Название: Integrating Metaheuristics in Computer Vision for Real-World Optimization Problems
Автор: Shubham Mahajan, Kapil Joshi, Amit Kant Pandit
Издательство: Wiley-Scrivener
Год: 2024
Страниц: 353
Язык: английский
Формат: pdf (true)
Размер: 16.2 MB

This book recognizes digital images and video use in fields of surveillance, manufacturing, and agriculture. Organized in two parts, a variety of readers learn about communication, automation, and beginning a career in research and innovation. The goal of the book is to provide research that addresses broad challenges in both theoretical and application aspects of soft computing and Machine Learning in image processing and computer vision.

Using cameras, data, and algorithms rather than the retina, optic nerve, and visual cortex, computer vision teaches a computer for carrying out these tasks. However, it must do so in a much shorter amount of time. Systems that have been taught to check items or keep an eye on production machinery may inspect hundreds of items or tasks per minute and find errors or issues that cannot be seen. The use of computer vision grows quickly across a range of sectors, from manufacturing and cars to energy and utilities.

A computer’s vision to comprehend and analyze visual input is achieved via the application of algorithms, methods, and models. Image capture, preprocessing, feature extraction, and classification are some of the procedures that are taken to accomplish this. Using cameras, sensors, or other imaging equipment, image acquisition includes acquiring visual data. The captured pictures are subsequently cleaned up and improved using preprocessing techniques, such as reducing noise, fixing distortions, and altering the color balance. Feature extraction includes detecting essential properties or aspects of the photos, such as edges, corners, textures, or form. CNNs, which are intended to learn from and extract characteristics from enormous volumes of visual input, are frequently used for this purpose. By classifying photos based on their attributes, machine learning techniques are used. Support vector machines (SVMs), decision trees, or deep learning algorithms like CNNs can all be used for this. In general, the topic of computer vision is complicated, involving a variety of methods and models, and it is ever-evolving as new studies are undertaken and new applications are created.

Included in This Book:

- Discussions on advancements in diagnostics and therapeutic techniques for ischemic stroke, object detection, and tracking face detection

- Talks about the comparative evaluation of machine learning algorithms for bank fraud detection and improving performance with feature selection, extraction, and learning

- Details the concept of trading cryptocurrency market-based strategies with comparative evaluation and prediction of exoplanets by using machine learning methods

- Explores the advancements of machine learning and highlights the strengths and limitations of swarm intelligence and computation