Machine Learning for Biometrics: Concepts,

Algorithms and Applications (Cognitive Data

Science in Sustainable Computing) 1st Edition -
eBook PDF download
https://ebookluna.com/download/machine-learning-for-biometrics-
concepts-algorithms-and-applications-cognitive-data-science-in-
sustainable-computing-ebook-pdf/

Download more ebook from https://ebookluna.com

 We believe these products will be a great fit for you. Click
the link to download now, or visit ebookluna.com
to discover even more!

Introduction to Algorithms for Data Mining and Machine Learning 1st edition
- eBook PDF

https://ebookluna.com/download/introduction-to-algorithms-for-data-mining-
and-machine-learning-ebook-pdf/

(eBook PDF) Machine Learning Refined: Foundations, Algorithms, and

Applications

https://ebookluna.com/product/ebook-pdf-machine-learning-refined-
foundations-algorithms-and-applications/

Cognitive Big Data Intelligence with a Metaheuristic Approach (Cognitive

Data Science in Sustainable Computing) 1st Edition - eBook PDF

https://ebookluna.com/download/cognitive-big-data-intelligence-with-a-
metaheuristic-approach-cognitive-data-science-in-sustainable-computing-
ebook-pdf/

Machine Learning for Planetary Science 1st Edition - eBook PDF

https://ebookluna.com/download/machine-learning-for-planetary-science-
ebook-pdf/
Artificial Intelligence and Machine Learning for EDGE Computing 1st Edition
Rajiv Pandey - eBook PDF

https://ebookluna.com/download/artificial-intelligence-and-machine-
learning-for-edge-computing-ebook-pdf/

Machine Learning and Data Science in the Power Generation Industry: Best
Practices, Tools, and Case Studies 1st Edition - eBook PDF

https://ebookluna.com/download/machine-learning-and-data-science-in-the-
power-generation-industry-best-practices-tools-and-case-studies-ebook-pdf/

Sustainable Developments by Artificial Intelligence and Machine Learning

for Renewable Energies 1st Edition- eBook PDF

https://ebookluna.com/download/sustainable-developments-by-artificial-
intelligence-and-machine-learning-for-renewable-energies-ebook-pdf/

Advances in Subsurface Data Analytics: Traditional and Physics-Based

Machine Learning 1st Edition - eBook PDF

https://ebookluna.com/download/advances-in-subsurface-data-analytics-
traditional-and-physics-based-machine-learning-ebook-pdf/

BIG DATA ANALYTICS: Introduction to Hadoop, Spark, and Machine-Learning 1st

Edition - eBook PDF

https://ebookluna.com/download/big-data-analytics-introduction-to-hadoop-
spark-and-machine-learning-ebook-pdf/
Machine Learning for Biometrics
Concepts, Algorithms, and Applications
Cognitive Data Science in Sustainable Computing

Machine Learning for

Biometrics
Concepts, Algorithms, and Applications

Edited by

Partha Pratim Sarangi

Assistant Professor, Computer Science and Engineering, Seemanta Engineering College,
Jharpokharia, Baripada, Odisha, India

Madhumita Panda
Assistant Professor, Master in Computer Applications, Seemanta Engineering College,
Jharpokharia, Baripada, Odisha, India

Subhashree Mishra
Assistant Professor, School of Electronics Engineering, KIIT University, Bhubaneswar,
Odisha, India

Bhabani Shankar Prasad Mishra

Associate Professor, School of Computer Engineering, KIIT University, Bhubaneswar,
Odisha, India
Dean, School of Computer Engineering, KIIT University, Bhubaneswar, Odisha, India

Banshidhar Majhi
Director, IIITDM, Kancheepuram, Chennai, Tamil Nadu, India

Series Editor
Arun Kumar Sangaiah
School of Computing Science and Engineering,Vellore Institute of Technology (VIT),
Vellore, India
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.

Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
ISBN 978-0-323-85209-8

For information on all Academic Press publications

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

Publisher: Mara Conner

Editorial Project Manager: Andrae Akeh
Production Project Manager: Swapna Srinivasan
Cover Designer: Christian J. Bilbow

Typeset by STRAIVE, India

Contributors

Numbers in paraentheses indicate the pages on which the authors’ contributions begin.
Sumitav Acharya (143), Department of Computer Science and Engineering, National
Institute of Science and Technology, Berhampur, India
A. Anandh (105), Department of Computer Science and Engineering, Kamaraj College
of Engineering and Technology, Madurai, India
Saurabh Bilgaiyan (155), School of Computer Engineering, KIIT Deemed to be
University, Bhubaneswar, Odisha, India
P.V.S.S.R. Chandra Mouli (65), Department of Computer Science, Central University
of Tamil Nadu, Thiruvarur, India
Manisha P. Dale (1), MES College of Engineering, Savitribai Phule Pune University,
Pune, India
Rupam Das (155), School of Electronics Engineering, KIIT, Deemed to be University,
Bhubaneswar, India
K. Devendran (87), Computer Science and Engineering, Kongu Engineering College,
Erode, Tamil Nadu, India
Sachit Dhamija (155), School of Electronics Engineering, KIIT, Deemed to be
University, Bhubaneswar, India
R. Jai Ganesh (129), K. Ramakrishnan College of Technology, Trichy, India
G.K. Kamalam (177), Department of Information Technology, Kongu Engineering
College, Erode, Tamil Nadu, India
Vaishali H. Kamble (1), AISSMS IOIT, Savitribai Phule Pune University, Pune, India
M. Kavitha (129), K. Ramakrishnan College of Technology, Trichy, India
P. Keerthika (87,177), Department of Computer Science and Engineering, Kongu
Engineering College, Erode, Tamil Nadu, India
Chirag Kyal (29), Department of Computer Science and Engineering, National Institute
of Science and Technology, Berhampur, India
K. Logeswaran (177), Department of Information Technology, Kongu Engineering
College, Erode, Tamil Nadu, India
V.M. Manikandan (201), Computer Science and Engineering, SRM University-AP,
Amaravati, Andhra Pradesh, India
R. Manjula Devi (87,177), Department of Computer Science and Engineering, Kongu
Engineering College, Erode, Tamil Nadu, India

xiii
xiv Contributors

Bhabani Shankar Prasad Mishra (47,155), School of Computer Engineering, KIIT

Deemed to be University, Bhubaneswar, India
Subhashree Mishra (47), School of Computer Engineering, KIIT Deemed to be
University, Bhubaneswar, Odisha, India
Samrat Mondal (217), Department of Computer Science and Engineering, Indian
Institute of Technology Patna, Bihta, India
H. Muthukrishnan (177), Department of Information Technology, Kongu Engineering
College, Erode, Tamil Nadu, India
K. Muthulakshmi (105), Department of Computer Science and Engineering, Kamaraj
College of Engineering and Technology, Madurai, India
Madhumita Panda (47), Master of Computer Applications, Seemanta Engineering
College, Jharpokharia, India
Suryakanta Panda (217), Department of Computer Science and Engineering, Indian
Institute of Technology Patna, Bihta, India
Harsh Poddar (29), Department of Computer Science and Engineering, National
Institute of Science and Technology, Berhampur, India
R. Ramya (105), Department of Computer Science and Engineering, Kamaraj College
of Engineering and Technology, Madurai, India
Motahar Reza (29,143), Department of Mathematics, GITAM Deemed to be
University, Hyderabad, India
C. Sagana (87), Computer Science and Engineering, Kongu Engineering College,
Erode, Tamil Nadu, India
M. Sangeetha (87), Computer Science and Engineering, Kongu Engineering College,
Erode, Tamil Nadu, India
Partha Pratim Sarangi (47,87), School of Computer Engineering, KIIT Deemed to be
University, Bhubaneswar, India
Priyanshu Sarmah (155), School of Electronics Engineering, KIIT, Deemed to be
University, Bhubaneswar, India
K. Sentamilselvan (177), Department of Information Technology, Kongu Engineering
College, Erode, Tamil Nadu, India
P. Suresh (87,177), Department of Information Technology, Kongu Engineering
College, Erode, Tamil Nadu, India
S. Venkatesh (105), Microland, Bengaluru, India
Dilip Kumar Yadav (65), Department of Computer Applications, National Institute of
Technology, Jamshedpur, India
Gaurav Yadav (65), Department of Computer Applications, National Institute of
Technology, Jamshedpur, India
Preface

The biometric systems automatically recognize the identity of a person by using

his/her physiological and behavioral characteristics. The improvement in sensor
technology attracts many researchers’ attention to introduce a number of new
biometric traits for providing privacy and security in various applications rang-
ing from industrial to healthcare systems. In the last decades, with the aim of
improving the recognition performance, a large number of machine learning
algorithms have been presented in numerous biometric applications. The signif-
icance of machine learning algorithm is to determine the identity of individuals
by associating biometric patterns with their respective enrolled users. However,
in high-security applications, security and recognition accuracy are major con-
cerns that can be addressed by designing multimodal biometric systems. As the
name depicts, multimodal biometrics incorporates two or more different bio-
metric characteristics (for example, fingerprint, iris, face, plamprint, etc.),
which works in the similar way like unimodal biometrics except an extra mod-
ule of information fusion. Due to the increasing number of biometric systems
and their applications, a profound amount of research and developments are still
required in this field of biometrics.
So far, a numerous biometric systems are exhibiting limited identification
rate or accuracy in unconstrained scenarios due to the poor quality of image
acquisition that includes poor contrast, varying illumination, and
noncooperation by users. These challenges encourage many researchers to work
in its diverse segments such as extracting discriminative features, employing
efficient classifiers, and fusion of information at different stages of the multi-
modal biometric systems. Furthermore, another direction of research is to pro-
vide security to the biometric information system in case of online mode of
identity recognition of individuals in many applications. This book covers
the most recent research progresses in the field of biometrics to enhance recog-
nition performance and security problems in several applications, such as online
transaction, e-commence, access control, law enforcement, boarder security,
healthcare, and so on.
In this book, we specifically address the current research progress in the
field of biometrics and biometric-based applications with the objective of
improving personal identity recognition performance. This book comprises
12 chapters, in which each chapter delivers the concepts and fundamentals of
a recent topic on biometrics, presents reviews on up-to-date methodologies,

xv
xvi Preface

reveals results, and compares with state-of-the-art, highlighting its effectiveness

in the recognition and future directions. Furthermore, the contents of this book
are organized according to the well-accepted biometric traits and a few novel
ones along with biometric security and recent applications, so that each chapter
can be read autonomously from the others. A concise and chapter-wise intro-
duction of this book is described below.
Chapter “Machine Learning Approach for longitudinal face recognition of
children” presents a novel application of identity recognition of children using
the face biometric trait. Face recognition of children below 6 years of age helps
to identify children for their healthcare services and investigation of missing
children. In this chapter, the authors developed a new face database in which
they collected face images of the individual over a period of time, especially
at a gap of few months, and named it as longitudinal face database. Extensive
experiments are performed using different machine learning classifiers such as
support vector machine (SVM), K-nearest neighbor (KNN), logistic regression
(LR), decision tree, Gaussian Naive Bayes, and convolutional neural networks
(CNN). Comparison of experimental results demonstrates the effectiveness and
superiority of the proposed method using CNN with improved accuracy.
Finally, the authors suggest to increase the database size and improvement in
the model as some possible research directions in future.
Chapter “TBFR: Thermal Biometric Face Recognition—A Noncontact Face
Biometry” proposes a novel thermal face recognition method to diminish the
effect of varying light intensity, poses, accessories, aging, etc. of the existing
face biometric systems. In this chapter, the authors fused face geometry features
with pretrained ResNet-50 features to improve recognition performance, and
experimental results revealed the proposed method to be more robust in com-
parison with the existing methods.
Chapter “Multimodal Biometric Recognition using Human Ear and Profile
Face: An Improved Approach” discusses an improved approach for ear biomet-
rics based on fusion of ear and profile face-handcrafted features at the feature-
level fusion and score-level fusion schemes. Recently, some convolutional neu-
ral network models have been proposed for ear biometrics, and their experiment
results show competent performance compared to state-of-the-art methods.
However, deep-learning-based biometrics applications need a large number
of training images, huge memory for data processing, and enormous computa-
tional complexity at the cost of improving the identification rate. In this chapter,
the authors focus on a multimodal approach based on ear and profile face, to
enhance the recognition performance with less processing complexity. Several
experiments with comparisons are performed to test the effectiveness and supe-
riority of the proposed approach.
Chapter “Statistical Measures for Palmprint Image Enhancement” provides
the overview of image enhancement methods and their quality measurements in
palmprint images. In the palmprint image enhancement techniques, the most
widely used databases are discussed with further details. However, the image
 Preface xvii

enhancement plays an important role in the preprocessing step of biometric sys-

tems to provide a better recognition performance in terms of accuracy and
identification rate.
In chapter “Retina Biometrics for Personal Authentication,” an attempt is
made to explore the retina biometrics using the ANFIS model. In this chapter,
ANFIS-based Retina Biometric Authentication System (ARBAS) for person’s
authentication and its significance are discussed. Besides, this chapter provides
readers with the required amount of knowledge to select suitable features’ set
and adequate techniques to develop robust research in this field.
Chapter “Gender Recognition from facial Images using Multi-Channel
Deep Learning Framework” demonstrates Multi-Channel Deep Learning
Framework based on an automated gender recognition approach. In this work,
the authors try to extract discriminative features from raw facial images from
GoogleNet. They fuse feature vectors obtained from two sources: first, feature
vectors are obtained from featured images generated from LDP filters and sec-
ond, feature vectors are directly obtained from raw facial images, and then both
are separately applied as inputs to GoogleNet. Finally, the feature sets are com-
bined together using CCA/DCA to provide input to SVM classifier for gender
recognition of the given input image. A series of experiments are conducted,
and a comprehensive explanation of experimental results is illustrated.
Chapter “Implementation of Cardiac Signal for Biometric Recognition from
Facial Video” discusses the Heartbeat Signal from Facial Video (HSFV) as a
biometric evidence for recognizing the identity of individuals. This chapter
impacts researchers to get motivation and knowledge toward working in a novel
direction in the field of biometrics.
Chapter “Real-Time Emotion Engagement Tracking of Students using
Human Biometric Emotion Intensities” describes an automatic method to detect
an absent-minded face present in a class using FAU (Facial Action Unit). Fur-
thermore, this chapter covers multiple aspects of tracking, ranging from the stu-
dent entering the classroom. Experimental results presented by the authors can
achieve excellent results for emotion detection and recognition of students’ face
images.
Chapter “Facial Identification Expression Based Attendance Monitoring
and Emotion Detection—A Deep CNN Approach” discusses the challenges
of taking attendance and gauging emotions of the students in a new online class-
room environment. In this chapter, with the help of advance machine learning
approach, the authors introduce an automatic biometric-based attendance and
facial expression recognition system.
Chapter “Contemporary Survey on Effectiveness of Machine and Deep
Learning Techniques for Cyber Security” reviews the practicality of biometrics
in cyber security based on efficient machine learning and deep learning tech-
niques. The objective of this survey is to provide an impact on readers to acquire
the fundamental concepts of biometric authentication and identification using
different machine learning techniques for cyber security systems.
xviii Preface

Chapter “A Secure Biometric Authentication System for Smart Environ-

ment using reversible Data Hiding through Encryption Scheme” introduces
the notion of reversible data hiding (RDH) scheme to achieve a secure online
biometric data communication. In a RDH-based encryption scheme, the bio-
metric images are encrypted by hiding the important information in an image.
In this chapter, the author presents a new model wherein the compressed finger-
print data is used as a secret message, and it will be embedded into the face
image through a reversible data hiding-based encryption scheme. In order to
maintain secure communication, encrypted image obtained after RDH scheme
is transmitted to the cloud service provider in a secure manner.
Chapter “An Efficient and Untraceable Authentication Protocol for Cloud
Based Healthcare System” provides Telecare Medical Information System
(TMIS), which is very important in pandemic situations, for example, Covid-
19. In this technique, the patient’s mobile device has a major role for commu-
nicating with doctors. Furthermore, mutual authentication and key agreement
scheme is required to protect vital information from any kind of mischief activ-
ities from adversaries in the insecure communication channel. In this chapter,
the authors propose a biometric hashing technique to provide secure authenti-
cation and key agreement for communication between the patient and
the doctor.
Finally, we hope that our readers including the graduate students, research
scholars, young researchers, academicians, and industrial professionals find the
contributed chapters in this book thought-provoking, and this piece of work will
motivate future research breakthrough to progress further advances in the
machine learning-based biometrics applications.

Partha Pratim Sarangi

Madhumita Panda
Subhashree Mishra
Bhabani Shankar Prasad Mishra
Banshidhar Majhi
Acknowledgments

The editors would like to gratefully acknowledge all of the contributors for con-
tinuous effort and timely submission of their chapters. This book would not
have been feasible without the cooperation of the chapter authors. All the chap-
ters have been reviewed for several rounds to facilitate the selection of final
chapters in our book. Valuable suggestions and guidance from the reviewers
helped the authors in refining individual chapters. Thanks also go to the
reviewers in enhancing the quality of the chapters of this book.
The editors would like to extend gratitude to Sonnini Ruiz Yura from
Elsevier for inspiration over the year. We would like to gratefully acknowledge
Andrae Akeh and Judith Clarisse Punzalan (Elsevier) for their patience during
the preparation of this book. In addition, we shall thank Swapna Srinivasan at
Elsevier for her sincere help and patience during the final preparation of this
book. Finally, we shall extend gratefulness to our family members and friends
for all their support.

xix
Chapter 1

Machine learning approach

for longitudinal face recognition
of children
Vaishali H. Kamblea and Manisha P. Daleb
a
AISSMS IOIT, Savitribai Phule Pune University, Pune, India, bMES College of Engineering,
Savitribai Phule Pune University, Pune, India

1 Introduction
Face of an individual is a popular and well-accepted biometric trait that can be
used to perform identity recognition of adults as well as children. Children are
the most valuable and jeopardy group in society; hence they should be under
supervision continuously. Security and healthcare of children is an important
aspect of all countries [1]. Automatic recognition of children using their face
is a useful investigative tool to help identify missing children. Though the
development of the face of a child starts in the mother’s womb from 3 months,
it is not proportional to the development of other parts of the body. Therefore,
recognition of children below 6 years is still an open research problem. Children
recognition using different modalities needs to be studied to solve the problems
related to security and healthcare. As per the literature survey, still there is not
even one commercial biometric system in use for recognition of toddlers. Var-
ious researchers have discussed about biometric recognition of adults; however,
very few papers on toddlers are available. A meager amount of work has been
traced in biometric identification of toddlers or children. It is most challenging
to recognize a toddler from his own single photograph after a few months. But,
in some instances, such as missing children, we have only the face image. So,
the recognition of children from their face image is very important. Facial
images can be acquired without users’ active involvement using ordinary cam-
eras from a distance. Also in survey, we noted that that most of the toddler’s
biometric recognition is in verification mode.
Sahar Siddique studied longitudinal face recognition using the Extended
Newborn Database and Children Multimodal Biometric Database. Identifica-
tion accuracy achieved using CNN is 62.7% and 85.1% on both the databases,

Machine Learning for Biometrics. https://doi.org/10.1016/B978-0-323-85209-8.00011-0

Copyright © 2022 Elsevier Inc. All rights reserved. 1
2 Machine learning for biometrics

respectively [2]. Rowden et al. studied the longitudinal face recognition of chil-
dren between 0 and 4 years of age. The same-session accuracy they achieved is
93% and cross-session accuracy 43% after 6 months [3]. Local Binary Pattern
(LBP) is widely used texture-based method for recognition of face biometrics
[4]. This method is also used for children face recognition [5,6].
This chapter is presenting a machine learning and deep learning approach
for face recognition of children. As per the literature review, the study of chil-
dren recognition with the help of their face modality started in 2010 in India
[5,6]. Most of the papers studied same-session face recognition. The meager
amount of work is carried out longitudinally, that is, images of the same subject
taken over a period of time. There are very few readily available databases for
newborns like FG-NET [7]. There are very few images of children below
5 years in these datasets. CMBD database of IIIT Delhi (India) [8] is one of
the children longitudinal face recognition databases, but due to security reasons
the face database is not publicly available [9]. Therefore, database collection is
the major task in infants and toddler’s recognition. For this study, the database
of 81 subjects for the same session and 48 subjects for the cross session is col-
lected. The span between the data acquisition sessions is 3–6 months. The major
stages of the proposed work are preprocessing of face images, which include
manually cropping of face images of size 120
120 and converting it into gray
scale. Feature extraction is based on principal component analysis and linear
discriminant analysis, and CNN is proposed. The classification of subjects is
done using machine learning classifiers on the children database of the same
session and cross session. Further, the work is extended using convolutional
neural network (CNN) in which we have proposed our own optimized model
with data augmentation used to compare the machine learning and deep learn-
ing classifiers on our database.
The major contributions of the proposed work are as follows.
1. Due to the nonavailability of reference databases for children below 6 years,
the collection of longitudinal databases of children faces itself is a great
challenge. In the proposed work, we have captured the face images of tod-
dlers with the mobile camera of resolution 20 MP, with consent from their
parents in two sessions. Time period between two sessions is 2 months to
1 year. In the same session, 730 images of 81 toddlers are taken. In the sec-
ond session, 485 images of 48 toddlers are captured.
2. In this chapter, we are proposing feature extraction using PCA, LDA, and
CNN approaches on the facial images and their comparative study. In PCA
and LDA approaches, we have applied different machine learning tech-
niques for classification of subjects, such as Support Vector Machine
(SVM), Logistic Regression (LR), Gaussian Naive Bayes, K-Nearest
Neighbors (KNN), Decision Tree, and Random Forest. In CNN, feature
extraction is done by convolution layers and classification is done by dense
layers.
 Machine learning approach Chapter 1 3

3. In this work, we implemented CNN from basic level to optimized level.

First, CNN is of three convolution layers for the same session and cross ses-
sion. Further, the CNN is modified in terms of data augmentation, and six-
layer convolution model is used with L2 kernel regularization and batch
normalization for enhancing the accuracy.
4. We compare the effectiveness of different feature extraction and classifica-
tion techniques in identification mode. Further, we optimize the CNN for
our children database to increase the accuracy. We compare the recognition
performance of the proposed CNN with the conventional PCA- and LDA-
based method using classification technique.

2 Face modality for children face recognition

Human beings can be recognized using its inherent biological feature’s analysis
and measurement. Science of recognizing human beings with the help of its bio-
logical features is called biometrics.
Biometric recognition is based on two prominent characteristics of human
beings:
(1) Physiological characteristics: face, fingerprint, iris, palmprint, footprint,
palm vein, ear, DNA, etc.
(2) Behavioral characteristics: voice, gait, signature, etc.
Indian biometrics market survey graph about the usage of different biometric
modalities for the year 2019 is shown in Fig. 1.
As per biometric market fingerprint is widely used for recognition of human
beings. Children’s fingerprints are tiny, and are not easy to capture. The second-
largest used biometric modality is face. Face images are easy to capture and

Indian Biometrics Market In US $ Million, 2019

54.8
27.8 80.5
307.4
Fingerprint Recognition
Facial Authentication
IRIS Recognition
Voice Recognition
Palm Recognition
Others
448.2
1679.4

FIG. 1 Indian biometric market share in US million [10].

4 Machine learning for biometrics

useful in the case of missing children. Therefore, we are using face biometrics
for recognition of children. An example of different biometrics of children from
our database is shown in Fig. 2.
Human face is a 3D model, we can recognize it by its features such as eyes,
shape of face, and color. Automatic face recognition is based on 2D photo-
graphs of a person. In 1964 and 1965, Woody Bledsoe, along with Helen Chan
and Charles Bisson, recognized human faces using computers for the first time.
Nowadays adult face recognition achieves very high accuracy in the range of
99.63% [11]. Face recognition of children is still an open challenge. All face
recognition algorithms evaluate on false negative identification rate (FNIR)
which is dependent on age. The FNIR of children is higher as compared to
adults or seniors. The comparison of FNIR and false positive identification rate
(FPIR) is given in Table 1 for various stages of child age.
Child face development: Studies from the paper by Farkas discuss that in
the first 2 years of a child face development, mouth width of child increases
whereas mouth height decreases. Mouth shape alters from “rosebud-like” to
a more adult type. Growth of facial features is very fast in the first year, less
rapid in the second year. Subsequent changes were slow and irregular from
the age of 3 to 9 years [12].

Face Footprint Fingerprint

FIG. 2 Different biometrics for recognition of children.

TABLE 1 Comparison of FNIR and FPIR of children for

different age.

Subjects FNIR FPIR

Babies 0.7 0.2
Kids 0.4 In between 0.2 and 0.05
Pre-teen 0.29 0.05
Discovering Diverse Content Through
Random Scribd Documents
to the end, its interest never flags. It is full of beautiful melody
charged with sentiment and passion, yet varied with lighter
passages, like Butterfly's "I am like the moon's little goddess"; "I
used to think if anyone should want me"; and the exquisite,
"Vogliatemi bene" (Ah, love me a little). There is a beautiful melody
for Pinkerton, "Love, what fear holds you trembling." The climax of
the love duet is reached in two impassioned phrases:—"Dolce notte!
Quante stelle" (Night of rapture, stars unnumbered),

[Listen]

and "Oh! Quanti occhi fisi, attenti" (Oh, kindly heavens).

[Listen]

Act II. Part I. Three years have elapsed. It is a long time since
Pinkerton has left Butterfly with the promise to return to her "when
the robins nest." When the curtain rises, after an introduction, in
which another Japanese theme is employed, Suzuki, although
convinced that Pinkerton has deserted her mistress, is praying for his
return. Butterfly is full of faith and trust. In chiding her devoted maid
for doubting that Pinkerton will return, she draws in language and
song a vivid picture of his home-coming and of their mutual joy
therein:—"Un bel dì vedremo" (Some day he'll come).
 [Listen]

In point of fact, Pinkerton really is returning to Nagasaki, but with no

idea of resuming relations with his Japanese wife. Indeed, before
leaving America he has written to Sharpless asking him to let
Butterfly know that he is married to an American wife, who will join
him in Nagasaki. Sharpless calls upon Butterfly, and attempts to
deliver his message, but is unable to do so because of the emotions
aroused in Butterfly by the very sight of a letter from Pinkerton. It
throws her into a transport of joy because, unable immediately to
grasp its contents, she believes that in writing he has remembered
her, and must be returning to her. Sharpless endeavours to make the
true situation clear to her, but is interrupted by a visit from
Yamadori, a wealthy Japanese suitor, whom Goro urges Butterfly to
marry. For the money left by Pinkerton with his little Japanese wife
has dwindled almost to nothing, and poverty stares her in the face.
But she will not hear of an alliance with Yamadori. She protests that
she is already married to Pinkerton, and will await his return.
When Yamadori has gone, Sharpless makes one more effort to open
her eyes to the truth. They have a duet, "Ora a noi" (Now at last), in
which he again produces the letter, and attempts to persuade her
that Pinkerton has been faithless to her and has forgotten her. Her
only reply is to fetch in her baby boy, born since Pinkerton's
departure. Her argument is, that when the boy's father hears what a
fine son is waiting for him in Japan, he will hasten back. She sings to
Trouble, as the little boy is called:—"Sai cos'ebbe cuore" (Do you
hear, my sweet one, what that bad man is saying). Sharpless makes
a final effort to disillusion her, but in vain. If Pinkerton does not
come back, there are two things, she says, she can do—return to
her old life and sing for people, or die. She sings a touching little
lullaby to her baby boy, Suzuki twice interrupting her with the
pathetically voiced exclamation, "Poor Madam Butterfly!"
A salute of cannon from the harbour announces the arrival of a man-
of-war. Looking through the telescope, Butterfly and Suzuki discover
that it is Pinkerton's ship, the "Abraham Lincoln." Now Butterfly is
convinced that Sharpless is wrong. Her faith is about to be
rewarded. The man she loves is returning to her. The home must be
decorated and made cheerful and attractive to greet him. She and
Suzuki distribute cherry blossoms wherever their effect will be most
charming. The music accompanying this is the enchanting duet of
the flowers, "Scuoti quella fronda di ciliegio" (Shake that cherry tree
till every flower). Most effective is the phrase, "Gettiamo a mani
piene mammole e tuberose" (In handfuls let us scatter violets and
white roses.)

[Listen]

Butterfly adorns herself and the baby boy. Then with her fingers she
pierces three holes in the paper wall of the dwelling. She, Suzuki,
and the baby peer through these, watching for Pinkerton's arrival.
Night falls. Suzuki and the boy drop off to sleep. Butterfly rigid,
motionless, waits and watches, her faith still unshaken, for the
return of the man who has forsaken her. The pathos of the scene is
profound; the music, with the hum of voices, borne upon the night
from the distant harbour, exquisite.
Act II. Part II. When the curtain rises, night has passed, dawn is
breaking. Suzuki and the baby are fast asleep, but Butterfly still is
watching. Again Puccini employs a Japanese melody (the "vigil"
theme).

[Listen]

When Suzuki awakes, she persuades the poor little "wife" to go

upstairs to rest, which Butterfly does only upon Suzuki's promise to
awaken her as soon as Pinkerton arrives. Pinkerton and Sharpless
appear. Suzuki at first is full of joyful surprise, which, however, soon
gives way to consternation, when she learns the truth. Pinkerton
himself, seeing about him the proofs of Butterfly's complete loyalty
to him, realizes the heartlessness of his own conduct. There is a
dramatic trio for Pinkerton, Sharpless, and Suzuki. Pinkerton, who
cannot bear to face the situation, rushes away, leaving it to
Sharpless to settle matters as best he can.
Butterfly has become aware that people are below. Suzuki tries to
prevent her coming down, but she appears radiantly happy, for she
expects to find her husband. The pathos of the scene in which she
learns the truth is difficult to describe. But she does not burst into
lamentations. With a gentleness which has been characteristic of her
throughout, she bears the blow. She even expresses the wish to
Kate, Pinkerton's real wife, that she may experience all happiness,
and sends word to Pinkerton that, if he will come for his son in half
an hour, he can have him.
Sharpless and Mrs. Pinkerton withdraw. In a scene of tragic power,
Butterfly mortally wounds herself with her father's sword, the blade
of which bears the inscription, "To die with honour when one can no
longer live with honour," drags herself across the floor to where the
boy is playing with his toys and waving a little American flag, and
expires just as Pinkerton enters to take away the son whom thus she
gives up to him.
From examples that already have been given of modern Italian
opera, it is clear that "atmosphere," local colour, and character
delineation are typical features of the art of Italy's lyric stage as it
flourishes today. In "Madama Butterfly" we have exotic tone colour
to a degree that has been approached but not equalled by Verdi in
"Aïda." Certain brief scenes in Verdi's opera are Egyptian in tone
colour. In "Madama Butterfly" Japanese themes are used in extenso,
and although the thrilling climaxes in the work are distinctively
Italian, the Japanese under-current, dramatic and musical, always is
felt. In that respect compare "Madama Butterfly" with a typical old
Italian opera like "Lucia di Lammermoor" the scene of which is laid
in Scotland, but in which there is nothing Scotch save the costumes
—no "atmosphere," no local colour. These things are taken seriously
by modern Italian composers, who do not ignore melody, yet also
appreciate the value of an eloquent instrumental support to the
voice score; whereas the older Italian opera composers were content
to distribute melody with a lavish hand and took little else into
account.
In character delineation in the opera Butterfly dominates. She is a
sweet, trusting, pathetic little creature—traits expressed in the music
as clearly as in the drama. The sturdy devotion of Suzuki is, if
possible, brought out in an even stronger light in the opera than in
the drama, and Sharpless is admirably drawn. Pinkerton, of course,
cannot be made sympathetic. All that can be expected of him is that
he be a tenor, and sing the beautiful music allotted to him in the first
act with tender and passionate expression.
The use of the "Star-Spangled Banner" motif as a personal theme for
Pinkerton, always has had a disagreeable effect upon me, and from
now on should be objected to by all Americans. Some one in
authority, a manager like Gatti-Casazza, or Ricordi & Co.'s American
representatives, should call Puccini's attention to the fact that his
employment of the National Anthem of the United States of America
in "Madama Butterfly" is highly objectionable and might, in time,
become offensive; although no offence was meant by him.
I "did" the first night of David Belasco's play "Madam Butterfly" for
the New York Herald. The production occurred at the Herald Square
Theatre, Broadway and Thirty-fifth Street, New York, March 5, 1900,
with Blanche Bates as Butterfly. It was given with "Naughty
Anthony," a farce-comedy also by Belasco, which had been a failure.
The tragedy had been constructed with great rapidity from John
Luther Long's story, but its success was even swifter. At the Duke of
York's Theatre, London, it was seen by Francis Nielsen, stage
manager of Covent Garden, who immediately sent word to Puccini
urging him to come from Milan to London to see a play which, in his
hands, might well become a successful opera. Puccini came at once,
with the result that he created a work which has done its full share
toward making the modern Italian lyric stage as flourishing as all
unprejudiced critics concede it to be.
The Milan production of "Madama Butterfly" was an utter failure. The
audience hooted, the prima donna was in tears. The only person
behind the scenes not disconcerted was the composer, whose faith
in his work was so soon to be justified.

LA FANCIULLA DEL WEST

(THE GIRL OF THE GOLDEN WEST)

Opera in three acts by Puccini; words by C. Zangarini

and G. Civini, after the play by David Belasco.
Produced, Metropolitan Opera House, New York,
December 10, 1910, with Destinn, Mattfeld, Caruso,
Amato, Reiss, Didur, Dinh-Gilly, Pini-Corsi, and De
Segurola.
Characters
Minnie Soprano
Jack Rance, sheriff Baritone
Dick Johnson (Ramerrez) Tenor
Nick, bartender at the "Polka" Tenor
Ashby, Wells-Fargo agent Bass
Sonora } Baritone
Trim } Tenor
Sid } Baritone
Handsome } Baritone
Miners
Harry } Tenor
Joe } Tenor
Happy } Baritone
Larkens } Bass
Billy Jackrabbit, an Indian redskin Bass
Mezzo-
Wowkle, Billy's squaw
Soprano
Jake Wallace, a travelling camp
Baritone
minstrel
José Castro, a greaser from
Bass
Ramerrez's gang
A Postillion Tenor
Men of the Camp

Time—1849-1850, the days of the gold fever.

Place—A mining-camp at the foot of the Cloudy
Mountains, California.
 Photo by White

Destinn as Minnie, Caruso as Johnson, and

Amato as Jack Rance in
“The Girl of the Golden West”

Successful in producing "atmosphere" in "La Bohème," "Tosca," and

"Madama Butterfly," Puccini has utterly failed in his effort to do so in
his "Girl of the Golden West." Based upon an American play, the
scene laid in America and given in America for the first time on any
stage, the opera has not been, the more's the pity, a success.
In the first act, laid in the "Polka" bar-room, after a scene of
considerable length for the miners (intended, no doubt, to create
"atmosphere") there is an episode between Rance and Minnie, in
which it develops that Rance wants to marry her, but that she does
not care for him. Johnson comes in. He and Minnie have met but
once before, but have been strongly attracted to each other. She
asks him to visit her in her cabin, where they will be undisturbed by
the crowd, which has gone off to hunt for Ramerrez, head of a band
of outlaws, reported to be in the vicinity but which soon may be
back.
The scene of the second act is Minnie's cabin, which consists of a
room and loft. After a brief scene for Billy and Wowkle, Minnie
comes in. Through night and a blizzard Johnson makes his way up
the mountainside. There is a love scene—then noises outside. People
are approaching. Not wishing to be found with Johnson, Minnie
forces him to hide. Rance and others, who are on the trail of
Ramerrez and hope to catch or kill him any moment, come in to
warn her that Johnson is Ramerrez. When they have gone, and
Johnson acknowledges that he is the outlaw, Minnie denounces him
and sends him out into the blizzard. There is a shot. Johnson, sorely
wounded, staggers into the cabin. A knock at the door. Rance's
voice. With Minnie's aid the wounded man reaches the loft where he
collapses.
Rance enters, expecting to find Johnson. He is almost persuaded by
Minnie that the fugitive is not there, when, through the loose
timbers of the loft, a drop of blood falls on his hand. Minnie
proposes that they play cards—Johnson to live, or she to marry the
sheriff. They play. She cheats, and wins.
The third act is laid in the forest. Johnson, who has recovered and
left Minnie's cabin, is caught, and is to be hung. But at the critical
moment Minnie arrives, and her pleading moves the men to spare
him, in spite of Rance's protests. They leave to begin a new life
elsewhere.
In the score there is much recitative. It is not interesting in itself, nor
is it made so by the insufficiently varied instrumental
accompaniment. For the action of the play is too vigorous to find
expression by means of the Debussyan manner that predominates in
the orchestra. The most genuinely inspired musical number is
Johnson's solo in the last act, when it seems certain that he is about
to be executed.—"Ch'ella mi creda libero e lontano" (Let her believe
that I have gained my freedom).

LA RONDINE
THE SWALLOW

The opera begins in Paris during the Second Empire. Magda, the
heroine, is a demi-mondaine living under the protection of the rich
banker Rambaldo. Satisfied with the luxuries he lavishes upon her,
she longs for true affection, and is unable to stifle the remembrance
of her first love, a poor young student. She meets Ruggero, who like
her earlier love, is young and poor, and a student. At Bouilliers, the
rendezvous of the gay life of Paris, Ruggero declares his love for
Magda. They leave Paris for Nice, where they hope to lead an idyllic
existence.
Ruggero looks forward to a life of perfect happiness. He writes to his
parents asking their consent to his marriage with Magda. The reply
is that if she is virtuous and honourable, she will be received with
open arms. Magda now considers herself (like Violetta in "La
Traviata") unworthy of Ruggero's love and lest she shall bring
dishonour upon the man she loves, she parts with him. Other
principal rôles are Lisetta and Prunia, and there are numerous
second parts requiring first-rate artists.
In the second act of "La Rondine" is a quartet which, it is said,
Puccini believes will rival that at the end of the third act in "La
Bohème." "I have let my pen run," he is reported to have said, "and
no other method suffices to obtain good results, in my opinion. No
matter what marvellous technical effects may be worked up by
lengthy meditation, I believe in heart in preference to head."
The opera was produced in March, 1917, in Monte Carlo, and during
the summer of the same year, in Buenos Aires. Puccini intended to
compose it with dialogue as a genuine opéra comique, but finally
substituted recitative. The work is said to approach opéra comique in
style. Reports regarding its success vary.
After the first Italian performance, San Carlo Theatre, Naples,
February 26, 1918, Puccini, according to report, decided to revise
"La Rondine." Revision, as in the case of "Madama Butterfly," may
make a great success of it.

ONE-ACT OPERAS
Three one-act operas by Puccini have been composed for
performance at one sitting. They are "Suor Angelica" (Sister
Angelica), "Il Tabarro" (The Cloak), and "Gianni Schicchi." The motifs
of these operas are sentiment, tragedy, and humour.
The scene of "Suor Angelica" is laid within the walls of a mountain
convent, whither she has retired to expiate an unfortunate past. Her
first contact with the outer world is through a visit from an aunt,
who needs her signature to a document. Timidly she asks about the
tiny mite, whom she was constrained to abandon before she entered
the convent. Harshly the aunt replies that the child is dead. Sister
Angelica decides to make an end to her life amid the flowers she
loves. Dying, she appeals for pardon for her act of self-destruction.
The doors of the convent church open, and a dazzling light pours
forth revealing the Virgin Mary on the threshold surrounded by
angels, who, intoning a sweet chorus, bear the poor, penitent, and
weary soul to eternal peace. This little work is entirely for female
voices.
The libretto of "Il Tabarro" is tragic. The great scene is between a
husband and his wife. The husband has killed her lover, whose body
he shows to his unfaithful wife, lifting from the ground the cloak (il
tabarro) under which it is hidden.
The scene of "Il Tabarro" is laid on the deck of a Seine barge at
sunset, when the day's work is over, and after dark. The husband is
Michele, the wife Giorgetta, the lover, Luigi, and there are two other
bargemen. These latter go off after the day's work. Luigi lingers in
the cabin. He persuades Giorgetta that, when all is quiet on the
barge, and it will be safe for him to return to her, she shall strike a
match as a signal. He then goes.
Michele has suspected his wife. He reminds her of their early love,
when he sheltered her under his cloak. Giorgetta, however, receives
these reminiscences coldly, feigns weariness, and retires to the
cabin.
It has grown dark. Michele lights his pipe. Luigi, thinking it is
Giorgetta's signal, clambers up the side of the barge, where he is
seized and choked to death by Michele, who takes his cloak and
covers the corpse with it.
Giorgetta has heard sounds of a struggle. She comes on deck in
alarm, but is somewhat reassured, when she sees Michele sitting
alone and quietly smoking. Still somewhat nervous, however, she
endeavours to atone for her frigidity toward him, but a short time
before, by "making up" to him, telling him, among other things, that
she well recalls their early love and wishes she could again find
shelter in the folds of his big cloak. For reply, he raises the cloak,
and lets her see Luigi's corpse.
I have read another synopsis of this plot, in which Michele forces his
wife's face close to that of her dead lover. At the same moment, one
of the other bargemen, whose wife also had betrayed him, returns
brandishing the bloody knife, with which he has slain her. The
simpler version surely is more dramatic than the one of cumulative
horrors.
When the action of "Gianni Schicchi" opens one Donati has been
dead for two hours. His relations are thinking of the will. A young
man of the house hands it to his mother but exacts the promise that
he shall marry the daughter of neighbour Schicchi. When the will is
read, it is found that Donati has left his all to charity. Schicchi is
called in, and consulted. He plans a ruse. So far only those in the
room know of Donati's demise. The corpse is hidden. Schicchi gets
into bed, and, when the Doctor calls, imitates the dead man's voice
and pretends he wants to sleep. The lawyer is sent for. Schicchi
dictates a new will—in favour of himself, and becomes the heir, in
spite of the anger of the others.
Riccardo Zandonai

FRANCESCA DA RIMINI
FRANCESCA OF RIMINI
Opera in four acts, by Riccardo Zandonai; words by
Tito Ricordi, after the drama of the same title by
Gabriele d'Annunzio. English version from Arthur
Symons's translation of the drama. Produced, Reggio
Theatre, Turin, February 1, 1914. Covent Garden
Theatre, London, July 16, 1914. Metropolitan Opera
House, New York, December 22, 1916, with Alda
(Francesca), Martinelli (Paolo), and Amato (Giovanni).
Characters

Giovanni, the
} { Baritone
lame
sons of
Paolo, the
} Malatesta da { Tenor
beautiful
Verrucchio
Malatestino, the
} { Tenor
one-eyed
Ostasio, son of Guido Minore da
Baritone
Polenta
Ser Toldo Berardengo, a notary Tenor
A Jester Bass
A Bowman Tenor
Tower Warden Baritone
Francesca, daughter of Guido and
Soprano
sister of Ostasio
Samaritana, sister of Francesca and
Soprano
Ostasio
Biancofiore } { Soprano
Garsenda } women of { Soprano
Francesca Mezzo-
Altichiara } {
Soprano
Mezzo-
Donella
Soprano
Smaradi, a slave Contralto
 Bowmen, archers, and musicians.
Time—Thirteenth century.
Place—First act, Ravenna, then Rimini.

A PRETENTIOUS but not wholly successful score based upon a

somewhat diffuse drama—such is the net impression made by
Zandonai's opera "Francesca da Rimini." The story of Francesca and
Paolo is one of the world's immortal tales of passion, and an opera
set to it should be inspired beyond almost any other. But as W.J.
Henderson wrote in the New York Sun the day after the production
of Zandonai's work in New York, "In all human probability the full
measure of 'love insatiable' was never taken in music but once, and
we cannot expect a second 'Tristan und Isolde' so soon."
Act I. The scene is a court in the house of the Polentani, in Ravenna,
adjacent to a garden, whose bright colours are seen through a
pierced marble screen. A colloquy between Francesca's brother
Ostasio and the notary Ser Toldo Berardengo informs us that for
reasons of state, Francesca is to be married to that one of the three
sons of Malatesta da Verrucchio, who although named Giovanni, is
known as Gianciotto, the Lamester, because of his deformity and
ugliness. As Francesca surely would refuse to marry Gianciotto, a
plot has been formed by which she is introduced to his handsome
younger brother Paolo, with whom, under the impression that he is
her destined bridegroom, she falls deeply in love at first sight, a
passion that is fully reciprocated by him, although they have only
beheld each other, and not yet exchanged a word.
Such is the procedure of the first act. When Francesca and Paolo
behold each other through the marble screen, which divides the
court from the garden, in which Paolo stands amid brightly coloured
flowers, the orchestra intones a phrase which may properly be called
the love motif.
 [Listen]

The act is largely lyric in its musical effect. Much charm is given to it
by the quartette of women who attend upon Francesca. Almost at
the outset the composer creates what might be called the necessary
love mood, by a playful scene between Francesca's women and a
strolling jester, who chants for them the story of "Tristan und
Isolde." The setting of the scene is most picturesque. In fact
everything in this act tends to create "atmosphere," and were the
rest of the opera as successful, it would be one of the finest works
of its kind to have come out of modern Italy.
Act II. The scene is the interior of a round tower in the fortified
castle of the Malatestas. The summit of the tower is crowned with
engines of war and arms. There are heavy cross-bows, ballistas, a
catapult, and other mediæval machinery of battle. The castle is a
stronghold of the Guelfs. In the distance, beyond the city of Rimini,
are seen the battlements of the highest Ghibelline Tower. A narrow
fortified window looks out on the Adriatic.
Soon after the act opens, an attack takes place. The battle rages.
Amid all this distracting, and therefore futile tumult, occurs the first
meeting between Francesca and Paolo, since the marriage into
which she was tricked. Their love is obvious enough. Paolo
despairingly seeks death, to which Francesca also exposes herself by
remaining on the platform of the tower during the combat. The
relation between these two principal characters of the opera is
clearly enough set forth, and the impression made by it would be
forcible, were not attention distracted by the fiercely raging
mediæval combat.
The Malatestas are victorious. The attacking foes are driven off.
Gianciotto comes upon the platform and brings news to Paolo of his
election as Captain of the people and Commune of Florence, for
which city Paolo departs.

Photo by White

Alda as Francesca and Martinelli as Paolo in

“Francesca da Rimini”

Act III. The scene is the beautiful apartment of Francesca, where,

from an old tome, she is reading to her women the story of Lancelot
and Guenevere. This episode has somewhat of the same charm as
that which pervaded portions of the first act. Especially is this true,
when to the accompaniment of archaic instruments, the women sing
their measures in praise of spring, "Marzo è giunto, e Febbraio gito
se n'è col ghiado" (March comes, and February goes with the wind
today).

[Listen]

The women dance and sing, until on a whispered word from her
slave, Francesca dismisses them. Paolo has returned. The greeting
from her to him is simple enough: "Benvenuto, signore mio cognato"
(Welcome my lord and kinsman), but the music is charged with
deeper significance.

[Listen]

Even more pronounced is the meaning in the musical phrase at

Francesca's words, "Paolo, datemi pace" (Paolo, give me peace).
 [Listen]

Together they read the story which Francesca had begun reading to
her women. Their heads come close together over the book. Their
white faces bend over it until their cheeks almost touch; and when
in the ancient love tale, the queen and her lover kiss, Francesca's
and Paolo's lips meet and linger in an ecstasy of passion.
Act IV. This act is divided into two parts. The scene of the first part
is an octagonal hall of gray stone. A grated door leads to a
subterranean prison. Cries of a prisoner from there have disturbed
Francesca. When she complains of this to the youngest brother of
Gianciotto, Malatestino, he goes down into the prison and kills the
captive. The introduction to this act is, appropriately enough, based
on an abrupt phrase.

[Listen]

Malatestino is desperately in love with Francesca, urges his suit upon

her, and even hints that he would go to the length of poisoning
Gianciotto. Francesca repulses him. Out of revenge he excites the
jealousy of Gianciotto by arousing his suspicions of Paolo and
Francesca, pointing out especially that Paolo has returned from
Florence much sooner than his duties there would justify him in
doing.
The scene of part two is laid in Francesca's chamber. It is night. Four
waxen torches burn in iron candlesticks. Francesca is lying on the
bed. From her sleep she is roused by a wild dream that harm has
come to Paolo. Her women try to comfort her. After an exchange of
gentle and affectionate phrases, she dismisses them.
A light knocking at the door, and Paolo's voice calling, "Francesca!"
She flings open the door and throws herself into the arms of her
lover. There is an interchange of impassioned phrases. Then a
violent shock is heard at the door, followed by the voice of
Gianciotto, demanding admission. Paolo spies a trap door in the floor
of the apartment, pulls the bolt, and bids Francesca open the door of
the room for her husband, while he escapes.
Gianciotto rushes into the room. Paolo's cloak has caught in the bolt
of the trap door. He is still standing head and shoulders above the
level of the floor. Seizing him by the hair, the Lamester forces him to
come up. Paolo unsheathes his dagger. Gianciotto draws his sword,
thrusts at Paolo. Francesca throws herself between the two men,
receives the thrust of her husband's sword full in the breast, and
falls into Paolo's arms. Mad with rage, her deformed husband with
another deadly thrust pierces his brother's side. Paolo and Francesca
fall at full length to the floor. With a painful effort, Gianciotto breaks
his bloodstained sword over his knee.
Where the drama is lyric in character, and where it concentrates
upon the hot-blooded love story, a tradition in the Malatesta family,
and narrated by a Malatesta to Dante, who, as is well known, used it
in his "Inferno," the music is eloquent. Where, however, the action
becomes diffuse, and attention is drawn to subsidiary incidents, as is
far too often the case, interest in the music flags. With great benefit
to the score at least a third of the libretto could be sacrificed.
Riccardo Zandonai was born at Sacco. He studied with Gianferrai and
at the Rossini Conservatory. "Conchita," another opera by him, Milan,
1912, was produced in this country in Chicago and New York in
1913.
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

ebookluna.com