Preface

A biometric system is a technological system that uses information about a person or other biological organism to identify that person. The biometric industry is rapidly changing and progressing at an astonishing speed. What used to be a futuristic concept has become a reality today. In order to work correctly and effectively, biometric systems depend and rely on data about specific biological traits. In the current age and time, biometric systems are of intensive needs and are widely used in various real-life applications. There are a number of potential applications that have different requirements at different places and times. These, for example, include personal recognition, identification, verification, and others. It may be needed for safety, security, permission, banking, crime prevention, forensics, medical applications, communication, face finding, and others.

The increasing trends, needs, and applications of biometric systems make it more prominent to make developments in this direction to achieve more recent and desired objectives. This leads to the idea of capturing, storing, finding, retrieving, analyzing, and using biometrics in everyday life under the computing environment. Being a computer-based technology, biometric systems carry out automatic processing, manipulation, and interpretation of personal information. It plays a significant and important role in various aspects of real life. It is also highly useful in many areas, disciplines, and fields of science and technology.

This book is specifically dedicated to biometric systems, research, applications, techniques, tools, and algorithms that originate from areas such as image processing, computer vision, pattern recognition, signal processing, artificial intelligence, intelligent systems, soft computing in particular, and in the fields of computer engineering, electrical engineering, and computer science in general. This book explores the latest developments, theories, methods, approaches, algorithms, analyses, and systems for advancements in biometrics and related systems. This publication provides an effective platform for helping and guiding readers, professionals, researchers, academicians, engineers, scientists, and policy makers involved in the area of biometrics. It will disseminate information about backgrounds, methodologies, technologies, and systems in this area together with an in-depth discussion of its latest advances.

The main objective of this book is to provide the international community with an effective platform in the area of people identity verification and authentication from physiological and behavioral aspects. It aims to publish the latest developments and insights for biometric innovations, systems, and applications. The book is also targeted to describe the latest emerging settings and requirements in biometric systems and technologies.

Sarfraz begins the book with an introductory chapter on fingerprint biometrics. He describes that, in the current age and time, biometrics have been deployed successfully in various fields of real life. Numerous methods, techniques, and systems in biometrics serve the sciences, security, military, medical area, and human identification. There are various kinds of biometrics being used, these include fingerprint, face, speaker/voice, infrared thermogram (facial, hand or hand vein), gait, keystroke, odor, ear, hand geometry, retina, iris, palmprint, signature, DNA, knuckle crease, brain/EEG, and heart sound/ECG. In recent years, a large amount of research has been undertaken regarding the evolution of biometricbased information on the fingerprint. Fingerprints are a very effective recording of their superior properties such as reliability and accuracy. This chapter presents the fingerprint recognition concept. It highlights and analyses the work done by different authors related to fingerprint recognition. A detailed comparative study is included in this chapter. It concludes with a discussion of future trends.

This is followed by the chapter titled "Biometric Authentication Based on Electrocardiogram" Bogdanov et al, in this chapter, aim to discuss biometric identification based on electrocardiograms. They state that biometric identification is a multi-stage process, including stages such as signal registration, signal preprocessing, extraction of biometric features, assessment of information content, and selection of the most informative features, as well as classification of biometric features. Each of the stages of biometric identification contributes to the final recognition accuracy. This chapter discusses each of the stages of biometric identification based on ECGs. The results discussed, in this chapter, have been obtained using a computational experiment conducted using programs written in Python. At the same time, the authors have used popular libraries, such as sklearn, scipy, wfdb, biosppy, tensorflow, and others. Most of the input data for the computational experiment has been taken from www.physionet.org. The authors have identified the main factors affecting the accuracy of biometric identification using ECGs. They state that traditional password-based authentication methods have a number of disadvantages related primarily to the human factor. Therefore, biometric methods of identification and authentication are much more reliable, although they have some disadvantages. Some of them, like fingerprints, retina, and voice, was compromised. However, it is not clear what to do if hackers gain access to a biometric database, because a person cannot change fingerprints as easily as a forgotten password. The development of wireless technologies and technologies of the Internet of medical things makes possible the emergence of new biometric identification scenarios.

Muqeet and Hameeduddin, in Chapter 3 of the book, follow with a discussion on "Face Identification using LBP-based Improved Directional Wavelet Transform." They assert that face identification is one of the most active areas of research in computer vision and biometric authentication. It is a well-known fact that various face identification methods have been developed over time. But still, numerous facial appearances including facial expression, pose, and illumination variation, have to be dealt with. Moreover, faces captured in unrestrained situations also impose immense concern in designing effective face identification methods. It is desirable to extract robust local descriptive features to effectively characterize such facial variations both in unrestrained and restrained situations. This chapter discusses such a face identification method that incorporates a popular local descriptor such as local binary patterns (LBP) based on the improved directional wavelet transform (IDW) method to extract facial features. This designed method is applied to complex face databases such as CASIA-WebFace and LFW, which consists of a large number of face images collected under an unrestrained environment with extreme facial variations in expression, pose, and illumination. The chapter describes experiments and includes comparisons with various other methods. These include local descriptive methods as well as local descriptive-based multiresolution analysis (MRA) based methods demonstrating the efficacy of the proposed LBP-based IDW method.

**V**

The next chapter, by Liang, et al, is on "Region of Interest Localization Methods for Publicly Available Palmprint Databases". There exist many publicly available palmprint databases that can be found in the current literature. However, not all of them provide the corresponding region of interest (ROI) images. If everyone uses their own extracted ROI images for performance testing, the final accuracy is not strictly comparable. Since ROI localization is the critical stage of palmprint recognition, location precision has a significant impact on the final recognition accuracy. This is very specifically true in unconstrained scenarios. This problem somehow limits the applications of palmprint recognition. However, various published surveys in the current literature focus only on feature extraction and classification methods. Many of the new ROI localization methods have been proposed in recent years. In this chapter, the authors have attempted to group the existing ROI localization methods into different categories. They have analyzed their basic ideas, reproduced some of the codes, made comparisons of their performances, and provided further directions. Hopefully, this chapter would be a

As a representative of biometric technology, palmprint recognition provides a reliable and efficient solution in many authentication scenarios. Palmprint images, even in low resolution, contain rich and discriminative biometric information and have the high antispoof capability, which is suitable for person recognition. Currently, many palmprint acquisition devices have been proposed, but how to design the systems is seldom studied. For example, how to choose the imaging sensor, the lens, and the working distance. Liang et al in Chapter 5, "Image Sharpness based System Design for Touchless Palmprint Recognition", aim to find the relationship between image sharpness and recognition performance. They utilize this information to direct the system design. In this chapter, firstly, the authors introduce the development of recent palmprint acquisition systems, and abstract their basic frameworks to propose the key problems needed to be solved when designing new systems. Secondly, the relationship between the palm distance in the field of view (FOV) and image pixels per inch (PPI) are studied based on the imaging model. They have provided suggestions about how to select the imaging sensor and camera lens. Thirdly, the authors have taken into consideration the image blur and depth of focus (DOF). They have analyzed the recognition performances of the image layers in the Gaussian scale space. Based on this, an image sharpness range is determined for optimal imaging. The experiment results are obtained using different algorithms on various touchless palmprint databases collected using different kinds of devices. These achievements could act as

The second to last chapter of the book is motivated by the Convolutional Neural Network (CNN) on "Transfer Learning of Pre-Trained CNN Models for Fingerprint Liveness Detection." In the current literature, many of the machine learning experts expect that transfer learning will be the next research frontier. Indeed, in the era of deep learning and big data, there are many powerful pre-trained CNN models that have been deployed. Therefore, using the concept of transfer learning, these pre-trained CNN models could be re-trained to tackle a new pattern recognition problem. Samma and Suandi, in this chapter, are aiming to investigate the application of the transferred VGG19-based CNN model to solve the problem of fingerprint liveness recognition. In particular, they have modified the transferred VGG19-based CNN model, re-trained and finely tuned, to recognize real and fake fingerprint images. Moreover, the authors have examined different architectures of the transferred VGG19-based CNN model, including the shallow model,

useful reference for researchers.

references for new system design.

The next chapter, by Liang, et al, is on "Region of Interest Localization Methods for Publicly Available Palmprint Databases". There exist many publicly available palmprint databases that can be found in the current literature. However, not all of them provide the corresponding region of interest (ROI) images. If everyone uses their own extracted ROI images for performance testing, the final accuracy is not strictly comparable. Since ROI localization is the critical stage of palmprint recognition, location precision has a significant impact on the final recognition accuracy. This is very specifically true in unconstrained scenarios. This problem somehow limits the applications of palmprint recognition. However, various published surveys in the current literature focus only on feature extraction and classification methods. Many of the new ROI localization methods have been proposed in recent years. In this chapter, the authors have attempted to group the existing ROI localization methods into different categories. They have analyzed their basic ideas, reproduced some of the codes, made comparisons of their performances, and provided further directions. Hopefully, this chapter would be a useful reference for researchers.

As a representative of biometric technology, palmprint recognition provides a reliable and efficient solution in many authentication scenarios. Palmprint images, even in low resolution, contain rich and discriminative biometric information and have the high antispoof capability, which is suitable for person recognition. Currently, many palmprint acquisition devices have been proposed, but how to design the systems is seldom studied. For example, how to choose the imaging sensor, the lens, and the working distance. Liang et al in Chapter 5, "Image Sharpness based System Design for Touchless Palmprint Recognition", aim to find the relationship between image sharpness and recognition performance. They utilize this information to direct the system design. In this chapter, firstly, the authors introduce the development of recent palmprint acquisition systems, and abstract their basic frameworks to propose the key problems needed to be solved when designing new systems. Secondly, the relationship between the palm distance in the field of view (FOV) and image pixels per inch (PPI) are studied based on the imaging model. They have provided suggestions about how to select the imaging sensor and camera lens. Thirdly, the authors have taken into consideration the image blur and depth of focus (DOF). They have analyzed the recognition performances of the image layers in the Gaussian scale space. Based on this, an image sharpness range is determined for optimal imaging. The experiment results are obtained using different algorithms on various touchless palmprint databases collected using different kinds of devices. These achievements could act as references for new system design.

The second to last chapter of the book is motivated by the Convolutional Neural Network (CNN) on "Transfer Learning of Pre-Trained CNN Models for Fingerprint Liveness Detection." In the current literature, many of the machine learning experts expect that transfer learning will be the next research frontier. Indeed, in the era of deep learning and big data, there are many powerful pre-trained CNN models that have been deployed. Therefore, using the concept of transfer learning, these pre-trained CNN models could be re-trained to tackle a new pattern recognition problem. Samma and Suandi, in this chapter, are aiming to investigate the application of the transferred VGG19-based CNN model to solve the problem of fingerprint liveness recognition. In particular, they have modified the transferred VGG19-based CNN model, re-trained and finely tuned, to recognize real and fake fingerprint images. Moreover, the authors have examined different architectures of the transferred VGG19-based CNN model, including the shallow model,

**IV**

identification scenarios.

proposed LBP-based IDW method.

vein), gait, keystroke, odor, ear, hand geometry, retina, iris, palmprint, signature, DNA, knuckle crease, brain/EEG, and heart sound/ECG. In recent years, a large amount of research has been undertaken regarding the evolution of biometricbased information on the fingerprint. Fingerprints are a very effective recording of their superior properties such as reliability and accuracy. This chapter presents the fingerprint recognition concept. It highlights and analyses the work done by different authors related to fingerprint recognition. A detailed comparative study is

included in this chapter. It concludes with a discussion of future trends.

This is followed by the chapter titled "Biometric Authentication Based on Electrocardiogram" Bogdanov et al, in this chapter, aim to discuss biometric identification based on electrocardiograms. They state that biometric identification is a multi-stage process, including stages such as signal registration, signal preprocessing, extraction of biometric features, assessment of information content, and selection of the most informative features, as well as classification of biometric features. Each of the stages of biometric identification contributes to the final recognition accuracy. This chapter discusses each of the stages of biometric identification based on ECGs. The results discussed, in this chapter, have been obtained using a computational experiment conducted using programs written in Python. At the same time, the authors have used popular libraries, such as sklearn, scipy, wfdb, biosppy, tensorflow, and others. Most of the input data for the computational experiment has been taken from www.physionet.org. The authors have identified the main factors affecting the accuracy of biometric identification using ECGs. They state that traditional password-based authentication methods have a number of disadvantages related primarily to the human factor. Therefore, biometric methods of identification and authentication are much more reliable, although they have some disadvantages. Some of them, like fingerprints, retina, and voice, was compromised. However, it is not clear what to do if hackers gain access to a biometric database, because a person cannot change fingerprints as easily as a forgotten password. The development of wireless technologies and technologies of the Internet of medical things makes possible the emergence of new biometric

Muqeet and Hameeduddin, in Chapter 3 of the book, follow with a discussion on "Face Identification using LBP-based Improved Directional Wavelet Transform." They assert that face identification is one of the most active areas of research in computer vision and biometric authentication. It is a well-known fact that various face identification methods have been developed over time. But still, numerous facial appearances including facial expression, pose, and illumination variation, have to be dealt with. Moreover, faces captured in unrestrained situations also impose immense concern in designing effective face identification methods. It is desirable to extract robust local descriptive features to effectively characterize such facial variations both in unrestrained and restrained situations. This chapter discusses such a face identification method that incorporates a popular local descriptor such as local binary patterns (LBP) based on the improved directional wavelet transform (IDW) method to extract facial features. This designed method is applied to complex face databases such as CASIA-WebFace and LFW, which consists of a large number of face images collected under an unrestrained environment with extreme facial variations in expression, pose, and illumination. The chapter describes experiments and includes comparisons with various other methods. These include local descriptive methods as well as local descriptive-based multiresolution analysis (MRA) based methods demonstrating the efficacy of the

medium model, and deep model. To assess the performances of each architecture, the LivDet2009 database was employed. Reported results indicated that the best recognition rate was achieved from the shallow VGG19-based CNN model with 92% accuracy.

The last chapter is about the "Assessment Methods of Cognitive Ability of Human Brains for Inborn Intelligence Potential using Pattern Recognitions" Raja et al, in this chapter, aim to analyze the scientific study related to fingerprint patterns and the brain lobes. Generally, this method is used to find and develop the inborn potential and personality, especially of children. Every person has inborn potential and personality, which helps to analyze each person's strengths and weaknesses. The work, in this chapter, is based on the analysis. It is mainly for reference purposes for scientific research in the field of Galtian. The statistical study is conducted based on the fingerprint analysis. The study describes that the human brain is divided into two parts: the left hemisphere and the right hemisphere. The fingers of the right hand represent the functions of the left brain and the fingers of the left hand represent the functions of the right brain. The human brain is divided into 10 lobes and each lobe is related with a finger. Each lobe represents different bits of intelligence. The chapter illustrates that the detailed analysis of the fingerprint helps people to find the inborn talents. It provides them with the most appropriate learning habits from a young age and improves learning ability effectively. The vital factor of an individual's intelligence is determined by the neural network connection of brain cells. Cognitive science is a scientific study that helps people to learn about themselves.
