Madagascar2011 - 08 - OTB segmentation and classification
0 likes1,840 views
The Orfeo Toolbox provides tools for image segmentation, classification, and feature extraction. It includes modules for meanshift segmentation, watershed segmentation, and level set segmentation. It also allows for unsupervised classification using k-means clustering and supervised classification using support vector machines. Example use cases demonstrate segmenting images using mean shift, the otbSegmentationApplication, k-means clustering, and SVM classification. Feature extraction filters can be used to extract metrics like variance, NDVI, and textures to aid in segmentation and classification.
Madagascar2011 - 08 - OTB segmentation and classification
- 1. Orfeo Toolbox Segmentation, classification Stéphane MAY stephane.may@cnes.fr orfeo-toolbox.org 1
- 2. Segmentation Definition of segmentation Extract the outlines of different regions in the image Divide the image the image into regions with pixels which have something in common OTB – Monteverdi Meanshift segmentation module OTB (integration of ITK library) Watershed segmentation Region growing segmentation Level set segmentation Hybrid segmentation, etc... orfeo-toolbox.org 2
- 3. Monteverdi – Mean-shift Filtering > Meanshift clustering orfeo-toolbox.org 3
- 4. Use case 1 : segmentation with mean-shift Menu File > Open ./theme2/extraitIm2_C/Im2_c_extrait.tif Menu Filtering > Mean-shift clustering Change radius : 5 Spectral radius : 15 Min region size : 15 Clusters : ON Change values and Click on Run button Click on Close button after selecting right set of parameters See also : ➢ Image filtered / Image clustered ➢ See OTB-Software-Guide.pdf for details orfeo-toolbox.org 4
- 5. Orfeo toolbox - otbSegmentationApplication otbSegmentationApplication orfeo-toolbox.org 5
- 6. otbSegmentationApplication (1/8) orfeo-toolbox.org 6
- 7. otbSegmentationApplication (2/8) orfeo-toolbox.org 7
- 8. otbSegmentationApplication (3/8) orfeo-toolbox.org 8
- 9. otbSegmentationApplication (4/8) orfeo-toolbox.org 9
- 10. otbSegmentationApplication (5/8) orfeo-toolbox.org 10
- 11. otbSegmentationApplication (6/8) orfeo-toolbox.org 11
- 12. otbSegmentationApplication (7/8) orfeo-toolbox.org 12
- 13. otbSegmentationApplication (8/8) orfeo-toolbox.org 13
- 14. Use case 2 : otbSegmentationApplication Command line application : otbSegmentationApplication Open ./theme2/extraitIm2_C/Im2_c_extrait.tif Segment homogeneous areas Save your results orfeo-toolbox.org 14
- 15. Monteverdi – Classification modules Menu Learning SVM classification K-Means clustering orfeo-toolbox.org 15
- 16. Monteverdi Menu Learning > K-means orfeo-toolbox.org 16
- 17. Use case 3 : unsupervised clustering with k-means Menu File > Open ./theme2/extraitIm2_C/Im2_c_extrait.tif Menu Learning > k-means clustering (doc OTBSoftwareGuide.pdf) Training 15% Number of classes : 5 Iteration number : 100 Convergence : 0.0001 Save your results Try with several parameters set Visualization > Viewer > Compare results orfeo-toolbox.org 17
- 18. Monteverdi Menu Learning > SVM classification (1/3) orfeo-toolbox.org 18
- 19. Monteverdi Menu Learning > SVM classification (2/3) orfeo-toolbox.org 19
- 20. Monteverdi Menu Learning > SVM classification (3/3) orfeo-toolbox.org 20
- 21. Use case 4 : supervised classification with SVM (1/2) Menu File > Open ./theme2/IM2/extraitIm2_C/Im2_c_extrait.tif Menu Learning > SVM Classification Create several classes (4-5) ➢ Add ➢ Select polygons (right click to end a polygon) ➢ Edit names ➢ Change colors Learn Display orfeo-toolbox.org 21
- 22. Use case 4 : supervised classification with SVM (2/2) Menu Learning > SVM Classification Deselect random validation set Select Display validation Select your classes 1 by 1 ➢ Select polygons (right click to end a polygon) Display Validate File > Export selected polygons orfeo-toolbox.org 22
- 23. Monteverdi – Feature extraction Filtering Feature Extraction (1/4) orfeo-toolbox.org 23
- 24. Monteverdi – Feature extraction Filtering Feature Extraction (2/4) orfeo-toolbox.org 24
- 25. Monteverdi – Feature extraction Menu Filtering > Feature Extraction (3/4) Mean, variance, Gradient, spectral angle Original data (=> no need to concatenate channels after filtering) Textures (energy, entropy, contrast, etc) Morphological filters Radiometric indexes Vegetation (NDVI, ARVI, etc), Soil, Built up, Water Edge density Mean shift orfeo-toolbox.org 25
- 26. Monteverdi – Feature extraction Menu Filtering > Feature extraction (4/4) Radiometric indexes ➢ Vegetation NDVI, RVI, PVI, etc ➢ Soil BI2 ➢ Built up ISU orfeo-toolbox.org 26
- 27. Use case 5 : Segment with Feature extraction Menu File > Open ./theme2/IM2/extraitIm2_C/Im2_c_extrait.tif Menu Filtering > Feature extraction Test the following features (See OTB-Software-Guide.pdf for technical details on algorithms) ➢ Original data (=> no need to concatenate channels after filtering) ➢ Spectral angle : choose one vegetation pixel ➢ Variance, mean ➢ NDVI ➢ Meanshift filtering, etc. Menu Learning > K-Means Menu Learning > SVM (import polygons) Compare your results orfeo-toolbox.org 27
- 28. Monteverdi Thank you for your attention ! orfeo-toolbox.org 28