Fuzzy Machine Learning Algorithms for Remote Sensing Image Classification  book cover
1st Edition

Fuzzy Machine Learning Algorithms for Remote Sensing Image Classification

  • Available for pre-order. Item will ship after July 20, 2020
ISBN 9780367355715
July 20, 2020 Forthcoming by CRC Press
208 Pages - 64 B/W Illustrations

SAVE ~ $26.00
was $130.00
USD $104.00

Prices & shipping based on shipping country

Book Description

This book covers the state of art image classification methods for discrimination of earth objects from remote sensing satellite data with an emphasis on fuzzy based learning methods including applications for preparing land cover classification outputs from actual satellite data. It provides details about the temporal indices database using proposed class-based sensor independent approach supported by practical examples. Fuzzy based algorithms with machine learning algorithms to prepare land cover maps is discussed. Accuracy assessment for soft classification outputs are included and all algorithms are supported by in-house developed tool as Sub-pixel Multi-spectral Image Classifier. Aimed at researchers, graduate students, professionals in earth remote sensing, remote sensing image and data processing, geography and geoinformation science, image classification, this book:

Exclusively focuses on using fuzzy classification to remote sensing images.

Covers Sub Pixel Multi-Spectral Image Classifier Tool (SMIC) to support discussed algorithms.

Explains fuzzy and learning based classifiers with in-house developed SMIC tool.

Discusses how ANN, CNN, RNN and hybrid learning classifiers application on remote sensing images.

Combines explanation of the algorithms with examples, graph, and charts.

Table of Contents

I.  Machine Learning
1. Introduction
2. Pattern Recognition
3. Machine Learning Algorithms for Pattern Recognition

II. Ground Truth Data for Remote Sensing Image Classification
1. Criteria for Ground Truth Data
2. Training Data
3. Testing Data

III. Fuzzy Classifiers
1. Soft Classifiers
2. Traditional Classifiers vs Soft Classifiers
3. Linear and non-linear classifiers
4. Fuzzy c-Means (FCM) Classifier
5. Possibilistic c-Means (PCM) Classifier
6. Noise Clustering (NC) Classifier
7. Why Noise Clustering?
8. Limitations of Possibilistic c-Means (PCM)
9. Improved Possibilistic c-Means (IPCM)
10. Advantages of IPCM over PCM
11. Modified Possibilistic c-Means (MPCM)

IV. Learning Based Classifiers
1. Artificial Neural Network (ANN)
2. Convolutional Neural Network (CNN)
3. Recurrent Neural Network (RNN)
4. Hybrid Learning Network (HLN)
5. Deep Learning Concepts
6. In-house Tool for Study of Learning Algorithms

V. Hybrid Fuzzy Classifiers
1. Entropy Based Hybrid Soft Classifiers
2. Fuzzy c-Means with Entropy (FCME)
3. Noise Clustering with Entropy (NCWE) Classifier
4. Similarity/Dissimilarity Measures in Fuzzy Classifiers
5. Kernels Concept in Fuzzy Classifiers
6. Theory behind Markov Random Field (MRF)
7. Types of MRF methods
8. Contextual Information using MRF
9. Convolution based Local Information in Fuzzy Classifiers

VI. Fuzzy Classifiers for Temporal Data Processing
1. Introduction
2. Indices Approaches
3. Fuzzy Based Algorithms for Single Class Extraction
4. Concept for Mono/Bi-sensor Remote Sensing Data Processing

VII. Assessment of Accuracy for Soft Classification
1. Generation of Testing Data
2. Methods for Assessment of Accuracy
3. Fuzzy Error Matrix  (FERM) and Other Operators
4. Entropy Method
5. Mean and Variance Method for Edge Preservation
6. Correlation Coefficient
7. Root Mean Square Error
8. Receiver Operating Characteristics (ROC)

Appendix A1
SMIC: Sub_Pixel Multi-spectral Image Classifier Tool

Appendix A2
Case Study 1    : Study of similarity and dissimilarity measures with IPCM and MPCM classifiers
Case Study  2    : Bi-sensor temporal data for paddy crop mapping
Case Study  3    : Handling non-linearity between classes using kernels in fuzzy classifiers
Case Study 4   : Handling noise through MRF based noise clustering classifier
Case Study 5 :Local convolution based contextual information in Possibilistic c-Means Classification
Case Study  6   : Optimization of local convolution based MPCM classifier and identification of paddy and burnt paddy fields
Case Study 7       : Semi-supervised training approach for PCM classifier
Case Study 8 : Study of hybridizing stochastic and deterministic measures with fuzzy based classifier
Case Study 9      : Kernal based PCM Classification approach
Case Study 10   : Effect of red edge bands in fuzzy classification: a case  study of sunflower crop
Case Study 11 : Discriminating sugar ratoon / plant crop using multi- sensor temporal data

View More



Anil Kumar is working as Scientist/Engineer-'SG’ & Head Photogrammetry and Remote Sensing Department at Indian Institute of Remote Sensing (IIRS), Indian Space Research organisation (ISRO), Dehradun, India. He received his B.Tech degree in Civil Engineering from University of Lucknow, India and M.E. degre as well as inservise Ph.D in soft computing from Indian Institute of Technology, Roorkee, India. He has published 46 papers in journals. Guided 36 masters and 5 Ph.D thesis. He has been recipient of the prestigious P. R. Pisharoty Memorial Award conferred by the Indian Society of Remote Sensing. He is a life member of the Indian Society of Remote Sensing. His current research interests are in Soft computing, Deep Learning, Multi-sensor temporal data processing, Digital Photogrammetry, GPS and LiDAR. Senthil Kumar is the Director of UN-affliated Centre for Space Science and Technology Education in Asia and the Pacific in Dehradun, India. He received M.Sc. (Engg.) and Ph.D. from the Indian Institute of Science, Bangalore in the field of image processing in 1985 and 1990 respectively. He joined ISRO in 1991. Since then he has served in Indian Remote Sensing programs in various capacities. He has published more than 120 technical papers in international journals and conferences and co-edited a book on Remote Sensing of Northwest Himalayan Ecosystems. He has received ISRO Team awards for his contributions to Chandrayaan-1 and Cartosat-1 missions. His research areas include remote sensing sensor characterization, radiometric data processing, image restoration, data fusion techniques and in soft computing techniques. He has also been a recipient of the prestigious Prof. Satish Dhawan Award conferred by the Indian Society of Remote Sensing. He is a life member of the Indian Society of Remote Sensing and the Indian Society of Geomatics. Priyadarshi Upadhyay is working as a Scientist/Engineer-SD in Uttarakhand Space Application Centre (USAC), Dehradun, India. He received his M.Sc. degree in Physics from Kumaun University Nainital, India and M.Tech. degree in Remote Sensing from Birla Institute of Technology, Mesra Ranchi, India. He has received his Ph.D. degree from Indian Institute of Technology Roorkee, India in the area of time series remote sensing for single crop identification. He has published 15 research papers in various International Journals, Internation and National Conferences. He has been awarded by presitigious CSIR-NET, GATE and MHRD Travel Grant Fellowships. He is a life member of Indian Society of Remote Sensing and The Institute of Engineers (India). His current research interest are Microwave Remote Sensing for Soil Moisture and Crop Mapping, Polarimatric and Inerferrometric SAR, Hyperspectral and Optical Remote Sensing, Climate Change, Ecological Studies in Himalayan Region for Economically Important Crops and Plants.