of the pulse-direction command at a precision of 1 μm. The
tool path is planned based on the mathematical function of the liner and circular
CNC codes and synchronous motion of the 3-axes is provided on the MATLAB
environment. After this process, continuous time axis motions are transformed to
pulse/direction command at a precision of 1 μm for each axis. In order to observe
the effectiveness and robustness of the proposed method, several experiments
are conducted on the 3-axis CNC machining using STM32F4 discovery kit. The
generated pulse/direction command from planned tool path shows high accuracy
control performance on the CNC.
Keywords: CNC, G Codes, Pulse/Direction, Step Motor, STM32F4 Kit
DESIGN AND IMPLEMENTATION OF A WHEELED
MOBILE ROBOT PLATFORM
Suat Karakaya, Hasan Ocak
suat.karakaya, hocak { @kocaeli.edu.tr }
Mechatronics Engineering Department, Kocaeli University
ABSTRACT
In this study, design of an indoor mobile robot platform (MAVI-BOT) is
presented. The platform consists of a chasis made of aluminum, a 2D light
detection and ranging sensor (LIDAR), an inertial measurement unit (IMU), an
infra-red camera and a LED projection and an industrial PC. The platform, based
on differential drive model, has 2-wheel drive. The active wheels are pneumatic
and the third wheel is positioned as a rubber-filled caster. A study was also
carried out to identify the actuators that met the given dynamic criteria. Realtime
applications made on the developed wheeed mobile robot (WMR) are discussed.
Keywords: Mobile robot, mechanical design, actuator
IWW 2019, 18-20 OCTOBER 2019
42 43
HOW IMAGE PROCESSING CHANGED
MINING ENGINEERING?
Amirmohammad Abhary
Amirkabir University of Technology- Department of Mining and Metallurgical Engineering,
ABSTRACT
From the prehistory to the present day, the mine has played a prominent role
in the sustainability of humanity. From the beginning of civilization, people
used ceramics and stone, which later used surface metals and near the ground.
All of this was used as a weapon of life. Mining engineering is a collection of
sciences, methods and techniques that begin with the exploration of a mine
and continues its processing. Mining engineers today must be equipped with
numerous knowledge and techniques. General geology, structural geology,
economic geology, rock mechanics, rock dynamics, mineral extraction,
descriptive mineralogy, lithology, geological survey, cartography and the
application of aerial photographs, optical mineralogy, mineralography, mineral
analysis, exploratory geophysics, exploratory drilling, well logging,
exploration geochemistry, geotechnics, drilling and blasting, crushing,
flotation, hydrometallurgy, engineering geology, paving and etc. are some
few example of hundreds different mining sciences and techniques. Mining
engineering can be very efficient in the field of executive and technical
affairs, in terms of specificity and sharing with a few other fields such as oil,
mechanics, civil, environmental, chemistry, metallurgy, and electrical of
engineering. In general, mining engineering has a very important role to play in
changing the economic and social face of each region and its economic
prosperity. Image has been the first means of communication between
humans. Man knew thousands of years ago that he could transcend many of
the concepts with the help of the image. Certainly mining engineering is no
exception image processing
has played a major role in the development and advancement of excellence in
mining engineering. Whether they want to know what they want and what
they do not know, engineers are surrounded by image processing.
Unfortunately, due to the lack of understanding of this area among the
professors and consequently educated students and researchers, the huge
capacity of image processing has not yet come as it should and may be not me
included in the field of mining engineering. However, it can be said with
certainty that there are no small and large mineral projects in different
trends in which at least part of the work is based on image processing and
the lack of processing image processing not only leads to a reduction in the
quality, but also some of the failure of the project. In this paper, we try to
highlight the importance of this topic with brief references to some of the
capabilities of image processing in various trends related to mining
engineering, and draw the attention of mining engineers as well as image
processing professionals to the importance of interdisciplinary work in this
regard, so that we will see an increase in efficiency in these areas.
Keywords: Mining Engineering, Image Processing, Exploration, Remote
Sensing, PCA
IWW 2019, 18-20 OCTOBER 2019
44 45
BOTTLE COUNTING SYSTEM FOR A RE-CYCLING
FACILITY BASED ON PROCESSING OF DEPTH IMAGES
Gürkan Küçükyıldız1, Hasan Ocak2
1 Usak University [email protected]
2Kocaeli University, [email protected]
ABSTRACT
In this study, a bottle counting system is developed based on processing of
depth images. 3D depth images are captured by Kinect-2 sensor, which was
mounted at a facility recycling line. The Kinect 2, developed by Microsoft, has
an infrared time of flight depth camera with 512*424 resolution. Compared to
structured light based depth cameras, the disturbances (light variations, shadows,
bottle impureness) does not affect the sensor measurements. Based on the facility’s
demand, three different types of bottles were considered in the bottle counting
system. Boxes contain either 20 or 12 bottles depending on the bottle type. The
facility recycling line moves at 1 m/s velocity and at least one box crosses from
facility line per second. Consequently, more than 40000 bottles are recycled in
the facility per day. The bottle counting system consists of an image processing
algorithm for processing captured depth images. The first step of the algorithm
includes camera calibration routine so as to convert sensor measurements from
camera frame to a fixed world frame. The camera intrinsic matrix is provided by
the supplier independently for each manufactured Kinect-2 sensor. Therefore,
the calibration routine includes steps to obtain the camera extrinsic matrix only
for fixed camera and world frames. In addition to providing the camera extrinsic
matrix, the calibration routine also computes the distance from the camera to
conveyer band plane, which is referred to as plate depth. Depth images are first
converted to world frame using the camera intrinsic and extrinsic matrices. If a
pixel is elevated (decreased depth value) by a certain threshold and projects on
the conveyer belt, it is considered to be a foreground pixel. All other pixels are
assumed to be background pixels. The threshold is computed based on bottle
height. A detection line on the conveyer belt is used to check the scene for the
presence of a box. Elevated pixels on the detection line triggers the measurement
process. Once the measurement is triggered, morphological operations are applied
to the binary image to remove noise and close the gaps. Connected component
analysis is performed to detect objects in the image. The largest object is marked
as the box if its area is consistent with the boxes in the database. The corners of
the minimum bounding rectangle are computed in the next step. Orientation of
the box is determined based on the corners of the minimum bounding rectangle.
Box pixels along with the box corner coordinates are projected to conveyer belt
surface. The projected image is then rotated to achieve zero orientation for the box.
Given the box and bottle type, the algorithm computes the locations within the
box where bottles are expected to be. If the number of pixels around a computed
bottle location is higher than a pre-determined threshold, the algorithm assumes
that a bottle is present at that location. The computational cost of algorithm is
an important parameter in real time operation. The cycle time for the developed
image processing scheme is 45 ms: 33 ms for image capturing (30 fps), 5 ms for
the pre-processing and 7ms for the bottle detection steps. It is observed from the
results that the developed algorithm can count the bottles with 99% accuracy for
each bottle type.
Keywords: Kinect, Depth image processing, Camera calibration, Bottle
counting
IWW 2019, 18-20 OCTOBER 2019
46 47
IMAGE PROCESSING BASED PUPIL TRACKING
SYSTEM FOR A WHEEL CHAIR SYSTEM
Gürkan Küçükyıldız1, Hasan Ocak2
1 Usak University [email protected]
2Kocaeli University, [email protected]
ABSTRACT
In this study, image processing and classification based pupil tracking system
is developed. Images are captured by a webcam mounted on a wheelchair.
Captured images are processed by a custom algorithm developed in MATLAB
environment. A low pass filter is first applied to captured images in order to
remove noise as a pre-processing step. Viola-Jones algorithm, which is an
embedded function (CascadeObjectDetector) in Computer Vision Toolbox of
MATLAB, is utilized for the face detection part of the algorithm. Similarly,
eyes are detected using the same function. Given the detected eye locations, the
algorithm then starts searching for eye pupils on the frame. The two eye sub-
images are binarized using an empirically determined threshold. Morphological
filters are applied to these binary images to remove isolated pixels. From each
image, the location of the pupil is detected by the Hough circle algorithm. The
circle which has the most number of black pixels, is finally selected as the eye
pupil circle.
After the detection of eye pupils’ locations, features are extracted in order
to train a Random Forest (RF) Model. The algorithm extracts the detected eye
pupil’s center coordinates as a feature. In addition to these center coordinates,
the ratio between black pixels and white pixels around the eye pupil is also
included in the feature vector. Three different classes were considered: Looking
left, looking right and looking ahead. It is observed from the results that the
developed algorithm detects corresponding classes with 95%, 98% and 99%
accuracy, respectively.
Keywords: Pupil Detection, Hough Circle Transform, Random Forest, Image
processing
MODEL OUTPUT STATISTICS (MOS) METHODS,
OPERATIONAL APPLICATIONS AND VERIFICATION
Bernd Richter ret.,
Deutscher Wetterdienst, Germany
ABSTRACT
Model Output Statistics (MOS) is a statistical post-processing procedure that has been
published by Glahn and Lowry (1972). It is based on a set* of multiple linear regression equations
between predictors and predictands. Predictors are variables from the direct model output (DMO)
forecast of a numerical weather prediction (NWP) model, predictands are the variables to be
predicted as final forecast. The multiple linear regression equations are developed on datasets
covering the recent few years of NWP model forecast fields and of recent observations. A special
feature of the MS-MOS (Meteo Service) presented here is a very sophisticated selection of specially
defined predictors in order to describe as much variance of the predictands as possible. Some
examples of these specially defined predictors are wind-components perpendicular and parallel to
the sea or to a mountain range, simple approximations describing the vorticity of the airflow, indices
for the vertical stability and other synoptical parametrizations. These specially defined predictors
showed to be very successful during the past twenty years since MS-MOS has been introduced into
the operational forecast section of Deutscher Wetterdienst. Verification results show that operational
forecaster can, in general, not improve the forecast quality of MS-MOS. MS-MOS is also in
operational use in other MET services, e.g. for spot forecasts in smartphone apps, which can, with
respect to numbers, not be accomplished by individual forecasters. As MOS is based on regression
equations with predictors of a specific NWP-model, it corrects for model errors of this NWP-model.
As a consequence, MOS has to be developed for each model. Another feature of MOS is the option
to predict all variables that have been observed, even if these variables are not element of DMO.
This feature is of special importance in aviation forecasts to predict visibility and ceiling. Probability
predictands are Important in aviation meteorology and can be forecast by MS-MOS, e.g. PROB
(VIS<1000m), (CIG<100ft), (Gust>25KT), (SH), (TS). MS-MOS is also in operational use to
predict non-meteorological parameters (depending on weather, though), e.g. underkeel water
clearance for large container vessels coming to Hamburg Harbour with a draught close to the depth
of the Elbe river. *set of equations: 15.000 stations worldwide, per station 200 predictands, lead
time increment of 3hrs up to 360 hrs, 4 seasons, 250 potential predictors (1-15 acrual predictors)
Keywords: Weather Forecasting, Numerical Weather Prediction (NWP) Post-Processing,
Model Output Statistics (MOS), Probability Forecasts,
IWW 2019, 18-20 OCTOBER 2019
48 49
LOSSY COMPRESSION OF HYPERSPECTRAL
IMAGES USING WAVELET TRANSFORM, PRINCIPAL COMPONENT ANALYSIS AND TENSOR
DECOMPOSITION
Ergün Can1, Ali Can Karaca2, Oğuzhan Urhan3, M.
Kemal Güllü
1 Piri Reis University, Department of Electrical and Electronics Eng.,
[email protected] 2 Kocaeli University, Department of Electronics and Communication Eng.,
[email protected] 3 Kocaeli University, Department of Electronics and Communication Eng.,
[email protected] 4 Kocaeli University, Department of Electronics and Communication Eng.,
ABSTRACT
Compression of hyperspectral image (HSI) has recently become a very
popular for remote sensing applications. During the compression of HSI,
different compression methods have been developed by making use of spectral
and spatial correlations. Recently, tensor decomposition has been used to improve
compression performance in the literature.
In this work, a hybrid compression method based on Discrete Wavelet
Transform (DWT) and nonnegative Tucker decomposition (NTD) is proposed
for HSI. In the first step, DWT of each band image is obtained. NTD algorithm
is used to decompose the wavelet domain sub-image into a core tensor and side
matrices. By this way, energy of the sub-images is efficiently compressed. To
improve the compression performance, Principal Component Analysis is applied
before DWT to perform uncorrelation. Obtained core tensor and side matrices
are encoded by making use of arithmetic encoding. We evaluate the effect of the
proposed method on real HSIs and also compare the results with the well-known
compression methods such as DWT+JPEG2000, 3D-SPECK, 3D-TARP and
JPEG2000. Experiments show that the proposed method provides better results
in terms of signal-to-noise ratio (SNR) and mean spectral angle (MSA) criteria.
Keywords: Discrete Wavelet Transform, Tucker decomposition, Hyperspectral
Image Compression, Principal Component Analysis
IWW 2019, 18-20 OCTOBER 2019
50 51
DETECTION OF BURNT REGIONS IN FOREST FIRES
FOR MULTISPECTRAL IMAGES USING WAVELET
TRANSFORM AND SUPERPIXEL SEGMENTATION
Ali Can Karaca1, M. Kemal GÜLLÜ2
1Kocaeli University, Department of Electronics and Communication Eng.,
2Kocaeli University, Department of Electronics and Communication Eng.,
ABSTRACT
According to the Ministry of Forestry and Water Management of Turkey,
11993 hectares are affected from 2411 forest fires just in 2017. Detection of burnt
regions is an important task for fast planning and land management. Although
there are some applications such as using cameras on the tower and wireless
systems, satellite systems are getting popular due to high spatial resolution. In
this work, a hybrid method that includes Discrete Wavelet Transform (DWT)
and superpixel segmentation steps is proposed for detection of burnt regions.
The proposed method presents a coarse to fine strategy for multispectral images.
Firstly, the normalized burn ratio (NBR) map is calculated using the multispectral
image. Afterwards, NBR map decomposed into wavelet coefficients using 4-level
DWT. Then superpixel segmentation is applied to its LL4 component and the
region is coarsely detected by spectral index. This process is repeated for LL2 and
original map. Finally, the burnt region is detected with details. The experiments
are performed on two multispectral images which are acquired by Sentinel-2A
satellite for the forest fire located in the İzmir-Menderes region (2017). From
experimental results, it is observed that the proposed method provides better
detection performance than other methods that use classical spectral indices.
Keywords: Forest Fire, Superpixel Segmentation, Wavelet Transform,
Sentinel-2A image
THRESHOLD DETERMINATION OF CANNY EDGE
DETECTION ALGORITHM BASED ON PARTICLE SWARM OPTIMIZATION
Nurgül Özmen Süzme1, Gür Emre Güraksın2
1Afyon Kocatepe Üniversitesi Biyomedikal Mühendisliği Bölümü,
[email protected] 2 Afyon Kocatepe Üniversitesi Biyomedikal Mühendisliği Bölümü,
ABSTRACT
Edge detection is one of the most important parts of image processing. Using
edge detection, the amount of data is significantly reduced. Thus by filtering out
the unwanted or insignificant information, a new set of features or image that
contains important information of the image can be presented. Proper detection
of the threshold value is critical for the performance of the edge detection. In this
study, the threshold values for Canny was determined using the Particle Swarm
Optimization (PSO) algorithm based on the image quality indices, Mean Squared
Error (MSE), Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index
Metrics (SSIM) and Correlation Coefficients (CorCoef). The threshold value
calculated by the PSO algorithm was accepted as the highest threshold value
and the lowest threshold value was accepted as half of the highest threshold
value. The results of the threshold values obtained by the proposed method were
compared with those of the threshold values determined by default for the Canny
Algorithm. As a result, when the visual evaluation of the image was made, a
quite successful result was obtained with this method.
Keywords: Image processing, Edge Detection, Particle Swarm Optimization
IWW 2019, 18-20 OCTOBER 2019
52 53
A SURVEY OF ROS SECURITY
Sheme Hamitaj1, A. Burak İnner2
1Kocaeli University [email protected]
2Kocaeli University [email protected]
ABSTRACT
Lately, the Robot Operating System (ROS) has gained more and more
popularity among robotic software. Its community has been increasingly
active and expanding further and farther. By using a straightforward
distributed framework ROS becomes very convenient on the data exchange,
but its communications could be easily interfered. Because robots have the
ability to directly interact with the outer world, their security issues must be
taken very seriously. This paper is a detailed research and an overview on
security of ROS 1.x, its vulnerabilities and reliability issues. Additionally, we
describe possible security solutions and recommend some of the best practices
to get past this concern.
Keywords: Robot Operating System, ROS, Cybersecurity, Robots, SROS.
DETECTING BOTS ON TWITTER USING MACHINE
LEARNING TECHNIQUES
Batuhan Meşeci, Abdulkadir Yapıcı, Murat Acar, Rıza Boçoğlu, Meltem
Kurt Pehlivanoğlu, Nevcihan Duru
Kocaeli University, Faculty of Engineering, Kocaeli, Turkey
ABSTRACT
Nowadays bot accounts increasing rapidly in social media. Also, their
structures are becoming more sophisticated and more complex day by day. Those
accounts have been influencing society by spamming commercials, propaganda
and inappropriate contents for many years. In this paper, we collected malicious
attributes of accounts by using the Twitter Application Programming Interface
(API). We used Decision Tree, Random Forest, Multi-Layer Perceptron (MLP),
k-Nearest Neighbor (kNN) and Naive Bayes algorithms to identify Turkish bot
accounts on Twitter. According to test results, the highest achieved F-measure is
92.70% obtained by using the MLP algorithm.
Keywords: Bot Detection, Social Media Data, Machine Learning, Twitter Bot
Accounts
IWW 2019, 18-20 OCTOBER 2019
54 55
AUTOMATIC DETECTION OF ELDERLY ON
TOUCH SCREEN INTERFACES USING TOUCH GESTURE MEASUREMENTS
Firdaws Faruk, Nevcihan Duru
a Computer Engineering, Kocaeli University, Izmit, 41001, Turkey
b Computer Engineering, Kocaeli University, Izmit, 41001, Turkey
* Firdaws Farukh Tel.: +90 552 2892747;
E-mail adresi: [email protected]
ABSTRACT
Nowadays, touch screen interfaces are widespread and used in everyday areas
of life ranging from personal smartphones, computers, tablets to public service
points like ATMs, restaurants, airports and even electronic voting systems.
With the prevalence of machine learning, many researchers have worked on
automatically detecting the demographics of a user for many reasons including
enhancing security, content filtering, customizing recommendations, anomaly
detection and improving usability. This paper follows the research done over the
years and reports on the verdicts attained. High accuracies have been reported
by most researchers thus proving that automatic detection of user demographics
based on different metrics is possible.
Keywords: Age group, Classification, Elderly, Touch screen
U-NET ARCHITECTURE OPTIMIZATION FOR OPTIC
DISC SEGMENTATION
Zafer Erdogan
Department of Computer Engineering Istanbul Technical University Istanbul, Turkey
ABSTRACT
Optic disc segmentation is essential for automated retinal analysis and further
diagnosis of retinal diseases such as glaucoma and diabetic retinopathy. These
diseases can cause irreversible damage to the optic disc and lead to visual
impairment or even blindness. Assisting technology for early detection of optic
disc related diseases can significantly decrease the incidence of blindness.
This work examines the optimization of a variant of fully convolutional neural
networks for optic disc segmentation, U-Net. The effect of intensity-based pre-
processing and the impact of optimization parameters are evaluated based on
validation loss. Publicly available REFUGE retinal image dataset is used in
testing the implemented approach. Results compared with a similar study in
optic disc segmentation to demonstrate the effectiveness of implementation.
Keywords: Deep learning, U-Net, optic disc, segmentation, image
processing
IWW 2019, 18-20 OCTOBER 2019
56 57
DROUGHT ESTIMATION OF SAKARYA BASIN
STATION WITH WAVELET MODEL‒ADAPTIVE NEURO‒FUZZY INFERENCE SYSTEM
Hatice Citakoğlu1, Yusuf Özeren2, Ömer Coşkun3
1Civil Engineering Department, Erciyes University, Turkey,
2General Directorate of State Hydraulic Works, Turkey,
3General Directorate of State Hydraulic Works, Turkey,
ABSTRACT
Drought indices in the literature allow for the determination of past drought.
Predicting the future drought is of great importance in terms of reducing the
negative effects of drought. Using drought index data from previous periods,
it will allow for the prediction of future drought and the establishment of early
warning systems. In this study, hydrological drought analysis was performed on
monthly flow data of Sakarya Basin 1234 and 1275 flow observation stations
between 1980‒2015. Flow Drought Index method was used to determine
drought. Adaptive neuro‒fuzzy inference system with Wavelet Model was used
to predict drought for the future. Drought index data between 1980‒2004 were
used as education and 2004‒2015 data were used as test data. Three different
models were developed at each time scale to estimate the drought index values
of 1, 3, 6 and 12 months. According to the determination coefficient, Adaptive
neuro‒fuzzy inference system models used together with wavelet analysis gave
more successful results.
Keywords: Flood, Current Drought Index, Wavelet Transform, Adaptive
neuro‒fuzzy inference system
IWW 2019, 18-20 OCTOBER 2019
58 59
DETECTION OF PARKINSON’S DISEASE VIA J48
ALGORITHM
Süleyman Çetinkaya1, Ali Demir2, Hülya Kodal Sevindir3
[email protected] [email protected]
ABSTRACT
Machine Learning is used in many areas around the world, including the
healthcare industry. It can play an essential role in predicting the presence/absence
of neurodegenerative disorder. If this information is well predicted, it may help
doctors to treat patients. Parkinson’s Disease is a worldwide health problem,
causing movement disorder and gait deficiencies. In this study, walking force
signals from the force sensitive receptors were used to separate one of the neuro-
degenerative Parkinson’s Disease (PD) from the normal individual. These signals
were first separated by the Daubechies 4 (db4) wavelet transform to the seventh
level. Then peak analysis was performed on the seventh level approximation
signal to find out the local maxima of the signal, the peak positions of these local
maxima, peak widths, and peaks. Then, two basic statistical characteristics were
obtained from each of these four peak features. So, a total of 16 attributes were
obtained, 8 for the left foot, and 8 for the right foot. After this stage, training data
was generated by using 50% of the data of each subject. In the next step, 87.1%
accuracy was obtained for determination of PD using the J48 algorithm.
Keywords: Parkinson’s Disease, J48 Algorithm, Wavelet Transform.
PERFORMANCE EVALUATION OF ADABOOST AND
NAIVE BAYES ALGORITHMS FOR THE DETECTION OF HEART ATTACK
Süleyman Çetinkaya1, Ali Demir2, Hülya Kodal Sevindir3
2 [email protected] 3 [email protected]
ABSTRACT
Machine Learning is used in many areas around the world, including the
healthcare industry. It can play an essential role in predicting the presence/absence
of heart diseases. If this information is well predicted, it may help doctors to
treat patients. One of the most important problems in the health sector is a heart
attack. In this study, we used Naive Bayes and Adaboost algorithms to determine
the heart attack. The data used in the study were taken from the UCI machine
learning repository. According to test results, the correct detection rate of Naive
Bayes algorithm is higher than the Adaboost algorithm to determine heart attack.
Keywords: Naïve Bayes, Adaboost, Heart Attack, Machine Learning.
IWW 2019, 18-20 OCTOBER 2019
60 61
DETERMINATION OF ATRIAL PREMATURE
BEAT BY USING ADABOOST ALGORITHM AND FRACTIONAL DERIVATIVE
Süleyman Çetinkaya1, Ali Demir2
ABSTRACT
Machine Learning is used in many areas around the world, including the
healthcare industry. It can play an essential role in predicting the presence/
absence of heart arrhythmias. If this information is well predicted, it may help
doctors to treat patients. One of the heart arrhythmias is atrial premature beat. In
this study, we used fractional derivative and Adaboost algorithm to determine
the atrial premature beat. The data used in the study were taken from MIT-BIH
arrhythmia database. According to test results, the correct detection rate of
Adaboost algorithm is 96.7 % to determine atrial premature beat.
Keywords: Adaboost algorithm, Fractional Derivative, Atrial Premature
Beat, ECG
THE ANALYTIC SOLUTION OF INITIAL BOUNDARY
VALUE PROBLEM INCLUDING SPACE-TIME FRACTIONAL DIFFUSION EQUATION
Süleyman Çetinkaya1, Ali Demir2
ABSTRACT
In this research, we determine the analytic solution of initial boundary value
problem with Dirichlet boundary conditions in one dimensional. By using
seperation of variables the solution is constructed in the form of a Fourier series
with respect to the eigenfunctions of a corresponding Sturm-Liouville eigenvalue
problem including fractional derivative in Caputo sense.
Keywords: Caputo fractional derivative, Time-space fractional diffusion
equation, Mittag-Leffler function, Initialboundary-value problems, Spectral
method
IWW 2019, 18-20 OCTOBER 2019
62 63
STABILITY IN RETARDED FUNCTIONAL
EQUATIONS
Cüneyt Yazıcı1, Ali Fuat Yeniçerioğlu2
Department of Mathematics, The Faculty of Education, Kocaeli University,
41380, Kocaeli, Turkey [email protected],
ABSTRACT
This study deals with the stability behavior of a scalar linear retarded
equation. Useful exponential estimations and stability criteria of the solutions
were established. Finally, two examples were given for the stability of the zero
solution of the retarded equation.
Keywords: Retarded equation, Characteristic equation, Stability, Trivial
solution.
ON THE BEHAVIOR OF THE SOLUTIONS FOR FIRST
ORDERDELAY INTEGRO-DIFFERENTIAL EQUATIONS
1Ali Fuat Yeniçerioğlu , 2Cüneyt Yazıcı
Department of Mathematics, The Faculty of Education, Kocaeli University,
41380, Kocaeli, Turkey
ABSTRACT
In this study, some results are given concerning the behavior of the solutions
for linear delay integro-differential equations. These results are obtained by the
use of two distinct real roots of the corresponding characteristic equation.
Keywords: Integro-differential equation, Stability, Delay, Real root.
IWW 2019, 18-20 OCTOBER 2019
64 65
RIVER WATER QUALITY ESTIMATION
USING WAVELET DECOMPOSED CONJUNCTED ARTIFICIAL INTELLIGENCE
Rashmi Bhardwaj
University School of Basic & Applied Sciences,
Non-Linear Dynamics Research Lab, Guru Gobind Singh Indraprastha University
(GGSIPU), Delhi, India.
E-mail: [email protected]
ABSTRACT
The survival of life on Earth is directly proportional to presence to water
among other important resources. Water remains to be a natural resource with
no replacement. Water is life and is the most precious resource on Earth. Earth
is covered with 70% of water, 2.5% is freshwater and 1% is easily accessible
freshwater; thus only 0.007% of Earth’s water is accessible. In today's era where
science and technology are growing every hour and innovating new technologies
and devices to make life easier and comfortable, but no artificial intelligence
could either replicate or replace the need for water on Earth. The present study
deals with the qualitative exploration of water quality components like potential-
of-Hydrogen (pH), chemical-oxygen-demand (COD); biochemical-oxygen-
demand (BOD); dissolved-oxygen (DO) of Yamuna River in India at different
sample sites. Various sample sites designated for highly reported pollutants using
artificial intelligence through Least Square Support Vector Regression (LSSVR)
and hybrid of Wavelet and LSSVR. It is observed that hybrid of Wavelet and
Least Square Support Vector Regression (W-LSSVR) predicted good quality
accurately among the two prototypes simulated on the basis of the simulation
errors i.e. root mean square error (RMSE); mean absolute error (MAE);
coefficient of determination (R2) and execution time for both prototypes. RMSE
values decrease overall on training and validating via W-LSSVR as compared
to LSSVR. It is observed that MAE values show a lesser decrease as it is in
RMSE; on an average MAE has lesser variability and R2 has a greater variability
as per simulations. The simulation is carried out to analyze the level of various
pollutants in the Yamuna River at different sites for the consideration of quality
of water. The observed pattern from the study may help for future prediction of
the quality of water parameters, so that it prohibits the further decay of water
quality which may prove to be lethal to the environment. These forecasts may be
helpful for the formulation of policies, planning and execution for protection of
environment and quality of water.
Keywords: Least Square Support Vector Regression, Wavelet Decomposition,
Water quality, Wavelet-LSSVR (WLSSVR), coefficient of determination
IWW 2019, 18-20 OCTOBER 2019
66 67
WATER PRODUCTIVITY ASSESSMENT
OF WHEAT AND SUNFLOWER IN RAINFED AREAS
USING FUZZY LOGIC APPROACH
Yeşim Ahi1,* Korkmaz Bellitürk2 Hüseyin T. Gültaş3
1 Water Management Institute, Ankara University, 06135 Ankara, TURKEY
2 Department of Soil Science and Plant Nutrition, Faculty of Agriculture, Tekirdag Namık
Kemal University, 59030 Tekirdağ, TURKEY 3 Biosystem Engineering Department, Agriculture and Natural Sciences Faculty, Bilecik Şeyh
Edebali University, 11230, Bilecik, TURKEY
*Corresponding author: [email protected] (Y. Ahi )
ABSTRACT
It is necessary to put forward to some solution proposals which provide
agricultural water efficiency as today that soil and water resources are gradually
decreasing in quality and quantity. The aim of this study is to evaluate the effects
of soil characteristics monitored in rainfed fields of wheat and sunflower on water
productivity (WP), find out the effects of these characteristics on soil productivity
using Fuzzy Logic and also, compare with each other. For this purpose, soil
fertility parameters (EC, pH, organic matter and lime) were determined from
soil samples taken from rainfed areas selected in Meriç-Ergene Basin of Turkey.
The fertility parameters change as following: pH 4.37-7.78, organic matter (OM)
0.47-2.51%, lime (CaCO3) 0-23.58%, electrical conductivity (EC) 0.18-5.92
dS m-1. In Fuzzy Logic model, fertility parameters, which directly effects soil
productivity, are considered as input variables, and productivity is considered as
output variable. While the average water productivity obtained during the pre-
October measurements from seven district was 0.83, 0.68, 0.87, 0.52, 0.79, 0.72
and 0.74 kg m-3 for wheat, they were found as 0.43, 0.43, 0.47, 0.37, 0.48, 0.42
and 0.42 kg m-3 for sunflower. According to the obtained results, it can be said
that Fuzzy Logic model which is used in many sciences should be useful in
agricultural field studies.
Key words: Soil productivity, evapotranspiration, water use effi , crop yield.
AKILLI ELEKTRİK ŞEBEKELERİ İÇİN TÜRKIYE
ELEKTRİKLİ EV ALETLERİ VERİ SETİ (TEEAVS) İLE DERİN ÖĞRENME KULLANARAK CİHAZ
SINIFLANDIRILMASI
Vahit Feryad1, ˙Ismail Hakkı Çavdar2
1Asis Otomasyon ve Akaryakıt Sistemleri [email protected]
2 Karadeniz Teknik Üniversitesi; [email protected]
ÖZET
Enerji ve sürdürülebilirlik sorunları, veri madenciliği ve makine öğrenmesi
yaklaşımları kullanılarak ele alınabilecek çok sayıda sorunu beraberinde
getirmektedir, Ancak, kamuya açık verilerin bulunmamasından dolayı bu tür
sorunların çözümü yavaş olmuştur. Bu çalışmada, çeşitli evlerden toplanan
ayrıntılı elektrik kullanım bilgilerini içeren ve enerji ayrıştırma (Disaggregation)
araştırmalarını ilerletmeyi amaçlayan bir veri seti olan Turkiye Elektrikli Ev Aletleri
Veri Seti (TEEAVS) sunulmuştur. Ayrıştırma konusundaki eski yaklaşımları ve
veri toplamadaki tasarım seçimleri nasıl etkilediklerini tartışarak veri toplama için
donanım ve yazılım kurulumları açıklınmıştır. Enerji yönetimi teknolojisi, talep
yönündeki yoğun yükler sırasında kullanıcıların enerji talebini azaltarak üretim
varlıklarının daha verimli kullanılmasına yardımcı olan akıllı şebekenin kilit bir
işlemidir. Akıllı şebeke ve akıllı ölçüm bağlamında, bu makale TEEAVS bilgisine
dayalı olarak ev aletlerini sınıflandırmasında müdahaleci olmayan (non-intrusive)
yük izlemesi (MOYİ) için Derin Öğrenme tekniği ile bir enerji ayrıştırma modeli
önerilmektedir. Ayrıca, uç kullanıcılara farkındalık kazandırmak ve pahalı akıllı priz
sensörlerine ihtiyaç duymadan talep tarafından ayrıntılı geri bildirim oluşturmak
için yüksek doğrulukta, tek boyutlu evrişimli derin sinir ağı (1D CNN) kullanan
denetimli (supervised) enerji ayrıştırma modeli tanıtılmıştır. 1D CNN ile TEEAVS
ortak kriter UK-DALE ve REDD veri setlerini kullanarak metriklere dayalı olarak
eğitimi hedef cihazlar üzerinde incelenmiştir. Önerilen model, Google Colab’ın
Tesla grafi işleme biriminde (GPU) Tensorflow kullanılarak eğitilmiştir.
Anahtar kelimeler: Enerji ayrıştırma, Derin Öğrenme, Müdahaleci olmayan
(non-intrusive) yük izlemesi, Tensorfl , Veri Seti.
IWW 2019, 18-20 OCTOBER 2019
68 69
EVALUATION OF SALINITY FOR GROUND WATER
RESOURCES BY USING WAVELET MODELING
Funda Dökmen1, Zafer Aslan2
1 Kocaeli University, The Vocational School of Food and Agricultural, Campus of Arslanbey,
41285, Arslanbey, Kartepe, Kocaeli, Türkiye,
[email protected] & [email protected] 2 Istanbul Aydın University, Faculty of Engineering, Department of Computer Engineering,
34295, Florya, Istanbul, Türkiye,
ABSTRACT
Approximately, 10 billion ha of agricultural lands/soils cannot be use due
to salinity every year because of wrong irrigation applications in the world.
Main elements (cations) are potassium (K+), sodium (Na+), calcium (Ca+2) and
magnesium (Mg+2) for salinity in dry and semi-dry climatic regions. Methods
of wrong irrigation and quality of irrigation water are so important for salinity
of agricultural soils/lands.In this study, ground water resources were analysed
in the laboratory conditions in terms of salinity parameters in Yalova Province
and its vicinity. Results of analyses were explained with variations and effects
based on quantity of salinity and also defining role of small, meso and large scale
factors by using wavelet model. And also, all salinity parameters were evaluated
by cluster test (model and evaluation on training test). Results of this study
would be helpful for estimation of salinity contents on soils by using irrigation of
groundwater at agricultural areas.
Keywords: Cations, groundwater, soil salinity, quality of irrigation water,
wavelet analyses.
MODELING OF WATER QUALITY FOR SALINITY
PARAMETERS BY USING FUZZY LOGIC SYSTEM
Funda Dökmen1, Nevcihan Duru2
1Kocaeli University, Food and Agricultural Vocational School, Campus of
Arslanbey, Kartepe-Kocaeli, Turkey 2Kocaeli University, Faculty of Engineering, Department of Computer,
Campus of Umuttepe, Kocaeli, Turkey
*Email: [email protected] & [email protected]
ABSTRACT
Water quality management has become an important issue in public policies
and agricultural activities throughout the world. Proper assessment of water
quality status in surface water resources based on different observations and also
analyses of chemical parameters. Various classification methods have been used
for estimating of usability of surface water resources.
This paper presents a comparative study, a methodology based on fuzzy
inference system to assess water quality is proposed. Using fuzzy logic system
was compared the output generated by fuzzy with that of conventional methods.
Three variables are employed for the quality assessment in terms of six different
surface water resources in Yalova region. Fuzzification unit will produce fuzzy
set, based on fuzzy variable namely magnesium (Mg+2), potassium (K+) and
sodium (Na+). Magnesium (Mg+2), potassium (K+) and sodium (Na+) values
were taken as input variables as they are factors which directly effect the usability
of agricultural activities and soil productivity. For this reason, usability was
taken as an output variable in return.
This study aims to determine the quality of water resources used in agricultural
areas selected in surface water resources of Yalova vicinity in Turkey and find
out the effects of these waters on productivity using Fuzzy Logic which have
recently been used in the world.
Keywords: Fuzzy inference system, monitoring, water quality assessment
IWW 2019, 18-20 OCTOBER 2019
70 71
INCREASING THE EFFICIENCY OF EMG SIGNALS
BY USING MACHINE LEARNING ALGORITHMS
Emre Parlak1, Çağdaş Özer2, Mustafa Takaoğlu3
1İstanbul Aydın University, [email protected] 2İstanbul Aydın University, [email protected]
3İstanbul Aydın University, [email protected]
ABSTRACT
In our study, long-term performance signals was measured; to minimize the
changes in the characteristics of the signals due to long term performance of
amputated arm patients, it is aimed to improve the read signals by using machine
learning algorithms. In our study, the data obtained from the measurements we
made through the Armband device of the right arms of seven people were used.
While the data were obtained, the hand was turned into a fist, and this movement
continued until fatigue occurred in the muscle. Naive Bayes, Generalized Linear
Model, Logistic Regression, Fast Large Margin, Deep Learning, Decision Tree,
Random Forest, and Gradient Boosted Trees algorithms are used to process
signals, and 16796 models are created. Data were analyzed based on Accuracy,
Classification Error, Area Under Curve, Precision, Recall, F Measure, Sensitivity,
Specificity. The algorithms that yield the best results were determined in each
variable, and the results were shared.
Keywords: EMG Signals, Machine Learning Algorithms, Signal Processing,
Signal Efficiency, MYO Armband.
RECONSTRUCTION OF MANDIBLE USING VIRTUAL
SURGICAL PLANNING
Kemalettin Yıldız
Bezmialem Vakif University, Dept. of Plastic Reconstructive & Aesthetic Surgery,
Istanbul, Turkey
ABSTRACT
Evolution in reconstructive microsurgery depends on the increase in demands,
advances in surgery and technology. We present mandible reconstruction using
virtual surgical planning and cutting-guides manufactured by 3D printers.
Traumatic or oncologic defects of mandible were reconstructed with free
vascularized fibula flaps using virtual surgical planning and cutting-guides.
The maxillofacial 3D computerized tomography scans of patients were taken.
DICOM data were transformed to STL format. Segmentation and virtual surgical
planning were performed with software. The cutting guides for osteotimies,
occlusal splint model, craniomaxillofacial model of patients designed virtually
and manufactured in 3D printers. We presented the application of virtual surgical
planning and custom made implants such as cutting guides, models, plates.
Implants were sterilized and used in vitro and intraoperatively. Precise and
delicate osteotomies have been performed. Also full bony contact in osteotomy
sites has been achieved. However, this technique needs a learning curve and the
cost is the main limitation of its use. Superior aesthetic and functional results can
be achieved with the use of this technique and implants.
Keywords: Microsurgery, Virtual Planning
IWW 2019, 18-20 OCTOBER 2019
72 73
THE IMAGE PROCESSING, WAVELET AND
APPLICATIONS IN NEUROSURGERY
Tolga Turan Dündar1
1 Bezm-i Alem Vakıf University, Faculty of Medicine, Department of Neurosurgery,
Fatih, Istanbul-TURKEY
E- mail : [email protected], [email protected]
ABSTRACT
The current aim of Neurosurgery is maximum neuro- functional protection ,
minimum tissue damage with individualized treatment. The Image Processing,
Wavelet and Applications are used for preoperative evaluation with several
imaging techniques such as magnetic resonans, computed tomography, and blood
diagnosis samples , perioperative asistant with several surgical and monitoring
equipment and postoperative several follow up techniques such as ICP,EEG and
EMG. Magnetic resonance imaging uses strong magnetic fields and radio waves
to produce detailed images of the inside of the part of body. Most of the human
body is made up of water molecules, which consist of hydrogen and oxygen atoms.
At the centre of each hydrogen atom is an even smaller particle called a proton.
Protons are like tiny magnets and are very sensitive to magnetic fields. Under the
powerful scanner magnets, the protons in cells line up in the same direction,
in the same way that a magnet can pull the needle of a compass. Short bursts
of radio waves are then sent to certain areas of the body, knocking the protons
out of alignment. When the radio waves are turned off, the protons realign. This
sends out radio signals, which are picked up by receivers. These signals provide
information about the exact location of the protons in the body.They also help to
distinguish between the various types of tissue in the body, because the protons
in different types of tissue realign at different speeds and produce distinct signals.
In the same way that millions of pixels on a computer screen can create complex
pictures, the signals from the millions of protons in the body are combined to
create a detailed image of the inside of the body. I think MRI as one of the main
issues for applications,software and image processing. Operation microscope and
endoscope systems are often used for minimal invasive surgery. İntraoperative
magnetic resonanse imaging , ultrasound and yellow green painting provide to
us real- time evaluation of brain tissue. Moreover , MEP,SSEP, free running
EMG,BAEP give knowledge about real time neuronal function. All of these
are software basis machine. In the near future, less tissue damage and maximum
benefit going to have been achieved with learning machines. We have to make
this process that individual , cheaper and accessible for every person in world
wide.
Keywords : Neurosurgery, hybrid surgery, learning machine
IWW 2019, 18-20 OCTOBER 2019
74 75
DIAGNOSTIC ANALYSES and WIND SPEED
ESTIMATION
Ş. Sibel Menteş1 , Zafer Aslan2, Duygu Akyıl1
1Istanbul Technical University, Department of Meteorological Engineering, Maslak 34469
Istanbul, TURKEY
[email protected], [email protected]
Istanbul Aydın University, Department of Computer Engineering, Florya, ISTANBUL
ABSTRACT
This paper covers time series analyses of wind speed data at three selected
areas in Turkey. Hourly wind speed observed in Aydın, Hatay and Mut in 2015
were analysed. Results of descriptive statistics and wavelet analyses with
MATLAB were discussed. In addition to temporal variations, local and large
scale factors and their role on wind speed variations were evaluated. In winter,
large scale factors are more effective on variation of hourly wind speed. In spring
and autumn, meso - scale factors play an important role. In summer period small
scale factors and their roles are more important than the other factors on wind
speed variations. In general, predictions show slightly decreasing trend on wind
speed in the last part of the year. This work was supported by the Scientific
and Technological Research Council of Turkey (TÜBİTAK) under the European
Union ERANET + Project titled “New European Wind Atlas (NEWA), Project
No: 215M386”.
Keyword: Wind speed, Wavelet, Time Series, Estimation.
INVESTIGATION OF THE TIDES OF NEWCASTLE
CITY IN AUSTRALIA
Emrah Tuncay Özdemir1, Zafer Aslan2
1Turkish State Meteorological Service, Atatürk Airport Meteorology Office, 34149, Yesilkoy,
Istanbul, Turkey.
2Istanbul Aydın University, Faculty of Engineering, Computer Engineering Department,
34392,
Florya, Istanbul, Turkey. [email protected]
ABSTRACT
Tides are caused by the gravitational force of the moon and the sun and earth’s
effects on the oceans and the seas due to the inertia (inertia attempts to keep the
water in place) of these gravitational forces. This tidal wave of the result of the
highest (crest) instead of the high tide, the lowest (trough) instead of low tide
is called. The height difference between low tide and high tide is called a tidal
range. In this study, the tides in Newcastle City, Australia in 2018-2019 were
investigated. Situated east of Newcastle City, the Pacific Ocean covers the entire
western part of the city. Daily tidal changes were investigated statistically in this
study. Every day at the ocean level, there are two high tides and two low tides.
According to the results, in 2018, the highest tide of 2.12 m occurred on July
13 and 14. Additionally, in 2019, the highest peak tide of 2.11 m occurred on
January 22. In addition, wavelet analysis was performed. In 2018, max and min
values are slightly higher than wave records in 2019. 1D Wavelet Packets and
Continuous Wavelets 1D show the role of small, meso and large scale factors on
the gravitational force of the moon and the sun.
Keywords: Newcastle, gravity, tides, high tide, low tide, tidal range
IWW 2019, 18-20 OCTOBER 2019
76 77
A COMPARATIVE STUDY
ON DECOLORIZATION METHODS
Gülsen Çimen1, Orhan Akbulut2
1Computer Engineering Department, Kocaeli University, [email protected]
2Computer Engineering Department, Kocaeli University, [email protected]
ABSTRACT
Color to gray conversion, which means decolorization, has an important
topic image processing and computer vision applications. In the literature, many
decolorization techniques are proposed to convert colored images into gray-
scaled format. The main objective of these methods is to preserve the contrast of
color images and especially perception of visual. In this paper, the performance
of the decolorization methods in terms of user-preferred results and objective
quality metrics are compared. The comparison is carried out by using color
contrast preserving ratio (CCPR) and color content fidelity ratio (CCFR) metrics
under different datasets such as COLOR250, Cadik. Experimental results on
decolorization can benefit future works.
Keywords: Decolorization, Color to Gray Conversion, Contrast Preserving.
AUTHOR(S)
A. ABHARY ..................42,43 H. ÇİTAKOĞLU ................ 56 O. COŞKUN .................... 56
A. B. NIAN ...................... 29 H. K. SEVİNDİR ...........58,59 O. N. UÇAN ...24,25,26,27,31
A. AL-NASER ................. 27 H. M. ERTUNÇ ............35, 38 O. URHAN ..............48,49,57
A. C. KARACA ........48,49,50 H. OCAK ..................41,44,46 P. MANCHANDA ........... 23
A. DEMIR............58,59,60,61 H. T. GÜLTAŞ ..................... 66 R. BHARDWAJ ................. 64
A. F. YENİÇERİOĞLU 62, 63 R. BOÇOĞLU ................. 53
A. H.SIDDIQI .................. 23 I. DOGAN ........................ 32
İ. H. ÇAVDAR .................. 67 S. AHMED ....................... 26
A. R. IBRAHIM. .............. 31 İ. M. KUNDAKCI ............ 38 S. BOZKURT .................... 36
K. BELLİTÜRK ............... 66 S. ÇETİNKAYA...58,59,60,61
A. YAPICI .......................... 53 K. KAPLAN ..................35,38 S. HAMITAJ ....................... 52
B. İNNER ......................... 52 S. KARAKAYA .................. 41
B. MEŞECI ...................... 53 K. YILDIZ ........................ 71 S. KİZİR.........................39,40
B. ORUÇ .......................... 36
L. BAŞAYİĞİT ................. 30 Ş. S. MENTEŞ ................. 74
C. YAZICI .....................62,63 L. H. AL-SAMARAI ........ 24 T. KURBAN .................... 33
Ç. OZER ………………….70 L. WANG ............................. 28 T. T. DÜNDAR ..............72,73
D. AKYIL ......................... 74 M. ACAR .......................... 53 T. YAREN ............................. 39
E. CAN...........................48,49 Ö. Evrim GÜNDOĞDU….37 V. FERYAD .......................... 67
B.RICHTER………….. 47 M. DEDEOĞLU ............... 30 Y. AHİ .............................. 66
E. KELEKÇI ..................39,40 M. K. GÜLLÜ...........48,49,50 Y. ÇOŞKUN ..................... 35
M. MAHMOOD ............... 25 Y. ÖZEREN ..................... 56
E. PARLAK ........................ 70 M. PEHLİVANOĞLU ...... 53
E. T. ÖZDEMIR ............... 75 M. TAKAOĞLU ............... 70 Z. ABOOD ....................... 31
F. DÖKMEN ............... 68-69 N. DURU .......................54,69 Z. ASLAN .................68,74,75
F. FARUKH ........................ 54 N. M. ATHAB ..................... 26 Z. ERDOĞAN ................. 55
G. ÇİMEN ........................ 76 N. Ö. SÜZME ................... 51
G. E. GÜRAKSIN ............ 51 O. ASHOUR ..................... 25
G. KÜÇÜKYILDIZ ..44,45,46 O. BAYAT ............24,25,26,27
78
10TH INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, WAVELET AND APPLICATIONS
IWW2019, 18-20 October 2019
KOCAELI, TURKEY
SPONSORS
Kocaeli Büyükşehir Belediyesi
(GOLD SPONSOR)
Kartepe Belediyesi
(GOLD SPONSOR)
Eurasian Universities Union
(GOLD SPONSOR)
ASİS Otomasyon ve
Akaryakıt Sistemleri A.Ş
(BRONZ SPONSOR)
Yilmazlar Ayakkabi Ltş., Istanbul
ABSTRACTBOOKLET
