Remote Sensing Digital Image Analysis : An Introduction

Remote Sensing Digital Image Analysis provides the non-specialist with an introduction to quantitative evaluation of satellite and aircraft derived remotely retrieved data. Since the first edition of the book there have been significant developments in the algorithms used for the processing and analysis of remote sensing imagery; nevertheless many of the fundamentals have substantially remained the same. This new edition presents material that has retained value since those early days, along with new techniques that can be incorporated into an operational framework for the analysis of remote sensing data. The book is designed as a teaching text for the senior undergraduate and postgraduate student, and as a fundamental treatment for those engaged in research using digital image processing in remote sensing. The presentation level is for the mathematical non-specialist. Since the very great number of operational users of remote sensing come from the earth sciences communities, the text is pitched at a level commensurate with their background. Each chapter covers the pros and cons of digital remotely sensed data, without detailed mathematical treatment of computer based algorithms, but in a manner conductive to an understanding of their capabilities and limitations. Problems conclude each chapter. Z......Page 0 Preface 5......Page 5 Contents ......Page 8 Atmospheric windows, 1......Page 19 Radiometric resolution, 4......Page 22 push broom, 6......Page 24 Polarisation, 10......Page 28 Spatial data sources, 18......Page 36 Image scale, 19......Page 37 Neogeography, 20......Page 38 1.8 How This Book is Arranged 21......Page 39 1.9 Bibliography on Sources and Characteristics of Remote Sensing Image Data 23......Page 41 1.10 Problems 25......Page 43 radiometric, 27......Page 45 Transfer characteristic of a detector, 28......Page 46 Solar radiation curve, 31......Page 49 compensation, 32......Page 50 measured, 33......Page 51 effect on imagery, 37......Page 55 broad waveband systems, 38......Page 56 scaled, 40......Page 58 haze removal, 44......Page 62 sources, 47......Page 65 earth rotation effect, 48......Page 66 platform velocity effect, 50......Page 68 panoramic, 51......Page 69 S-bend, 53......Page 71 2.16 Geometric Distortion Caused by Instrumentation Characteristics 54......Page 72 correction, 56......Page 74 mathematical modelling, 64......Page 82 Registration image to image, 67......Page 85 71......Page 89 Zooming, 72......Page 90 2.23 Problems 73......Page 91 Photointerpretation, 79......Page 96 Quantitative analysis, 83......Page 100 Thematic mapping, 84......Page 101 spectral, 86......Page 103 parametric, 92......Page 109 unsupervised, 93......Page 110 3.8 Bibliography on Interpreting Images 94......Page 111 3.9 Problems 95......Page 112 Look up table (LUT), 99......Page 115 modification, 100......Page 116 uniform, 106......Page 122 Mosaicing, 114......Page 130 4.6 Density Slicing 118......Page 134 4.7 Bibliography on Radiometric Enhancement of Images 120......Page 136 multicycle, 122......Page 138 template, 127......Page 142 mean value, 130......Page 145 modal, 133......Page 148 the Roberts operator, 135......Page 150 the Sobel operator, 136......Page 151 Spatial derivative, 137......Page 152 5.5 Edge Detection 139......Page 154 Spot detection, 141......Page 156 Transfer function, 142......Page 157 Fourier transform, 145......Page 160 5.10 Geometric Properties of Images 146......Page 161 Minimum distance classifier. See Classifier Morphology, 150......Page 165 Shape recognition, 157......Page 172 5.14 Problems 158......Page 173 Vegetation index, 161......Page 175 162......Page 176 Principal components transformation, 163......Page 177 noise adjusted, 186......Page 200 Tasseled cap transform, 189......Page 203 kernel, 192......Page 206 195......Page 209 Pan sharpening, 197......Page 211 6.9 Bibliography on Spectral Domain Image Transforms 198......Page 212 6.10 Problems 199......Page 213 7.1 Introduction 203......Page 216 Imaginary numbers, 204......Page 217 Unit step function, 207......Page 220 210......Page 223 Discrete Fourier transform, 212......Page 225 Fast Fourier transform, 215......Page 228 Sampling theory, 218......Page 231 of an image, 221......Page 234 image processing, 224......Page 237 two dimensional, 226......Page 239 Window functions, 227......Page 240 Wavelet transform, 229......Page 242 233......Page 246 upsampling, 237......Page 250 7.13.6.3 Relationship Between the Low and High Pass Filters 238......Page 251 of an image, 241......Page 254 7.16 Bibliography on Spatial Domain Image Transforms 243......Page 256 7.17 Problems 244......Page 257 soft, 247......Page 260 248......Page 261 MAP, 250......Page 263 Gaussian mixture models, 260......Page 273 Expectation-maximisation, 265......Page 278 parallelepiped, 269......Page 282 Mahalanobis, 271......Page 284 Table look up classifier, 272......Page 285 Euclidean, 273......Page 286 spectral angle mapper, 274......Page 287 276......Page 289 Slack variables, 281......Page 294 283......Page 296 sigmoidal, 286......Page 299 committee, 288......Page 301 Threshold logic unit (TLU), 290......Page 303 Spatial context, 299......Page 312 neighbourhood function, 303......Page 316 compatibility coefficient, 304......Page 317 8.20.4.3 Determining the Compatibility Coefficients 305......Page 318 stopping rule, 306......Page 319 8.20.4.5 Examples 307......Page 320 iterated conditional modes, 312......Page 325 8.22 Problems 315......Page 328 Clustering, 319......Page 332 Minkowski, 320......Page 333 migrating means, 322......Page 335 merging and deleting, 323......Page 336 splitting clusters, 325......Page 338 cost, 326......Page 339 single pass algorithm, 327......Page 340 331......Page 344 within cluster scatter, 333......Page 346 mountain, 335......Page 348 9.14 Bibliography on Clustering and Unsupervised Classification 339......Page 352 within class, 340......Page 353 Feature reduction, 343......Page 355 Separability, 344......Page 356 divergence, 345......Page 357 Jeffries-Matusita distance, 350......Page 362 Transformed divergence, 351......Page 363 in clustering, 353......Page 365 by spectral transformation, 354......Page 366 358......Page 370 10.8.2.3 The Generalised Eigenvalue Equation 359......Page 371 Eigenvalue equation generalised, 360......Page 372 block diagonal, 370......Page 382 regularisation, 375......Page 387 10.11 Bibliography on Feature Reduction 377......Page 389 10.12 Problems 378......Page 390 11.1 Introduction 381......Page 393 382......Page 394 384......Page 396 hybrid, 388......Page 400 393......Page 405 394......Page 406 396......Page 408 413......Page 425 Spectroscopic interpretation, 422......Page 434 Unmixing, 424......Page 436 Space measurement, 426......Page 438 11.12 Bibliography on Image Classification in Practice 431......Page 443 433......Page 445 multisource, 437......Page 448 Stacked vector, 438......Page 449 statistical multisource, 439......Page 450 Theory of Evidence, 444......Page 455 Knowledge-based image analysis, 448......Page 459 458......Page 469 12.7 Bibliography on Multisource Image Analysis 461......Page 472 12.8 Problems 463......Page 474 Satellite orbital velocity, 465......Page 476 Bit, 467......Page 478 Vector column, 469......Page 480 conditional, 479......Page 489 Loss function, 483......Page 493 Index 487......Page 496 Accuracy confidence limits, 412......Page 424 Kappa coefficient, 401......Page 413 omission, 397......Page 409 User’s accuracy, 398......Page 410 400......Page 412 Processing element (PE), 291......Page 304 Planck’s law, 3......Page 21 Boosting, 289......Page 302 Aliasing, 219......Page 232 Quantity disagreement, 404......Page 416 466......Page 477 ATREM, 42......Page 60 MODTRAN4, 43......Page 61 log residuals, 46......Page 64 the flat field method, 45......Page 63 water vapour, 41......Page 59 35......Page 53 2......Page 20 Orthogonal functions, 230......Page 243 Pseudocolouring, 119......Page 135 Binomial distribution, 407......Page 419 Bispectral plot, 390......Page 402 392......Page 404 Byte, 468......Page 479 by canonical analysis, 356......Page 368 Maximum likelihood classifier. See Classifier Mean vector, 164......Page 178 Grid searching, 395......Page 407 Water absorption bands, 12......Page 30 source-specific data, 440......Page 451 Gaussian maximum likelihood, 252......Page 265 Training data, 88......Page 105 Training field, 249......Page 262 484......Page 494 logarithmic, 442......Page 453 random forests, 420......Page 432 383......Page 395 CART, 415......Page 427 441......Page 452 ECHO, 302......Page 315 256......Page 269 Normal probability distribution, 90......Page 107 minimum distance, 89......Page 106 one-against-one, 287......Page 300 progressive two-class, 421......Page 433 dendrogram, 332......Page 345 critical distance, 328......Page 341 strip generation parameter, 329......Page 342 Clustering criteria SSE, 321......Page 334 Volume scattering, 16......Page 34 Compression, 24......Page 42 Gaussian stretch, 115......Page 131 automatic, 103......Page 119 piecewise linear, 105......Page 121 saturating linear, 102......Page 118 the Taylor method, 178......Page 192 Control point. See Ground control point Convolution, 129......Page 144 discrete, 217......Page 230 theorem, 216......Page 229 Correlation, 200......Page 214 Covariance matrix, 165......Page 179 Variance, 481......Page 491 Crabbing, 75......Page 93 Curse of dimensionality, 258......Page 271 Decision boundary feature extraction (DBFE), 368......Page 380 prior, 251......Page 264 Threshold maximum likelihood classifier, 253......Page 266 minimum distance classifier, 268......Page 281 unconditional maximum likelihood, 485......Page 495 minimum distance classifier, 267......Page 280 Decision tree. See Classifier binary, 414......Page 426 Orthogonal sum, 446......Page 457 dot product, 473......Page 484 Impulse function, 206......Page 219 Discrete inverse Fourier transform, 213......Page 226 properties, 214......Page 227 Fisher criterion, 362......Page 374 341......Page 354 347......Page 359 use in feature selection, 348......Page 360 Support vector machine. See Classifier Support vectors, 279......Page 292 Eigenvector, 168......Page 182 Emissivity spectra, 14......Page 32 Endorsement of a proposition, 455......Page 466 Evidential support, 445......Page 456 386......Page 398 Instantaneous field of view (IFOV), 5......Page 23 dyadic, 232......Page 245 downsampling, 235......Page 248 Filtering edge detection, 140......Page 155 kernel, 128......Page 143 unsharp masking, 138......Page 153 Fourier series, 208......Page 221 Generalisation, 427......Page 439 cubic convolution, 60......Page 78 choice, 61......Page 79 Ground control points, 57......Page 75 earth rotation, 65......Page 83 panoramic effects, 66......Page 84 nearest neighbour, 59......Page 77 scan nonlinearities, 55......Page 73 430......Page 442 Histogram anomalous equalisation, 112......Page 128 cumulative, 108......Page 124 169......Page 183 429......Page 441 8......Page 26 Image display colour infrared, 81......Page 98 System function, 143......Page 158 Gini, 416......Page 428 Inference engine, 451......Page 462 Scattering diffuse, 34......Page 52 Qualitative reasoning, 454......Page 465 radial basis function, 285......Page 298 Mercer condition, 284......Page 297 460......Page 471 Lagrange multipliers, 264......Page 277 278......Page 291 Regularisation parameter, 282......Page 295 Margin, 277......Page 290 Markov Random Field, 308......Page 321 partition function, 310......Page 323 Ising model, 311......Page 324 row, 470......Page 481 transpose, 472......Page 483 adjugate, 475......Page 486 block diagonal, 371......Page 383 373......Page 385 eigenvectors, 476......Page 487 singular, 474......Page 485 square, 471......Page 482 raised to a power, 477......Page 488 pseudo inverse, 425......Page 437 opening, 155......Page 170 dilation, 153......Page 168 erosion, 152......Page 167 structuring element, 151......Page 166 Multinomial distribution, 411......Page 423 Neural network. See Classifier backpropagation, 292......Page 305 296......Page 309 No free lunch theorem, 428......Page 440 Noise fraction, 187......Page 201 Non-parametric discriminant analysis (NDA), 364......Page 376 Non-parametric weighted feature extraction (NWFE), 369......Page 381 Sampling theorem, 220......Page 233 near-polar, 26......Page 44 Pattern vector, 87......Page 104 Speckle, 17......Page 35 redundancy reduction, 182......Page 196 image compression, 181......Page 195 image enhancement, 176......Page 190 origin shift, 173......Page 187 segmented, 374......Page 386 480......Page 490 337......Page 350 supervised, 443......Page 454 Production rules, 452......Page 463 36......Page 54 instrumentation errors, 30......Page 48 specular, 15......Page 33 sequential similarity detection algorithms, 68......Page 86 69......Page 87 to a map grid, 62......Page 80 fixed point, 318......Page 331 Scatterplot, 387......Page 399 434......Page 446 unit, 171......Page 185 translation, 231......Page 244 mechanical line, 7......Page 25 Scatter matrix among class, 366......Page 378 between class, 365......Page 377 within class, 367......Page 379 Variogram, 147......Page 162 smoothing, 131......Page 146 median, 132......Page 147 223......Page 236 Spatial gradient, 134......Page 149 Spectral reflectance characteristics, 11......Page 29 phase, 209......Page 222 Synthetic aperture radar, 9......Page 27 number required, 255......Page 268 405......Page 417 entropy, 149......Page 164 grey level co-occurrence matrix (GLCM), 148......Page 163 Thermal infrared, 13......Page 31 385......Page 397 Training pixel, 293......Page 306 234......Page 247 Haar, 240......Page 253 239......Page 252 scaling vector, 236......Page 249 Weight vector, 275......Page 288 Remote Sensing Digital Image Analysis provides the non-specialist with a treatment of the quantitative analysis of satellite and aircraft derived remotely sensed data. Since the first edition of the book there have been significant developments in the algorithms used for the processing and analysis of remote sensing imagery; nevertheless many of the fundamentals have substantially remained the same. This new edition presents material that has retained value since those early days, along with new techniques that can be incorporated into an operational framework for the analysis of remote sensing data. The book is designed as a teaching text for the senior undergraduate and postgraduate student, and as a fundamental treatment for those engaged in research using digital image processing in remote sensing. The presentation level is for the mathematical non-specialist. Since the very great number of operational users of remote sensing come from the earth sciences communities, the text is pitched at a level commensurate with their background. Each chapter covers a different aspect of the analysis of digital remotely sensed data, without an excessively detailed mathematical treatment of computer based algorithms, but in a manner conductive to an understanding of their capabilities and limitations. Problems conclude each chapter.