Applications of Computational

Intelligence in Multi-Disciplinary
Research Ahmed A. Elngar
Visit to download the full and correct content document:
https://ebookmass.com/product/applications-of-computational-intelligence-in-multi-dis
ciplinary-research-ahmed-a-elngar/
Applications of Computational Intelligence in
Multi-Disciplinary Research
Advances in Biomedical Informatics

Applications of
Computational Intelligence
in Multi-Disciplinary
Research

Edited by
Ahmed A. Elngar
Faculty of Computers and Artificial Intelligence, Beni-Suef University, Beni-Suef City, Egypt
College of Computer Information Technology, American University in the Emirates,
United Arab Emirates

Rajdeep Chowdhury
Department of Computer Application, JIS College of Engineering, Kalyani, West Bengal, India

Mohamed Elhoseny
College of Computing and Informatics, University of Sharjah, United Arab Emirates
Faculty of Computers and Information, Mansoura University, Egypt

Valentina Emilia Balas

Department of Automatics and Applied Software, Faculty of Engineering,
“Aurel Vlaicu” University of Arad, Arad, Romania

Series Editor
Valentina Emilia Balas
Department of Automatics and Applied Software, Faculty of Engineering,
“Aurel Vlaicu” University of Arad, Arad, Romania
Academic Press is an imprint of Elsevier
125 London Wall, London EC2Y 5AS, United Kingdom
525 B Street, Suite 1650, San Diego, CA 92101, United States
50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom
Copyright © 2022 Elsevier Inc. All rights reserved.
No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including
photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.
Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with
organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website:
www.elsevier.com/permissions.
This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be
noted herein).
Notices
Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding,
changes in research methods, professional practices, or medical treatment may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information,
methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their
own safety and the safety of others, including parties for whom they have a professional responsibility.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury
and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of
any methods, products, instructions, or ideas contained in the material herein.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
A catalog record for this book is available from the Library of Congress
ISBN: 978-0-12-823978-0

For Information on all Academic Press publications

visit our website at https://www.elsevier.com/books-and-journals

Publisher: Mara Conner

Acquisitions Editor: Chris Katsaropoulos
Editorial Project Manager: Isabella C. Silva
Production Project Manager: Selvaraj Raviraj
Cover Designer: Greg Harris
Typeset by MPS Limited, Chennai, India
Contents

List of contributors ix 3. Behavioral malware detection and

classification using deep learning
1. Iris feature extraction using approaches 29
three-level Haar wavelet transform
T. Poongodi, T. Lucia Agnes Beena,
and modified local binary D. Sumathi and P. Suresh
pattern 1
3.1 Introduction 29
Prajoy Podder, M. Rubaiyat Hossain Mondal and 3.1.1 Digital forensics—malware
Joarder Kamruzzaman detection 30
Abbreviations 1 3.1.2 Malware evolution and its
1.1 Introduction 1 taxonomy 32
1.2 Related works 3 3.1.3 Machine learning techniques for
1.3 Iris localization 4 malware analysis 32
1.4 Iris normalization 6 3.1.4 Behavioral analysis of malware
1.5 The proposed feature extraction detection 33
scheme 7 3.2 Deep learning strategies for malware
1.6 Matching results 11 detection 35
1.7 Performance evaluation 11 3.2.1 Feature extraction and data
1.8 Conclusion 13 representation 35
References 14 3.2.2 Static Analysis 36
3.2.3 Dynamic analysis 38
3.2.4 Hybrid analysis 38
3.2.5 Image processing techniques 38
2. A novel crypt-intelligent 3.3 Architecture of CNNs for malware
cryptosystem 17 detection 41
3.3.1 Preprocessing 41
Pratyusa Mukherjee and
3.3.2 Classification using CNNs 41
Chittaranjan Pradhan
3.3.3 Evaluation 42
2.1 Introduction 17 3.4 Comparative analysis of CNN
2.2 Related work 18 approaches 42
2.2.1 Machine learning contributions in 3.5 Challenges and future research
cryptology 18 directions 43
2.2.2 Genetic algorithm contributions in 3.6 Conclusion 43
cryptology 20 References 43
2.2.3 Neural network contributions in
cryptology 21 4. Optimization techniques and
2.2.4 Background of DNA cryptography 23 computational intelligence with
2.3 Proposed methodology 23 emerging trends in cloud
2.3.1 Proposed encryption scheme 24 computing and Internet of Things 47
2.3.2 Proposed decryption scheme 25
2.4 Discussion 25 Jayesh S Vasudeva, Sakshi Bhargava and
Deepak Kumar Sharma
2.5 Conclusion and future work 26
References 26 4.1 Introduction 47

v
vi Contents

4.1.1 Introduction to optimization 48 6.2 Supervisory control and data acquisition

4.1.2 Introduction to cloud computing with systems 84
emphasis on fog/edge computing 48 6.3 The very small aperture terminal
4.2 Optimization techniques 49 networks 87
4.2.1 An optimization problem 49 6.3.1 The satellite communication
4.2.2 Solution to the optimization problem 52 systems 87
4.3 Understanding fog/edge computing 54 6.3.2 Architecture very small aperture
4.3.1 What is fog? 54 terminal networks 89
4.3.2 Prelude to our framework 54 6.3.3 Connectivity 92
4.3.3 Our goal 55 6.3.4 Multiple access 93
4.3.4 Framework for fog computing 55 6.4 Algorithm for estimating the satellite
4.4 Optimizing fog resources 57 bandwidth 95
4.4.1 Defining optimization problem for 6.4.1 Determining the bandwidth
fog layer resources 57 required for data transmission 95
4.4.2 Optimization techniques used 58 6.4.2 Case study 98
4.5 Case studies 60 6.4.3 Overview of some recent
4.5.1 Case study I: floorplan optimization 60 algorithms in detail 99
4.5.2 Case study II: Gondwana— 6.4.4 Validation of bandwidth
optimization of drinking water calculations 104
distribution system 63 6.5 Challenges and future work 106
4.6 Scope of advancements and future 6.6 Conclusions 107
research 63 References 107
4.7 Conclusion 64
References 65
7. Using artificial intelligence search
5. Bluetooth security architecture in solving the camera placement
cryptography based on genetic problem 109
codons 67 Altahir A. Altahir, Vijanth S. Asirvadam,
Nor Hisham B. Hamid and Patrick Sebastian
Asif Ikbal Mondal, Bijoy Kumar Mandal,
Debnath Bhattacharyya and Tai-Hoon Kim Nomenclature 109
7.1 Introduction 109
5.1 Introduction 67
7.1.1 The roles of visual surveillance
5.1.1 Bluetooth 67
systems 110
5.1.2 Bluetooth security architecture 67
7.1.2 The camera placement problem
5.2 Survey of literature 69
from an artificial intelligence
5.3 Plaintext-to-ciphertext conversion
perspective 111
process 71
7.1.3 Chapter description 112
5.3.1 Basic workflow 71
7.2 Background 112
5.3.2 Algorithm 72
7.3 Modeling the visual sensors 113
5.3.3 Analysis and discussion 78
7.3.1 The sensor space modeling 114
5.4 Conclusion 79
7.3.2 The camera coverage modeling 114
5.5 Future work 79
7.3.3 The analysis of camera visibility 115
References 80
7.4 Solving the camera placement problem
using artificial intelligence search 116
6. Estimation of the satellite bandwidth 7.4.1 Generate and test algorithm 117
required for the transmission of 7.4.2 Uninformed search 118
information in supervisory control 7.4.3 Hill climbing strategy 119
and data acquisition systems 83 7.5 Further discussion 121
7.5.1 The efficiency of the algorithms 122
Marius Popescu and Antoanela Naaji
7.5.2 The performance of the algorithms 123
Abbreviations 83 7.6 Conclusion 123
6.1 Introduction 84 References 124
 Contents vii

8. Nanotechnology and applications 129 9.10 Nanotechnology in pesticides and

fertilizers 151
Kanika Dulta, Amanpreet Kaur Virk, 9.11 Control of plant pests 152
Parveen Chauhan, Paras Bohara and 9.12 Concluding remarks 152
Pankaj Kumar Chauhan Consent for publication 152
8.1 Introduction 129 Conflict of interest 153
8.2 Nanoscience and nanotechnology 130 References 153
8.3 Computational nanotechnology 130
8.3.1 Molecular modeling 131 10. A methodology for designing
8.3.2 Nanodevice simulation 133
knowledge-based systems and
8.3.3 Nanoinformatics 133
8.3.4 High-performance computing 135
applications 159
8.3.5 Computational intelligence 135 Hien D. Nguyen, Nhon V. Do and
8.4 Applications of computational Vuong T. Pham
nanotechnology 137
10.1 Introduction 159
8.4.1 Nanotube-based sensors and
10.2 Related work 160
actuators 138
10.3 Design the knowledge-based system 160
8.4.2 Nanoinformatics for drugs 138
10.3.1 The architecture of a
8.4.3 Molecular docking 138
knowledge-based system 160
8.4.4 Nanotoxicology 138
10.3.2 The process for designing the
8.4.5 Other applications 139
knowledge-based system 162
8.5 Conclusion 139
10.4 Knowledge base and inference engine
References 139
of a knowledge-based system 163
9. Advances of nanotechnology in 10.4.1 Design the knowledge base 163
plant development and crop 10.4.2 Design the Inference engine 164
10.5 Applications 168
protection 143
10.5.1 Design an intelligent problem
Rokeya Akter, Md. Habibur Rahman, solver for solving solid geometry
Md. Arifur Rahman Chowdhury, at high school 168
Manirujjaman Manirujjaman and 10.6 Conclusion and Future work 183
Shimaa E. Elshenawy References 184
9.1 Introduction 143
9.2 Agriculture’s nanofarming: a modern 11. IoT in healthcare ecosystem 187
frontier 143 Poonam Gupta and Indhra Om Prabha M
9.3 Synthesis of green nanoparticles and its
sources 144 11.1 Introduction 187
9.4 Good distribution possibilities allowed 11.2 Applications of Internet of Things in
by nanoparticles: a modern sustainable healthcare 188
agriculture portal 145 11.2.1 Patient-centric IoT 188
9.5 Nanofertilizers: a good food supply for 11.2.2 Hospital-centric IoT applications 189
crops 146 11.2.3 IoT benefitting health insurance
9.6 Germination, field production, and companies 190
efficiency enhancement of seed 11.2.4 Pharmaceutical governance 190
nanomaterials 148 11.3 Implementation methodologies 190
9.7 Plant sensory systems and responses to 11.3.1 Fog computing 191
radical climate change influences 11.3.2 Edge computing 192
nanomaterials 149 11.4 Implementation models 193
9.8 Nanosensors and nanomaterials: 11.4.1 Heart disease prediction 193
perturbation detection and control 150 11.4.2 Healthcare IoT-based affective
9.9 Pesticide-based plant safety state mining using deep
nanomaterials 150 convolutional neural networks 194
viii Contents

11.5 Challenges in healthcare IoT 195 11.7.3 Quarantined patient care 199
11.5.1 Technology-oriented challenges 195 11.7.4 Public surveillance 200
11.5.2 Adapting to remote healthcare 11.7.5 Safeguarding hygiene 200
and telehealth 195 11.7.6 IoT and robotics 200
11.5.3 Data security 195 11.7.7 Smart disinfection and sanitation
11.6 Security issues and defense tunnel 201
mechanisms and IoT 196 11.7.8 Smart masks and smart medical
11.6.1 Security requirements in equipment 201
healthcare IoT 196 11.8 Future of IoT in healthcare 202
11.6.2 Attacks on IoT devices 196 11.8.1 IoT and 5G 202
11.6.3 Defensive mechanism 197 11.8.2 IoT and artificial intelligence 202
11.7 Covid 19—how IoT rose to the global 11.9 Conclusion 203
pandemic 198 References 203
11.7.1 About Covid 19 199
11.7.2 Decoding the outbreak and Index 205
identifying patient zero 199
List of contributors
T. Lucia Agnes Beena Department of Information
Technology, St. Josephs College, Tiruchirappalli,
India
Rokeya Akter Department of Pharmacy, Jagannath
University, Dhaka, Bangladesh
Altahir A. Altahir Department of Electrical and
Electronics Engineering, Universiti Technologi
PETRONAS, Perak, Malaysia
Vijanth S. Asirvadam Department of Electrical and
Electronics Engineering, Universiti Technologi
PETRONAS, Perak, Malaysia
Sakshi Bhargava Department of Physical Sciences and
Engineering, Banasthali Vidyapith, Tonk, India
Debnath Bhattacharyya Computer Science and
Engineering Department, Koneru Lakshmaiah
Education Foundation, Guntur, India
Paras Bohara Faculty of Applied Sciences and
Biotechnology, Shoolini University of Biotechnology
and Management Sciences, Solan, India
Pankaj Kumar Chauhan Faculty of Applied Sciences
and Biotechnology, Shoolini University of
Biotechnology and Management Sciences, Solan, India
Parveen Chauhan Faculty of Applied Sciences and
Biotechnology, Shoolini University of Biotechnology
and Management Sciences, Solan, India
Md. Arifur Rahman Chowdhury Department of
Pharmacy, Jagannath University, Dhaka, Bangladesh;
Department of Bioactive Materials Science, Jeonbuk
National University, Jeoju, South Korea
Nhon V. Do Hong Bang International University, Ho Chi
Minh City, Vietnam
Kanika Dulta Faculty of Applied Sciences and
Biotechnology, Shoolini University of Biotechnology
and Management Sciences, Solan, India
Shimaa E. Elshenawy Center of Stem Cell and
Regenerative Medicine, Zewail City for Science,
Zewail, Egypt
Poonam Gupta G H Raisoni College of Engineering and
Management, Pune, India
Nor Hisham B. Hamid Department of Electrical and
Electronics Engineering, Universiti Technologi
PETRONAS, Perak, Malaysia
Joarder Kamruzzaman School of Engineering and
Information Technology, Federation University
Australia, Churchill, VIC, Australia
Tai-Hoon Kim Computer Science and Engineering
Department, Global Campus of Konkuk University,
Chungcheongbuk-do, Korea
Deepak Kumar Sharma Department of Information
Technology, Netaji Subhas University of Technology
(formerly known as Netaji Subhas Institute of
Technology), New Delhi, India
Indhra Om Prabha M G H Raisoni College of
Engineering and Management, Pune, India
Bijoy Kumar Mandal Computer Science and
Engineering Department, NSHM Knowledge Campus,
Durgapur, India
Manirujjaman Manirujjaman Institute of Health and
Biomedical Innovation (IHBI), School of Clinical
Sciences, Faculty of Health, Queensland University of
Technology, Brisbane, QLD, Australia
Asif Ikbal Mondal Computer Science and Engineering
Department, Dumkal Institute of Engineering &
Technology, Murshidabad, India
Pratyusa Mukherjee School of Computer Engineering,
KIIT Deemed to be University, Bhubaneshwar,
India
Antoanela Naaji Faculty of Economics, Computer
Science and Engineering, “Vasile Goldis” Western
University of Arad, Arad, Romania
Hien D. Nguyen University of Information Technology,
Ho Chi Minh City, Vietnam; Vietnam National
University, Ho Chi Minh City, Vietnam
Vuong T. Pham Sai Gon University, Ho Chi Minh City,
Vietnam
Prajoy Podder Bangladesh University of Engineering
and Technology, Institute of Information and
Communication Technology, Dhaka, Bangladesh
ix
x List of contributors
T. Poongodi School of Computing Science and
Engineering, Galgotias University, Greater Noida,
India
Marius Popescu Faculty of Economics, Computer
Science and Engineering, “Vasile Goldis” Western
University of Arad, Arad, Romania
Chittaranjan Pradhan School of Computer
Engineering, KIIT Deemed to be University,
Bhubaneshwar, India
Md. Habibur Rahman Department of Pharmacy,
Southeast University, Dhaka, Bangladesh
M. Rubaiyat Hossain Mondal Bangladesh University of
Engineering and Technology, Institute of Information
and Communication Technology, Dhaka, Bangladesh
Patrick Sebastian Department of Electrical and
Electronics Engineering, Universiti Technologi
PETRONAS, Perak, Malaysia
D. Sumathi SCOPE, VIT-AP University, Amaravati,
India
P. Suresh School of Mechanical Engineering, Galgotias
University, Greater Noida, India
Jayesh S Vasudeva Department of Instrumentation and
Control Engineering, Netaji Subash University of
Technology (formerly known as Netaji Subhas
Institute of Technology), New Delhi, India
Amanpreet Kaur Virk Faculty of Applied Sciences and
Biotechnology, Shoolini University of Biotechnology
and Management Sciences, Solan, India
Chapter 1

Iris feature extraction using three-level

Haar wavelet transform and modified
local binary pattern
Prajoy Podder1, M. Rubaiyat Hossain Mondal1 and Joarder Kamruzzaman2
1
Bangladesh University of Engineering and Technology, Institute of Information and Communication Technology, Dhaka, Bangladesh, 2School of
Engineering and Information Technology, Federation University Australia, Churchill, VIC, Australia

Abbreviations
^ yÞ
iðx; 2D cepstrum with ðx; yÞ representing quefrency coordinates
Iðu; vÞ 2D discrete-time Fourier Transform
Gðx; yÞ 2D Gabor function
FðU; VÞ 2D discrete cosine transform (DCT) coefficient matrix
W Angular frequency
σx andσy Standard deviations of x and y
xc The x-axis coordinate of the iris circle
yc The y-axis coordinate of the iris circle
r Radius of the iris circle
gc Gray level of the center pixel, c
gp Gray level of the neighboring pixel, p


ψ" sp Binary iris code obtained as XOR output
LBPp MLBP operator

1.1 Introduction
The concern of high security and surveillance in the present world has made the identification of people an increasingly
important issue. Among various identification modes, biometric has been considered over the last few decades for its
reliable and accurate identification [15]. Commonly used biometric features include the face, fingerprint, iris, retina,
hand geometry, and DNA identifications. Among them, nowadays, iris recognition has attracted significant interest in
research and commercialization [615]. Iris recognition has several applications in the security systems of banks, bor-
der control, restricted areas, etc. [13]. One key part of such a system is the extraction of prominent texture information
or features in the iris. This feature extraction method generates feature vectors or feature codes. The feature vectors of
the unknown images are used to match those of the stored known ones. In an iris recognition system, the matching pro-
cess matches the extracted feature code of a given image with the feature codes previously stored in the database. In
this way, the identity of the given iris image can be known.
A generalized iris recognition scheme is presented in Fig. 1.1. There are two major parts of Fig. 1.1, one showing
the feature extraction and the other describing the identification portion of an iris. The system starts with image acquisi-
tion and ends with matching, that is, the decision of acceptance or rejection of the identity. In between, there are two
main stages: iris image preprocessing and feature extraction [3,4]. Furthermore, iris image preprocessing includes the
stages of iris segmentation, normalization, and enhancement [5,11]. In the acquisition stage, cameras are used to capture
images of the iris. The acquired images are then segmented. In iris segmentation, the inner and the outer boundaries are

Applications of Computational Intelligence in Multi-Disciplinary Research. DOI: https://doi.org/10.1016/B978-0-12-823978-0.00005-8

© 2022 Elsevier Inc. All rights reserved. 1
2 Applications of Computational Intelligence in Multi-Disciplinary Research

FIGURE 1.1 Example of a typical iris recognition

system: (A) process of feature extraction from an iris
image; (B) identification of an iris.

detected to separate the iris from the pupil and sclera. A circular edge detection method is used to segment the iris
region by finding the pixels of the image that have sharp intensity differences with neighboring pixels [3]. Estimating
the center and the radius of each of the inner and outer circles refers to iris localization. After iris segmentation, any
image artifacts are suppressed. Next is the normalization step in which the images are transformed from Cartesian to
pseudo polar scheme. This is shown in Fig. 1.1, where boundary points are aligned at an angle. Image enhancement is
then performed. As a part of feature extraction, the important features are extracted and then used to generate an iris
code or template. Finally, iris recognition is performed by calculating the difference between codes with the use of a
matching algorithm. For this purpose, Hamming and Euclidian are well known and also considered in this chapter [15].
The matching score is compared with a threshold to determine whether the given iris is authentic or not.
Despite significant research results so far [39,11,12,14], there are several challenges in iris recognition
[13,1526]. One problem is the occlusion, that is, the hiding of the iris caused by eyelashes, eyelids, specular reflec-
tion, and shadows [21]. Occlusion can introduce irrelevant parts and hide useful iris texture [21]. The movement of the
eye can also cause problems in iris region segmentation and thus accurate recognition. Another issue is the computation
time of iris identification. For large population sizes, the matching time of the iris can sometimes become exceedingly
high for real-time applications, and the identification delay increases with the increase in the population size and the
length of feature codes. It has been reported in the recent literature [13,18,22] that the existing iris recognition methods
still suffer from long run times apart from other factors. This is particularly true when the sample size is very large, and
the iris images are nonideal and captured from different types of cameras. Hence, devising a method that reduces the
run time of iris recognition without compromising accuracy is still an important research problem. The identification
delay can be reduced by reducing the feature vector of iris images. Thus this chapter focuses on the issue of reducing
the feature vector which will lead to a reduction in identification delay without lowering the identification accuracy.
For lowering the feature vector, the concept of Haar wavelet along with modified local binary pattern (MLBP) is used
in this work. Note that in the context of face recognition [2730] and fingerprint identification [31], the Haar wavelet
transform demonstrates an excellent recognition rate at a low computation time. In Ref. [32], the Haar wavelet is also
proposed without the use of MLBP.
The main contributions of this chapter can be summarized as follows.
1. A new iris feature extraction method is proposed. This new method is based on repeated Haar wavelet transforma-
tion (HWT) and MLBP. Note that MLBP is the local binary pattern (LBP) operation followed by Exclusive OR
(XOR). This proposed method is different from the technique described in Ref. [30], which uses single-level HWT
and LBP (without XOR) in the context of face recognition.
 Iris feature extraction using three-level Haar wavelet transform and modified local binary pattern Chapter | 1 3

2. The efficacy of the HWTMLBP method is evaluated using three well-known benchmark datasets: CASIA-Iris-V4
[33], CASIA-Iris-V1 [34], and MMU iris database [35].
3. A comparison is made of this new technique with the existing methods of feature extraction in terms of feature vec-
tor length, false acceptance rate (FAR), and false rejection rate (FRR). It is shown here that the proposed method
outperforms the existing ones in terms of feature vector length.
The remainder of this chapter is organized as follows. Section 1.2 provides a literature survey of the relevant
research. Section 1.3 shows the iris localization part where the inner boundary and outer boundary can be detected.
Section 1.4 describes iris normalization. Section 1.5 illustrates our proposed approach for the purpose of encoding the
iris features. Section 1.6 describes the iris recognition process by matching score. The effectiveness of the new method
is evaluated in Section 1.7. Finally, Section 1.8 provides a summary of the research work followed by the challenges
and future work.

1.2 Related works

A number of research papers describe iris feature extraction techniques, which are discussed in the following.
Ma et al. [3] applied a bank of spatial filters to acquire local details of the iris. These spatial filters generate discrim-
inating texture features for an iris image based on the characteristics of the iris. Ma et al. [4] considered a bank of circu-
lar symmetric filters for iris feature extraction. These filters [4] are modulated by a circular symmetric sinusoidal
function, which is different from the Gabor filter modulated by an orientated sinusoidal function. Monro et al. [5] used
discrete cosine transform (DCT) for iris recognition. Daugman [6] introduced the idea of using a 2D Gabor wavelet fil-
ter for extracting features of an iris image. Furthermore, Masek et al. [9] used 1D and 2D Log-Gabor filters for feature
extraction. Li et al. [8] used a convolutional neural network (CNN) algorithm, which is a form of deep learning, to
extract iris features. Umer et al. [12] used a novel texture code defined over a small region at each pixel. This texture
code was developed with vector ordering based on the principal component of the texture vector space. Soliman et al.
[11] considered feature extraction using the Gabor filter, where the original Gabor features were masked via a random
projection scheme. The masking was performed to increase the level of security. In this scheme, the effects of eyelids
and eyelashes were removed. An iris feature extraction method using wavelet-based 2D mel-cepstrum was proposed in
Ref. [14], where the cepstrum of a signal is the inverse Fourier transform of the logarithm of the estimated signal spec-
trum. The 2D cepstrum of an image can be defined by the following expression:
 
^ yÞ 5 IDFTðlogðI ðu; vÞ2 ÞÞ;
iðx;
^ yÞ is the 2D cepstrum with ðx; yÞ representing quefrency coordinates, IDFT represents the inverse discrete
where iðx;
Fourier transform, and Iðu; vÞ is the 2D discrete-time Fourier Transform of the image. This scheme applied the
CohenDaubechiesFeauveau 9/7 filter bank for extracting features. In wavelet cepstrum, nonuniform weights are
assigned to the frequency bins. In this way, the high-frequency components of the iris image are emphasized, resulting
in greater recognition reliability. Furthermore, this wavelet cepstrum method helps to reduce the feature set.
Barpanda et al. [15] used a tunable filter bank to extract region-based iris features. These filters were used for recog-
nizing noncooperative images instead of high-quality images collected in cooperative scenarios. The filters in this filter
bank were based on the halfband polynomial of 14th order where the filter coefficients were extracted from the polyno-
mial domain. To apply the filter bank, the iris template was divided into six equispaced parts and the features were
extracted from all the parts except the second one, which mainly contains artifacts. Betancourt et al. [13] proposed a
robust key pointsbased feature extraction method. To identify distinctive key points, three detectors, namely
HarrisLaplace, HessianLaplace, and Fast-Hessian detectors, were used. This method is suitable for iris recognition
under variable image quality conditions.
For iris feature extraction, Sahua et al. in [22] used phase intensive local pattern (PILP), which consists of density-
based spatial clustering and key-point reduction. This technique groups some closely placed key points into a single
key point, leading to high-speed matching. Jamaludin et al. [18] used a 1D Log-Gabor filter and considered the subiris
region for feature extraction. This filter has a symmetrical frequency response on the log axis. In this case, only the
lower iris regions that are free from noise, as well as occlusions, are considered.
In Ref. [17], combined discrete wavelet transform (DWT) and DCT were used for the extraction of iris features.
Firstly, DWT was performed where the output of this stage was in the spatial domain. Next, DCT was performed to
transform the spatial domain signal to the frequency domain and to obtain better discriminatory features. Another fea-
ture extraction method is the discrete dyadic wavelet transform reported in Ref. [16]. In dyadic wavelet transform, the
4 Applications of Computational Intelligence in Multi-Disciplinary Research

decomposition at each level is done in a way that the bandwidth of the output signal is half of that of the input. In Ref.
[26], a PILP technique is used for feature extraction and to obtain a feature vector of size 1 3 128. In this PILP method,
there are four stages: key-point detection via phase-intensive patterns, removal of edge features, computation of ori-
ented histogram, and formation of the feature vector. Iris features were extracted using 1D DCT and relational measure
(RM), where RM encodes the difference in intensity levels of local regions of iris images [21]. The matching scores of
these two approaches were fused using a weighted average. The score-level fusion technique compensates for some
images that are rejected by one method but accepted by the other [21]. Another way of extracting feature vectors from
iris images is by the use of linear predictive coding coefficients (LPCC) and linear discriminant analysis (LDA) [24].
Llano et al. in [19] used a 2D Gabor filter for feature extraction. Before applying this filter, the fusion of three different
algorithms was performed at the segmentation level (FSL) of the iris images to improve the textual information of the
images. Oktiana et al. [36] proposed an iris feature extraction system using an integration of Gradientface-based nor-
malization (GRF), where GRF uses an image gradient to remove the variation in the illumination level. Furthermore,
the work in Ref. [19] concatenated the GRF with a Gabor filter, a difference of Gaussian (DoG) filter, binary statistical
image feature (BSIF), and LBP for iris feature extraction in a cross-spectral system. Shuai et al. proposed [37] an iris
feature extraction method based on multiple-source feature fusion performed by a Gaussian smoothing filter and texture
histogram equalization. Besides, there have been some recent studies in the field of iris recognition [3849], with some
focusing on iris feature extraction methods [38,4042,45,49] and some on iris recognition tasks [39,44,46,48].
The 2D Gabor function can be described mathematically by using the following expression:
     
1 1 x2 y2
Gðx; yÞ 5 exp 2 1 1 2 3 π 3 j 3 Wx
2 3 π 3 σx 3 σy 2 σx 2 σy 2
and 2D DCT can be defined as:
pffiffiffi M21 N21    
2XX ð2Y 1 1ÞUπ ð2X 1 1ÞVπ
F ðU; V Þ 5 f ð X; Y Þcos cos
M X50 Y50 2M 2M

where f(X,Y) is the image space matrix; (X, Y) is the position of the current image pixel and
FðU; V ÞðU; V 5 1; 2; . . . ::; M 2 1Þ is the transform coefficient matrix; W is the angular frequency; and σx and σy are the
standard deviations of x and y, respectively.
The concepts of machine learning (ML)-driven methods for example, neural networks and genetic algorithms have
been reported [46], while the idea of deep CNNs has also been applied [40]. Moreover, researchers are now investigat-
ing the effectiveness of multimodal biometric recognition systems [43,47].
A comparative summary of some of the most relevant works on iris feature extraction is shown in Table 1.1. It can
be seen that there are several algorithms and these are applied to different datasets, achieving varying performance
results.

1.3 Iris localization

This section discusses the iris localization step that employs circular Hough transformation, which is capable of prop-
erly detecting circles in the images. Hough transform searches for a triplet of parameters (xc ; yc ; r) determining (xi ; yi ),
where xc ; yc , and r represent the x-axis coordinate, y-axis coordinate, and the radius of the iris circle, respectively. In
this case, (xi ; yi ) represents the coordinates of any of the i points on the circle. With this consideration, the Hough trans-
form can be defined as follows.
X
n
H ðxc ; yc ; r Þ 5 hðxi ; yi ; xc ; yc ; rÞ (1.3)
i51

In this regard, edge points are detected first. For each of the edge points, a circle is drawn having the center in the
middle of the edge. In this way, each of the edge points constitutes circles with the desired radius. Next, an accumulator
matrix is formed to track the intersection points of the circles in the Hough space, where the accumulator has the num-
ber of circles. The largest number in the Hough space points to the center of the image circles. Several circular filters
with different radius values are considered and the best one is selected.
 Iris feature extraction using three-level Haar wavelet transform and modified local binary pattern Chapter | 1 5

TABLE 1.1 Summary of literature review.

References Adopted technique in reference Remarks Database

[3] Spatial filters constructed based on Extraction of features is performed only in CASIA-Iris-V1
observations the upper portion of the normalized iris
region as it provides useful texture
information. The feature vector length is
large, being of size 1 3 1536.
[4] Circular symmetric filters About 75% of the top-most of the
unwrapped iris images are used for texture
information. The variation of the texture of
the iris in the local region is not focused on
in this paper.
[5] Patch-coding technique for extracting The method has low complexity with a high CASIA-Iris-V1
features from normalized iris. The amount of accuracy. Additionally, the
features are derived by using fast Fourier dimensionality of the feature vector is
transformation. 1 3 2343. However, nonideal images are not
considered.
[6] 2D Gabor filter The dimensionality of the feature vector is
1 3 2048.
[9] 1D and 2D Log-Gabor filters This method cannot produce features of CASIA-Iris-V1
different frequencies, and the size of the iris
template is 1 3 4800.
[8] Deep learning CNN as deep learning is used to extract iris CASIA-Iris-V4
features, and the features are then used for
image encryption.
[12] Texture code cooccurrence matrix Feature vector size of 1 3 400. The method UPOL, CASIA-Iris-V3-
uses only an effective portion of the iris Interval, MMU1, and
images to avoid the occlusion part caused by IITD
not only eyelashes but also eyelids.
[11] 1D Gabor filter where Gabor features Masks the original Gabor features for CASIA-Iris-V3-Interval
are masked improving security while excluding the
effects of eyelids and eyelashes. Moreover,
only the upper half of the normalized iris
portion is considered.
[7] 2D kernel and hybrid MLPNNPSO Feature extraction is performed on a small CASIA-Iris-V3
algorithm sample of 140 images at an accuracy rate of
95.36%. In this case, 1000 iterations are
performed, which leads to high
computational time.
[14] 2D wavelet cepstrum technique for False acceptance rate is 10.45%; recognition CASIA-Iris-V3,
feature extraction accuracy is 89.93%. UBIRISv1, IITD
[15] Tunable filter bank based on halfband False acceptance rate is 8.45%; recognition CASIA-Iris-V3,
polynomial of 14th order accuracy is 91.65%. UBIRISv1, IITD
[13] Key pointsbased feature extraction Considers only salient key points in the CASIA-Iris-V4-Interval,
method whole region. The feature extraction stage is MMU 2, UBIRIS 1
time consuming.
[23] Low-density parity check and SHA-512 Comparatively high false rejection rate.
[22] Density-based spatial clustering and key- For feature extraction and feature vector BATH and CASIA-Iris-
point reduction to be applied on PILP reduction, postprocessing is required, V3
leading to additional time consumption.
[18] Subiris technique Does not extract features of the unoccluded CASIA-Iris-V4
upper part of the iris region.

(Continued )
6 Applications of Computational Intelligence in Multi-Disciplinary Research

TABLE 1.1 (Continued)

References Adopted technique in reference Remarks Database

[17] DWT and DCT combination for feature Provides good performance only for low- Phoenix and IITD iris
extraction contrast images. A recognition rate of 88.5% database
was achieved on the Phoenix database.
[16] Discrete dyadic wavelet transform Iris images of only 10 people are used and a
feature vector with a size of 1 3 256 is
achieved. Results need to be validated with a
higher number of subjects.
[26] Local feature based on phase-intensive Feature extraction is based on key point BATH, CASIA-Iris-V3,
patterns detection via phase-intensive patterns. UBIRISv2, and
Obtains a feature vector of size 1 3 128. FERETv4
[21] DCT and RM Based on the dissimilarity score of DCT and CASIA-Iris-V4 Interval,
RM and using the Hamming distance metric, Lamp, and self-
the matching of images is performed. This is collected IITK
used to compensate for images rejected by
either DCT or RM but accepted by the other.
[24] LPCC and LDA The method has high complexity, and in the CASIA-Iris-V1
case of LPCC, results in a feature vector with
a size of 1 3 546.
[19] Textural information development and This method has three stages: quality MBGC-V2, CASIA-Iris-
exploration evaluation, automatic segmentation, and V3, CASIA-Iris-V4,
fusion at the segmentation level. This and UBIRIS v1 (for iris
method rejects images that have low quality. image)
The obtained feature vector is of size
1 3 2048.
[36] Gabor filter, a DoG filter, BSIF, and LBP The feature extraction is done using the Hong Kong
fusion of GRF with a Gabor filter, a DoG Polytechnic University
filter, a BSIF, and LBP. Hamming distance is Cross-Spectral Iris
used for matching purposes. Images Database
[37] Convolutional neural network Feature extraction is done using the concept JLU iris library
of feature fusion, which is achieved by using
a Gaussian filter and a texture histogram
equalizer.

1.4 Iris normalization

This section describes the iris normalization step. The size of different acquired iris images will vary because of the var-
iation in the distance from the camera, angle of image capturing, illumination level, etc. For the purpose of extracting
image features, the iris image is to be segmented and the resultant segments must not be sensitive to the orientation,
size, and position of the patterns. For this, after segmentation, the resultant element is transformed to Cartesian. In other
words, the circular iris image is transformed into a fixed dimension.
Fig. 1.2 illustrates the normalization of iris images from three datasets. For each of the datasets, the one original
input image is shown, followed by its inner and outer boundary detection, and then its segmented version, and finally
its normalized version. Fig. 1.2A describes Daugman’s rubber sheet model for iris recognition. Three original images
from three datasets are shown in Fig. 1.2B, F, and J. First of all, Fig. 1.2B is one original image from the CASIA-Iris-
V4 dataset [33]. For the iris image in Fig. 1.2BE represent the corresponding inner and outer boundaries, the
segmented version, and the normalized version, respectively. Secondly, Fig. 1.2F is one original image from the
CASIA-Iris-V1 dataset [34]. For the iris image in Fig. 1.2FI represent the corresponding inner and outer boundaries,
the segmented version, and the normalized version, respectively. Thirdly, Fig. 1.2J is one original image from the
MMU iris database [35], and Fig. 1.2KM represent the corresponding inner and outer boundaries, the segmented ver-
sion, and the normalized version, respectively.
 Iris feature extraction using three-level Haar wavelet transform and modified local binary pattern Chapter | 1 7

FIGURE 1.2 Illustrations of (A) Daugman’s rubber sheet model; (B, F, J) original input images; (C, G, K) images with inner and outer boundary
detection; (D, H, L) segmented iris regions, and (E, I, M) iris images after normalization.

1.5 The proposed feature extraction scheme

This section describes the proposed iris feature extraction method. Fig. 1.3 represents the block diagram of the proposed
three-level HWT and MLBP. The decomposition of the image three times by HWT results in the reduction in feature
size without significant loss in the image quality or important attributes. The use of MLBP further reduces the feature
vector size without loss in image attributes. Fig. 1.4 shows the three-level HWT. It can be seen from the figure that at
each level of HWT, the input image is divided into four output images. These output images are denoted as horizontal
detail (HL), vertical detail (VL), diagonal detail (HH), and approximation (LL) images. The LL subimage, also known
as the LL subband, contains significant information about the original image. In other words, the LL subband is a
coarse approximation of an image and it does not contain high-frequency information. Next, the three-level HWT algo-
rithm is discussed.
Algorithm 1: HWT
Input: Normalized iris image
8 Applications of Computational Intelligence in Multi-Disciplinary Research

FIGURE 1.3 Block diagram of the proposed approach for iris feature extraction.

FIGURE 1.4 Three-level HWT.

Output: Approximation part of level three

Main Process:
Step 1: Apply first-level HWT to the normalized iris image to generate its wavelet coefficients.
Step 2: Apply second-level HWT on the approximation part obtained from Step 1 to generate its wavelet
coefficients.
Step 3: Apply third-level HWT on the approximation part obtained from Step 2 to generate its wavelet coefficients.
Step 4: Get the level three approximation part obtained from Step 3.
The main idea of using HWT is that wavelet decomposition can transform a detailed image into approximation
images. The approximation parts contain a major portion of the energy of the images. The HWT is repeatedly executed
to shrink the information size. The results of the three-level decomposition produce a reduced characteristics region
having little loss. This is shown in Fig. 1.5. It can be noted that most of the information of the iris image is contained
in the extracted LL (low-frequency) region on the multidivided iris image as indicated by Fig. 1.5. The other regions
have less information as indicated by their low intensity (dark) levels. Fig. 1.6 illustrates the size of each level for the
three-level HWT. The application of level 1 HWT to the normalized image of size 64 3 512 results in wavelet
 Iris feature extraction using three-level Haar wavelet transform and modified local binary pattern Chapter | 1 9

FIGURE 1.5 Three-level wavelet

decomposition of normalized iris.

FIGURE 1.6 Three-level HWT

with the size of each level.

coefficients of LL1, LH1, HL1, and HH1. In this case, the approximation part of level 1, denoted as LL1, becomes of
size 32 3 256. Next, level 2 HWT is applied to LL1, which generates wavelet coefficients of LL2, LH2, HL2, and HH2.
In this case, the approximation part of level 2 (LL2) becomes of size 16 3 128. After that, level 3 HWT is applied to
LL2 to generate its wavelet coefficients LL3, LH3, HL3, and HH3. In this case, the approximation part of level 3 (LL3)
becomes of size 8 3 64. Hence a major distinctive region LL3 is obtained by performing the wavelet transformation
three times. Next, the LL3 region is used for the MLBP tasks.
Now consider the MLBP operation [25], which generates robust binary features. Furthermore, MLBP has low
computational complexity. MLBP labels each pixel based on the neighboring pixels and considering a given threshold.
MLBP then produces outputs in the binary format. This binary code can describe the local texture pattern. Note that
MLBP is an LBP followed by an XOR operation. Next, MLBP operation is described in the following.
For a center pixel c, and neighboring pixels p within a neighborhood of P pixels, the MLBP operation can be
expressed as follows.
X
P21

LBPp 5 S gp 2 g c 3 2p (1.4)
p50

where LBPp is the MLBP operator, gc is the gray level of c, and gp is the gray level of p pixels. Moreover, SðxÞ in (4)
refers to the sign function defined as,
1 ifx $ 0
Sð x Þ 5 (1.5)
0 otherwise
10 Applications of Computational Intelligence in Multi-Disciplinary Research

FIGURE 1.7 Center element of a 3 3 3 pixel

image.

FIGURE 1.8 MLBP operation of

a 3 3 3 subregion: (A) the neigh-
borhood of a pixel within the
image, (B) the threshold version of
the neighborhood, and (C) the
MLBP pattern where the middle
pixel has been computed.

Next, the center pixel value is generated by applying XOR operation on the values of LBPp . This results in the fol-
lowing expression.
ψ" ðsp Þ 5 so ". . ."sP-1 (1.6)


where " denotes the XOR operator and ψ" sp is the binary iris code obtained as the XOR output. Since it is a com-
mutative operation of XOR, this can be performed by circularly shifting on sp in the clockwise or anticlockwise direc-
tion. Now XOR is performed to reduce the size from 8 3 64 to 1 3 64. XOR is computed in the column vector. In other
words, the eight-row iris signature is reduced to only a single row. Figs. 1.7 and 1.8 describe the MLBP operation.
Fig. 1.7 shows the center pixel in a 3 3 3 neighborhood, while Fig. 1.8 illustrates the computation of LBP8;1 with XOR
for a single pixel.
Algorithm 2: Feature encoding using the proposed MLBP
Input: Level three approximation part of the normalized image
Output: Binary sequence of the normalized iris image.
Main Process:
Step 1: Read the intensity values of the level three approximation part of the normalized image.
Step 2: Convert the RGB image to grayscale form.
Step 3: Resize the image if required and then store the size ½M; N of the image.
Step 4: Divide the image into eight segments.
Step 5: For each of the image segments, apply a 3 3 3 kernel.
Step 6: For i 5 1: P // P 5 8 for a 3 3 3 kernel.
Step 7: Compute DðiÞ 5 gp ðiÞ 2 gc ðiÞ // gp is the gray level for neighboring pixels and gc is the center pixel.
Step 8: If DðiÞ , 0
setUSðiÞ 5 0
else
SðiÞ 5 1
end.
Step 9: Compute LBP_p 5 XOR(SðiÞ); // Apply XOR operation to get the binary mask.
Step 10: Place the binary output of the XOR operation in the center pixel.
Step 11: Move the kernel in order to obtain a binary template.
Step 12: Apply XOR operation across the columns.
So, for the case of MLBP, the first LBP operation extracts the distinctive features to generate a unique iris code.
This code is reduced from 8 3 64 features to 1 3 64 by applying the XOR operation.
 Iris feature extraction using three-level Haar wavelet transform and modified local binary pattern Chapter | 1 11

1.6 Matching results

This section provides results on iris recognition using the matching process of iris codes. In order to find the similarity
or to measure the closeness of an unknown iris code with a template iris code, the distance between these two is calcu-
lated. The distance is a method of defining the degree of matching between two iris codes. For this, Euclidean and
Hamming distances are considered. The Euclidean distance is calculated as follows.
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ED 5 ðX2 2X1 Þ2 1 ðY2 2Y1 Þ2 (1.7)

where ED is the Euclidean distance between two coordinate points: (X1,Y1) and (X2,Y2). In the case of Hamming dis-
tance between two iris codes, the number of unmatched bits is divided by the number of bits used for comparison. The
main operation in the Hamming distance is the use of an XOR gate which computes the disagreement between two
input bits. If P and Q are two bitwise templates of iris images and N is the number of bits of each iris code, then the
Hamming distance can be mathematically expressed as follows.

1X N
HD 5 Pj ðXORÞQj (1.8)
N j51

where HD denotes the Hamming distance. Hence according to (6), HD 5 0 indicates complete similarity between two
iris codes, while HD 5 1 means total dissimilarity between the codes. In practice, the two iris codes are assumed to be
the same, if the Hamming distance is lower than a threshold. Similar to the work in Ref. [6], this chapter considers a
Hamming distance value of 0.32 for iris templates to be identical.

1.7 Performance evaluation

This section discusses the experimental results of the proposed method. For the experimentation, images are obtained
from three different datasets [3335]. Figs. 1.91.11 correspond to images from the CASIA-IRIS-V4 [33], CASIA-
IRIS-V1 [34], and MMU [35] datasets, respectively. The datasets are described in Table 1.2. The CASIA-IRIS-V4 data-
set consists of 2639 images of 249 subjects/persons. On the other hand, the CASIA-IRIS-V1 dataset has 756 iris images
from 108 eyes of 54 subjects, while the MMU dataset consists of 450 images of 45 subjects. Firstly, one original iris
image from Ref. [33] is illustrated in Fig. 1.9A, whereas Fig. 1.9B is the corresponding template after the application of
LBP to LL3 of size 8 3 64. For clarity, Fig. 1.9C illustrates a larger view of the final template shown in Fig. 1.9B. The
template is further reduced to a size of 1 3 64 by applying XOR operation through column vectors. Secondly,
Fig. 1.10AC illustrate another original iris image from Ref. [34], its corresponding template after applying LBP, and
a larger view of the template, respectively. Thirdly, Fig. 1.11AC illustrate the same for an original iris image from
Ref. [35].
The performance of the proposed method is evaluated for the three datasets mentioned above. For each dataset, 90%
(rounded up to the next integer) of the images are considered for training while the remaining are considered for testing.
Table 1.3 presents the successful recognition rate of the proposed method using Hamming distance and Euclidean dis-
tance. The results are obtained only for the testing iris images. It can be seen from Table 1.3 that for the CASIA-IRIS-
V1 dataset, the proposed algorithm obtains an average correct recognition rate of 98.30% and 97.60% for the case of
Hamming distance and Euclidean distance, respectively. The recognition rates for the other two datasets are slightly
lower, as shown in Table 1.3.
Next, the proposed new method is compared with the existing techniques reported in the literature [3,4,6,16,24].
Table 1.4 presents the comparative results of the proposed method with the previous ones. For the proposed method,

FIGURE 1.9 (A) An original iris

image from the CASIA-IRIS-V4
dataset [33], (B) the final gener-
ated iris template, and (C) a larger
view of the binarized template.
Another random document with
no related content on Scribd:
habían de estar ciegos los que
juegan con ellos. Y todo es
sufridero para con otras
tacañerías que se usan, y la
mayor de todas es cuando meten
dados cargados, que llaman
brochas, los cuales hacen de esta
manera: que á los que llaman de
mayor, por la parte del as hacen
un agujero hueco y allí meten un
poco de azogue, que es muy
pesado, y á los de menor donde
están los seis puntos; y después
tapan el agujero, que es muy sutil,
y encima pintan uno ó dos puntos
para que no se vean, y estos
dados llevan los chocarreros
escondidos, y cuando tienen una
suerte de doce ó trece ó catorce
puntos, echan los dados de
manera que se les caya alguno
en el suelo, y haciendo que se
baxan por él, sacan otro de los de
mayor, que meten en su lugar, y
como está cargado en el as, cae
siempre para abaxo y el seis para
arriba; y de la mesma manera
hacen cuando tienen por suerte
siete ú ocho puntos, que meten
un dado cargado en el seis
porque vaya el as para arriba,
yendo el seis para abaxo, y si es
menester meten dos dados de
esta suerte cargados de mayor, y
cuando tienen suerte de doce ó
de trece, alárganse en el parar y
en el decir, de arte que, no siendo
entendidos, todo el dinero es
suyo. Otros dados hay que llaman
falsos, que son mal pintados
porque tienen dos ases y fáltales
el seis, ó tienen dos seises,
faltándoles el as, y conforme á la
suerte que echan y á la necesidad
que tienen, se aprovechan dellos
metiéndolos en el juego tan bien
como las brochas. Y cuando
juegan á las tablas no penséis
que se descuidan los hombres
desta professión, que lo mesmo
hacen con los dados, y
verdaderamente yo tengo por
malo y dañoso también este
juego, assí por jugarse con
dados, como por ser trabajoso y
mohino. Á todos los otros juegos
podéis levantaros y os toman en
una petrera; habéis de esperar á
que se acabe el juego, perdiendo
á cada mano y cada vez que
echáis los dados sabiendo que se
echa para perder y no para ganar,
y assí es el juego más aparejado
de todos para perder la paciencia,
porque es menester esperar á
que el juego ó el dinero se
acaben. Y aunque yo no os he
dicho de diez partes la una de los
males y trabajos y fatigas y
persecuciones y desasosiegos y
afrentas, menguas y deshonras y
infamia que se siguen del juego,
de lo dicho podréis collegir cuán
perjudicial es, assí para la salud
como para la hacienda y la honra
de las gentes que lo siguen;
porque pocos hay que jueguen,
por ricos y caballeros y grandes
señores que sean, que no les
pese de perder, y muchos destos
se acodician á jugar mal por
ganar, y assí veréis muchas
personas de muy gran autoridad,
y de quien apenas se podría
creer, que hacen malos juegos,
por la buena estima y reputación
en que están tenidos que,
apremiados de la conciencia,
restituyen dineros mal ganados,
de los cuales yo conozco algunos
que lo han hecho.
Bernardo.—¿De manera que
queréis condenar á todos los
juegos del mundo y no dejar
ninguno para recreación de la
vida y para poder pasar la
ociosidad del tiempo?
Antonio.—No digo yo tal cosa,
que otros juegos hay lícitos, assí
como birlos, pelota y axedrez y
los semejantes á éstos, y esto se
entiende jugando pocos dineros y
que se tome más por recreación
que no por vía de vicio y exercicio
continuo, de manera que por ellos
dexen de entender las gentes en
lo que les conviene, que si esto
se hace ya dexan de ser buenos y
honestos y se convierten en la
naturaleza de los que habemos
reprobado, y aun de tal manera
se podrían usar los juegos de
naipes y dados que no pudiesen
tener reprensión; pero hay pocos
que no comiencen por poco que
si tienen aparejo no vengan á
picarse y á perder ó ganar en
mucha cantidad, y por esto tengo
por mejor dexarlos del todo. Y si
queréis que concluya, todo lo
dicho es poco y casi nada, porque
son trabajos y premios y
galardones del mundo. Lo que
toca á la ánima y á la conciencia
es lo que hace al caso, y lo que
más debríamos temer y
ponérsenos delante de los ojos,
para no solamente dexar de jugar,
pero para acordarnos de jamás
tener memoria dello; y si no
hobiera prometido de no pasar
más adelante en esta materia,
todavía dixera algo que
aprovechara; pero assí quiero
dexarlo para cuando tengáis más
voluntad de oir lo que sobre esto
puedo deciros.
Bernardo.—Agora que habéis
comenzado, queremos que no
quede nada por decir, y estáis
obligado á hacerlo, pues de tan
buena gana os escuchamos y
estamos atentos al discurso de
vuestra plática.
Antonio.—Pues que assí es, yo
lo diré tan brevemente cuanto he
sido largo en lo pasado; porque
en esto no podré decir cosa
nueva, ni que dexe de estar
escrita por muchos doctores,
canonistas y legistas y teólogos
que desmenuzan y apuran esta
materia de las restituciones
declarando los decretos y leyes
en ella, altercando cuestiones y
determinando la verdad dellas,
hasta dexarlo todo en limpio; y
quien quisiese satisfacerse y verlo
todo á la clara, lea á Santo Tomás
y á Grabiel, y al Antonio,
arzobispo de Florencia, al
Cayetano, que éstos sin otros
muchos le dirán lo cierto, y
porque no dexéis de llevar alguna
cosa en suma de que podáis
aprovecharos, digo que todos les
que ganan en los juegos con
naipes ó dados falsos ó con otro
cualquier género de las
chocarrerías y traiciones que he
dicho, están obligados á
restituirlo, so pena de irse al
infierno, conforme á lo que dice
San Agustín: Non dimittitur
peccatum, nisi restituatur ablatum.
Pues lo que assí se gana, tomado
y hurtado es, siendo encubierto,
como si fuese robo manifiesto.
Anssí mesmo, todo lo que se
gana á personas que lo que
juegan no es suyo, ni pueden
disponer dello sin licencia de otra
persona, así como los criados que
juegan los dineros ó haciendas de
sus amos, los esclavos que
juegan las de sus señores, los
hijos que para esto toman las
haciendas á sus padres, los que
tienen curadores y por falta de
edad no pueden disponer de sus
haciendas, y también los que
ganan dineros á otros que saben
que los han ganado mal y están
(antes que los juegen) obligados
á la restitución dellos. Lo que se
gana á personas simples y á
enfermos necesitados, lo que se
gana atrayendo á uno por fuerza
ó por engaño ó por grandes
persuaciones á que juegue, todo
esto obliga á restitución; y en
otros muchos casos que dexo de
decir, en que hay la mesma
obligación, el cómo y cuándo y en
qué manera se haya de restituir,
déxolo para que lo veáis en los
doctores que os he dicho, y
también porque los confesores os
avisarán de ello, aunque lo mejor
sería no tener en este casso
necesidad de sus consejos.
Solamente quiero agora que
consideréis, señores, entre
vosotros, pues sois tahures y
habéis conversado y tratado con
tahures, ¿cuántos habéis visto tan
limpios y tan recatados que
tengan advertencia á estas cosas,
sino, bien ó mal, juegan con quien
quiera, trayan dineros suyos ó
sean cuyos fuesen, sean libres ó
siervos, padres ó hijos, bobos ó
sabios, los dineros que traen mal
habidos? Por cierto pocos ó
ninguno hay que dexen de hacer
á cualesquiera dineros destos, y
procurar de ganarlos de la
manera que pudieren, alegando
que no están obligados á la
especulación destas cosas, ni á
saberlas; sabiendo que la
ignorancia no excusa el pecado y
que San Pablo dice (Ad. Cor.,
xiii): Ignorans ignorabitur. Y si
queréis que os diga lo que siento
verdaderamente de los que esto
hacen, se puede presumir que no
son verdaderos cristianos, ni
sienten bien de la fe, porque más
adoran á los naipes que á Dios,
más quieren los dados qué todos
los santos, que por jurar no oyen
misa ni sermón los días de
fiestas, por el juego pierden todos
los otros oficios divinos, y se
estarán una semana sin entrar en
la iglesia; si hacen alguna oración
ó devoción es por ganar; las
cuentas que traen y lo que por
ellas rezan es echar cuentas
cómo ganarán las haciendas á
sus prójimos. Si pierden es
abominable cosa su decir mal á
Dios y blasfemar, y si lo dexan de
decir en público, es porque temen
más el castigo del cuerpo que el
del alma y el del mundo más quel
del infierno. Así que siendo
cristianos usan tan mal de la
cristiandad, que roban las
haciendas ajenas y se
aprovechan dellas, pierden el
tiempo y muchas veces pagan de
sus haciendas lo que han ganado
de las otras, de los que viven de
la manera que ellos, quedando
todos debaxo de la obligación de
restituirlas. ¿Qué diremos sin esto
de los que buscan supersticiones
y hechicerías para ganar con ellas
diciendo que tienen virtud para
ello? Y assí unos traen consigo
nóminas con nombres no
conoscidos, ó por mejor decir de
demonios, otros traen sogas de
ahorcados, otros las redecillas ó
camisas en que nacen vestidos
los niños, algunos traen
mandrágulas y otras mil
suciedades y abominaciones. Por
cierto éstos tienen en tan poco
sus ánimas, que las darán á
trueque de ganar cuatro reales
por ellas. Pues decidme, señor
Bernardo, ¿qué os parece cómo
es bueno el juego para el cuerpo
y para el alma? ¿y qué provechos
son tan grandes los que dél se
sacan? ¿No es bien dexar su
amistad y trato y conversación á
cualquier tiempo que sea, pues
que debaxo los halagos y
placeres y deleites que dél se
siguen hay tantos y tan grandes
desabrimientos, tantas afrentas y
menguas, tan terribles
desasosiegos, tanta turbación y
peligros, principalmente para la
salvación de nuestras almas?
Mirad bien en ello y consideraldo
todo, que aunque nosotros como
malos cristianos no tuviésemos
atención al daño y perjuicio de
nuestras conciencias, la
habríamos de tener á que ningún
contentamiento ni descanso de el
juego hay que después no se
vuelva en doblado trabajo y
tristeza; y nunca dió ganancia que
no se pagase con doblada
pérdida; y en fin, es siempre
mayor el dolor que se causa del
perder que la alegría que trae
consigo el ganar; y no aleguéis á
dos ó tres ó cuatro personas que
por ventura sabéis que se hayan
hecho ricos por el juego, que
éstos son como una golondrina
en el invierno, porque por ellos
veréis mill millones de gentes
perdidas y abatidas por haber
perdido cuanto tenían. Dicho os
he mi parecer y dado os he
consejo, como pienso tomarlo
para mí, y el que estoy obligado á
daros como vuestro amigo; si os
pareciere bien, seguilde, y si no
vuestro será el daño, que á mí no
me cabrá dello más de pesarme
de ver que os quedáis tan ciegos
como hasta aquí habéis estado.
Bernardo.—No penséis, señor
Antonio, que no he caído en la
cuenta de todo lo que habéis
dicho; porque vuestras palabras
me han alumbrado el juicio y
destapado los ojos del
entendimiento, que tenía ciegos, y
con firme propósito y
determinación quedo desde agora
de no jugar en mi vida, y si jugare,
á lo menos de manera que me
puedan llamar tahur por ello, que
pues decís que pasar el tiempo
entre amigos es algunas veces
lícito, no se ganando tantos
dineros que el que los perdiese
reciba daño por ello, cuando
alguna vez me desmandase será
á esto y no á más.
Antonio.—Y aun eso no ha de
ser muy continuo, porque, si
muchas veces se hiciese, de
pasatiempo se volvería en vicio, y
si pudiésedes acabar con vos de
dexar de todo punto el juego,
sería lo más seguro; pero no
quiero agora apretaros tanto que
con ello quiebre este lance que os
he armado y prisión en que de
vuestra voluntad os vais metido.
Luis.—Pues en pago de vuestra
buena intención, señor Bernardo,
y porque me prometáis de seguir
lo que agora tenéis determinado,
os quiero prestar los treinta
ducados que quedasteis
debiendo, para que, pagándolos,
cumpláis con vuestra fe y palabra.
Bernardo.—Muy gran merced es
la que me hacéis, y de los
primeros que vinieren á mí poder
seréis muy bien pagado dellos.
Antonio.—Con esto nos
podremos ir, que platicando se
nos ha passado el día y yo tengo
mucho que hacer.
Luis.—Pues comenzad á
caminar, que nosotros os
acompañaremos hasta dexaros
en vuestra posada.

Finis.
 COLLOQUIO

En que se trata lo que los

médicos y boticarios están
obligados á hacer para cumplir
con sus oficios, y así mesmo
se ponen las faltas que hay en
ellos para daño de los
enfermos, con muchos avisos
necesarios y provechosos.
Divídese en dos partes: en la
primera se trata lo que toca á
los boticarios, y en la segunda
lo de los médicos.

INTERLOCUTORES

Médico, Licenciado Lerma.

Boticario, Dionisio.—Enfermo, D.
Gaspar.
Caballero, Pimentel.

Lerma.—Dios dé salud á vuestra

merced, mi señor D. Gaspar.
D. Gaspar.—Así haga á vuestra
merced para que en tiempo tan
necesario no me olvide tanto
como hoy lo ha hecho; que si no
fuera con la buena conversación
del señor Pimentel, que me ha
entretenido, muy largo se me
hubiera hecho el día, y aun con el
señor Dionisio no he holgado
poco, porque tiene gran cuidado
de visitarme, y cuando los
médicos se descuidan, es bien
que los boticarios (como uno de
sus miembros) vengan á cumplir
sus faltas con los enfermos.
Lerma.—Buena manera es essa
de reñir conmigo una falta que
hago por no poder hacer menos;
y no la hiciera sino con dexar á
vuesa merced esta mañana en
tan buena disposición, que creo
que debe estar ya sin calentura.
D. Gaspar.—Mejor viva yo que
estoy sin ella.
Lerma.—Muéstreme vuestra
merced el pulso. En verdad que
no es tanta que se pueda decir
calentura, y de aquí á mañana yo
sé cierto que no habrá ninguna.
D. Gaspar.—Menos cuenta tengo
con ella que con este dolor que
siento en el hígado, porque yo os
digo, señor licenciado, que me
atormenta tanto, que le temo, y
esto es lo principal para que yo
querría que me buscásedes
remedio.
Pimentel.—A lo que yo siento,
más debe proceder el accidente
de la calentura del mal que hay
en el hígado que no el mal ó dolor
del hígado de la calentura, y
pocas veces el señor Gaspar
estará sin ella hasta que esté
remediada la causa principal de á
donde se sigue el daño.
Lerma.—Vuestra merced dice
gran verdad, pero, según esto,
Dionisio no ha hecho el emplasto
de melliloto que yo dexé
ordenado, ni vuestra merced lo
debe tener puesto.
Dionisio.—Así es verdad.
Lerma.—¿Pues por qué no se
hizo?
Dionisio.—Porque no ha tantas
horas que vuestra merced lo
ordenó que no se pueda haber
sufrido sin él, como se han
pasado tantos días que el señor
don Gaspar lo hubiera de haber
tenido con otros beneficios que se
le pudieran haber hecho antes de
ahora.
Lerma.—¿Y qué descuidos
parece á vos que se ha tenido en
esso?
Dionisio.—Yo no he visto que
hayan precedido los remedios
universales á los particulares que
agora se hacen; pues no se han
hecho las evacuaciones conforme
á las reglas de medicina, las
cuales han de preceder á las
unciones y emplastos, según la
doctrina de Ipocras en sus
aforismos.
Lerma.—No es malo que queráis
vos haceros dotor en Medicina sin
saber letra della y que os parezca
que estoy yo obligado á sufrir
vuestra desvergüenza de
enmendarme la cura que yo hago.
¿Sabéis vos por ventura la
intención principal que yo he
llevado en ella, y si ha habido
otros accidentes más principales
y que tienen más necesidad de
remediarse?
Dionisio.—Lo que yo sé es que
no está toda la fuerza en el
emplasto para sanar el hígado.
Lerma.—Si no tuviera respeto á
estos señores que están
presentes, yo os respondiera
como vos merecíades; pero assí
no quiero deciros más de que
atendáis á hacer bien lo que toca
á vuestro oficio, y no haréis poco.
Dionisio.—Vuestra merced se ha
apasionado sin razón, y en lo que
toca á mi oficio, yo lo hago de
manera que no hay de qué
reprehenderme.
Lerma.—¿Qué podéis vos hacer
más que los otros boticarios, pues
en fin sois boticario como ellos?
Dionisio.—¿Y qué suelen hacer
los boticarios que no sea muy
bien hecho?
Lerma.—Por vuestra honra
quiero callarlo, y aun por la de los
médicos, pues lo sabemos y no lo
remediamos.
Dionisio.—Si vuestra merced lo
dixese, no faltará para ello
respuesta; pues no es justo que
en esse caso paguen justos por
pecadores.
Pimentel.—Lo que aquí se dixere
no saldrá desta puerta afuera, y
con esta condición, y con que sea
sin ningún enojo, el señor don
Gaspar y yo recebiremos muy
gran merced en que se trate algo
desta materia para satisfacerme
de algunas cosas que me han
puesto duda y sospecha de que
algunos boticarios no cumplen
con el mundo y con Dios lo que
son obligados.
Lerma.—Ningún engaño recibe
vuestra merced en esso, y plega
á Dios que no sean todos los que
esso hacen, y pues que aqui
puede pasar, menester es que
todas sean verdades las que se
dixeren.
Dionisio.—Diga vuestra merced
lo que quisiere, que ninguna pena
recebirá dello con tal que yo sea
también oído antes que la
cuestión se determine, pues estos
señores han de ser jueces della.
D. Gaspar.—Razón tiene Dionisio
en lo que pide.
Lerma.—Yo soy contento de que,
cuando sea tiempo, pueda
replicar y alegar de su derecho. Y
porque vuestras mercedes
entiendan que no me muevo sin
razón á lo que he dicho, sepan
que las condiciones que han de
tener los boticarios escriben
muchos autores, y quien
particularmente las trata, es
Saladino en la primera parte de
su obra; y porque referir todo lo
que dice sería confusión y
prolijidad, diré algunas cosas
dellas. Y lo primero es que el
boticario ha de ser de muy buen
ingenio, hombre sin vicios, sabio y
experimentado en su oficio; no ha
de ser avariento, ni deseoso de
adquirir hacienda; sobre todo ha
de ser muy fiel para que no haga
cosa contra su conciencia, ni por
su parecer, sino con consejo de
médico docto, y que en el precio
de las medicinas sea convenible.
Estas son cosas tan necesarias,
que obligan tanto al boticario á
guardarlas y cumplirlas, que no lo
haciendo, no es poco el daño ni
pocos los inconvenientes que
dello se siguen á los enfermos;
pero yo he hablado sin perjuicio
de los buenos boticarios (que son
tan pocos, que apenas se hallará
uno entre ciento), diré lo que
cerca desto hacen. Lo primero en
lo que toca á ser hombre sabio y
experimentado en su oficio no
tienen ellos toda la culpa, que la
mayor parte se puede dar á los
protomédicos porque examinan y
dan por hábiles y suficientes á
muchos que ni saben ni entienden
qué cosa son medicinas, ni tienen
experiencia dellas ni conocimiento
para alcanzar cuál es una ni cuál
es otra, sino que si van á la feria á
comprar sus drogas, no
solamente se engañan en
distinguir y apartar lo malo de lo
bueno, pero muchas veces toman
uno por otro sin conocerlo, porque
ignoran la condición y calidades
que han de tener para ser aquella
medicina que piensan; y por no se
mostrar ignorantes, quieren más
dexarse engañar de los que los
venden que tomar consejo con
quien podría desengañarlos para
que no errasen.
Pimentel.—Pues, ¿por qué los
protomédicos hacen una cosa tan
fuera de razón como essa?
Lerma.—O por no perder el
interese de los derechos que los
pagan ó porque reciben servicios
con que se obligan á hacer lo que
no deben, y sin esto aprovechan
mucho los favores de personas
señaladas ó de algunos amigos á
quien estiman en más que á las
conciencias, y así veréis que
muchos vienen examinados y con
su carta de examen muy bien
escrita y iluminada, que podrían
con más justa razón traer una
albarda que usar el oficio. Y con
poner sus boticas muy
compuestas con cajas doradas y
botes pintados, y las redomas con
unos rétulos muy grandes, á
muchas gentes hacen entender
que es oro todo lo que reluce, y
que vayan á tomar medicinas á
sus tiendas, que aprovechan más
para enfermar con ellas los sanos
que para dar salud á los
enfermos.
D. Gaspar.—En esto también me
parece que tienen la culpa los
médicos como los boticarios,
pues lo saben y lo permiten.
Lerma.—Yo no quiero excusar á
los que esso hacen.
Dionisio.—Ni podría vuesa
merced hacerlo aunque quisiese,
pero yo lo guardo todo para mi
respuesta, porque no quiero
quebrar el hilo satírico que
vuestra merced lleva tan bien
ordenado.
Lerma.—Bien es que lo hagáis
así, que también, como ya he
dicho, os oiré yo lo que en favor
vuestro y de los boticarios
alegásedes. Y tornando al
propósito, digo que es cosa recia
la desorden que en esto se tiene,
que en una cosa que va la salud y
vida de los hombres, no se ponga
mayor diligencia en conocer á los
que pueden tratar dello.
Pimentel.—¿Y qué se podría
hacer para remediarlo?
Lerma.—No dar el oficio de los
protomédicos á hombres que
hubiesen de llevar derechos ni
dineros algunos á los que
examinaren, porque así cesaría la
codicia y no los cegaría el interés
que se les sigue. Y demás desto
habíanlos de buscar personas
muy santas, temerosas de Dios y
de sus conciencias, para que no
permitiesen que ninguno tratasse
en esta arte que no la entendiese
y supiese muy bien lo que hacía.
D. Gastar.—Harto buena
gobernación sería essa, y aun
bien necesaria, si se hiciesse lo
que decís, y aun las justicias y
regimientos de los pueblos habían
de entender en remediar esta
falta, cuando saben que un
boticario no es bastante, por el
daño que dello se sigue á la