x
Contents
9 Hyperspectral Classification Using ICA Based Mixture Model
217
9.1 Introduction ....................................................................... 217
9.2 Independent Component Analysis Mixture Model (ICAMM) -
Theory .............................................................................. 219
9.2.1 ICAMM Classification Algorithm ................................. 220
9.3 Experimental Methodology .................................................. 222
9.3.1 Feature Extraction Techniques ..................................... 223
9.3.2 Feature Ranking ........................................................ 224
9.3.3 Feature Selection ....................................................... 224
9.3.4 Unsupervised Classification ........................................ 225
9.4 Experimental Results and Analysis ........................................ 225
9.5 Summary ........................................................................... 233
10 Support Vector Machines for Classification
of Multi- and Hyperspectral Data
237
10.1 Introduction ....................................................................... 237
10.2 Parameters Affecting SVM Based Classification ....................... 239
10.3 Remote Sensing Images ........................................................ 241
10.3.1 Multispectral Image ................................................... 241
10.3.2 Hyperspectral Image .................................................. 242
10.4 SVM Based Classification Experiments ................................... 243
10.4.1 Multic1ass Classification ............................................. 243
10.4.2 Choice of Optimizer ................................................... 245
10.4.3 Effect of Kernel Functions ........................................... 248
10.5 Summary ........................................................................... 254
11 An MRF Model Based Approach
for Sub-pixel Mapping from Hyperspectral Data
257
11.1 Introduction ....................................................................... 257
11.2 MRF Model for Sub-pixel Mapping ........................................ 259
11.3 Optimum Sub-pixel Mapping Classifier .................................. 261
11.4 Experimental Results ........................................................... 265
11.4.1 Experiment 1:
Sub-pixel Mapping from Multispectral Data .................. 266
11.4.2 Experiment 2:
Sub-pixel Mapping from Hyperspectral Data ................. 271
11.5 Summary ........................................................................... 276
12 Image Change Detection and Fusion Using MRF Models
279
12.1 Introduction ....................................................................... 279
12.2 Image Change Detection using an MRF model ........................ 279
12.2.1 Image Change Detection (lCD) Algorithm .................... 281
12.2.2 Optimum Detector .................................................... 284
12.3 Illustrative Examples ofImage Change Detection .................... 285
12.3.1 Example 1: Synthetic Data ........................................... 287
12.3.2 Example 2: Multispectral Remote Sensing Data .............. 290