Название: Machine Learning and Biometrics
Автор: Adele Kuzmiakova
Издательство: Arcler Press
Год: 2022
Страниц: 248
Язык: английский
Формат: pdf (true)
Размер: 10.2 MB

This comprehensive guide provides a detailed overview of modern biometrics, which allows a person to be identified and authenticated based on recognizable, unique, and verifiable data. Biometrics technologies include detection of dormant fingerprints, iris, gait, or facial and voice recognition. Today, biometrics powers cutting-edge security algorithms. Biometrics is used for access into banks, airports or personal smartphones, which means that your money and personal information can be stored safely. Each chapter starts with a comprehensive review of the biometrics technology and its algorithmic description, describes how it works, and outlines all of the applicable and modern use cases of that technology. This book will be an invaluable companion guide to students wishing to become system designers, micro-engineers, security algorithms creators, security experts, and electronic security system manufacturers working on controls or microchips.

This textbook provides a comprehensive outline of machine learning designs for biometric and perceptual tasks. The book synthesizes cutting-edge research on applications of CNN (convolutional neural networks) in fingerprint, iris, face, and vascular biometric systems, and soft biometric surveillance systems. MIL-STD-1553 remains the principal network for military avionics integration, despite the advent of higher-speed solutions, such as fiber channel loom. Biometric security issues are also discussed.

Machine Learning has been employed to tackle several exciting yet difficult real-world problems, with biometrics being one of the most popular. This book offers several new biometrics and Machine Learning approaches and recommendations for using Machine Learning techniques in biometrics. This book covers a variety of “Biometrics” and “Machine Learning” core principles. This book discusses how machine learning may be used to improve biometrics identification performance across an extensive range of biometrics modalities.