Recipes

This chapter presents guidelines to perform various remote sensing and/or image processing tasks. Its goal is not to be exhaustive, but rather to familiarise users with the OTB package functionality and demonstrate how they can be applied.

• From raw image to calibrated product
  • Optical radiometric calibration
  • Pan-sharpening
  • Digital Elevation Model management
  • Ortho-rectification and map projections
    • Beware of “ortho-ready” products
    • Ortho-rectification with OTB Applications
• SAR processing
  • Calibration
  • Despeckle
  • Polarimetry
    • Matrix conversions
    • Polarimetric decompositions
    • Polarimetric synthetis
    • Polarimetric data visualization
• Residual registration
  • Extract metadata from the image reference
  • Extract homologous points from images
  • Geometry refinement using homologous points
  • Orthorectify image using the affine geometry
• Image processing and information extraction
  • Simple calculus with channels
  • Images with no-data values
  • Segmentation
  • Large-Scale Mean-Shift (LSMS) segmentation
    • Step 1: Mean-Shift Smoothing
    • Step 2: Segmentation
    • Step 3 (optional): Merging small regions
    • Step 4: Vectorization
    • All-in-one
  • Dempster Shafer based Classifier Fusion
    • Fuzzy Model (requisite)
    • First Step: Compute Descriptors
    • Second Step: Feature Validation
• BandMathImageFilterX (based on muParserX)
  • Fundamentals: headers, declaration and instantiation
  • Syntax: first elements
  • New operators and functions
  • Application Programming Interface (API)
• Enhance local contrast
  • Principles
  • Advanced parameters
• Classification
  • Feature classification and training
    • Train a classifier with features
    • Feature classification
    • Validating classification
  • Pixel based classification
    • Samples statistics estimation
    • Sample selection
    • Samples extraction
    • Working with several images
    • Images statistics estimation
    • Training the model
    • Using the classification model
    • Validating the classification model
    • Fancy classification results
    • Example
  • Unsupervised learning
  • Fusion of classification maps
    • Majority voting for the fusion of classifications
    • Dempster Shafer framework for the fusion of classifications
    • Recommendations to properly use the fusion of classification maps
  • Majority voting based classification map regularization
    • Majority voting for the classification map regularization
    • Handling ambiguity and not classified pixels in the majority voting based regularization
    • Recommendations to properly use the majority voting based regularization
    • Example
  • Regression
    • Statistics estimation
    • Training
    • Prediction
• Feature extraction
  • Local statistics extraction
  • Edge extraction
  • Radiometric indices extraction
  • Morphological features extraction
    • Binary morphological operations
    • Gray scale morphological operations
  • Textural features extraction
    • Haralick texture features
    • SFS texture extraction
• Stereoscopic reconstruction from VHR optical images pair
  • Estimate epipolar geometry transformation
  • Resample images in epipolar geometry
  • Disparity estimation: Block matching along epipolar lines
  • From disparity to Digital Surface Model
  • One application to rule them all in multi stereo framework scheme
  • Stereo reconstruction good practices
  • Algorithm outline
• OTB processing in Python
  • Basics
  • Numpy array processing
  • In-memory connection
  • Interactions with OTB pipeline
  • Corner cases
    • Calling UpdateParameters()
    • No metadata in NumPy arrays
    • Setting of EmptyParameter