Название: Measurements and Instrumentation for Machine Vision
Автор: Oleg Sergiyenko, Wendy Flores-Fuentes
Издательство: CRC Press
Год: 2024
Страниц: 466
Язык: английский
Формат: pdf (true)
Размер: 42.3 MB

A comprehensive reference book that addresses the field of machine vision and its significance in cyber-physical systems. It explores the multidisciplinary nature of machine vision, involving electronic and mechatronic devices, Artificial Intelligence algorithms, embedded systems, control systems, robotics, interconnectivity, Data Science, and cloud computing. The book aims to provide advanced students, early career researchers, and established scholars with state-of-the-art knowledge and novel content related to the implementation of machine vision in engineering, scientific knowledge, and technological innovation.

The chapters of the book delve into various topics and applications within the realm of machine vision. They cover areas such as camera and inertial measurement unit calibration, technical vision systems for human detection, design and evaluation of support systems using neural networks, UV sensing in contemporary applications, fiber Bragg grating arrays for medical diagnosis, color model creation for terrain recognition by robots, navigation systems for aircraft, object classification in infrared images, feature selection for vehicle/non-vehicle classification, visualization of sedimentation in extreme conditions, quality estimation of tea using machine vision, image dataset augmentation techniques, machine vision for astronomical images, agricultural automation, occlusion-aware disparity-based visual servoing, Machine Learning approaches for single-photon imaging, and augmented visual inertial wheel odometry.

Measurements are values assigned to refer to a physical quantity or phenomenon; they play an important role in science development. One of the most crucial points in the measurement of machine vision is the proper estimation of information transform quality: such parameters as sensitive part calibration, scene size traceability, accuracy/uncertainty, receiver operating characteristic, repeatability, reproducibility, etc. The use of current technology requires measuring essential attributes from objects, health data, dimensions of a surface and weather, to mention some. These are necessary to do breakthrough innovations in a wide range of fields. The Artificial Intelligence (AI) field is one of them. This field is aimed at the research for imitating human abilities, above all, how they learn. AI can be divided into two main branches such as Machine Learning and Deep Learning. These components are methods and algorithms that are used for prediction and analysis in the case of machine perception. Although Deep Learning is also used for forecasting as well as Machine Learning, the way how the model is created is different. This sub-branch is inspired by how the human brain can learn. In other words, this is based on behavior’s neurons to make connections among others to solve a particular problem. In other contexts, measurements are known as data or instances and depending on the quality of them a problem can be solved. Convolutional neural networks (CNN) have catalyzed several vision recognition systems.

Each chapter is a result of expert research and collaboration, reviewed by peers and consulted by the book's editorial board. The authors provide in-depth reviews of the state of the art and present novel proposals, contributing to the development and futurist trends in the field of machine vision.

"Measurements and Instrumentation for Machine Vision" serves as a valuable resource for researchers, students, and professionals seeking to explore and implement machine vision technologies in various domains, promoting sustainability, human-centered solutions, and global problem-solving.