Fuzzy Machine Learning Algorithms for Remote Sensing Image Classification

This book provides detailed descriptions of state-of-the-art image classification methods for discriminating earth objects from remote sensing satellite data, with an emphasis on fuzzy machine learning and deep learning algorithms. It covers thematic mapping of multiple-class or specific-class landcover from multispectral optical remote sensing data and monitoring specific stages of a particular class. It also discusses other related algorithms based on distance, kernel-based, and spatial information. The book is aimed at graduate, postgraduate, research scholars, and working professionals in earth observation and satellite image processing.

Format: Paperback / softback
Length: 194 pages
Publication date: 25 September 2023
Publisher: Taylor & Francis Ltd

This comprehensive book delves into the latest advancements in image classification methods for discriminating Earth objects from remote sensing satellite data, with a particular focus on fuzzy machine learning and deep learning algorithms. Providing detailed explanations, these algorithms are designed to be implemented directly for thematic mapping of multiple-class or specific-class landcover from multispectral optical remote sensing data. Moreover, these algorithms, when combined with multi-date, multi-sensor remote sensing, enable the monitoring of specific stages, such as the phenology of growing crops, for a particular class.

Fuzzy machine learning algorithms find significant applications in diverse fields such as crop insurance, forest fire mapping, stubble burning, post-disaster damage mapping, and more. Additionally, this book offers insights into the temporal indices database using the proposed Class Based Sensor Independent (CBSI) approach, supported by practical examples. Furthermore, it addresses other related algorithms based on distance, kernel-based, and spatial information through Markov Random Field (MRF)/Local convolution methods to handle mixed pixels, non-linearity, and noisy pixels.

Furthermore, this book explores techniques for quantitatively assessing soft classified fraction outputs from soft classification, supported by the in-house developed tool called the sub-pixel multi-spectral image classifier (SMIC). This resource is aimed at graduate, postgraduate, research scholars, and working professionals from various branches, including Geoinformation sciences, Geography, Electrical, Electronics, and Computer Sciences, who are engaged in Earth observation and satellite image processing. The learning algorithms discussed in this book have potential applications in other related fields, such as medical imaging.

In summary, this book serves as a valuable resource for advancing our understanding and utilization of image classification methods in Earth observation and satellite image processing. It provides detailed insights into cutting-edge algorithms, practical applications, and techniques, making it essential for researchers, professionals, and students in this field.

Weight: 408g
Dimension: 234 x 156 (mm)
ISBN-13: 9780367355746