Classification of Remotely Sensed Data

1. Classification of Remotely Sensed Data General Classification Concepts Unsupervised Classifications

2. Learning Objectives • What is image classification, and what are the three general classification strategies? • What are the main steps required to classify images? • What is unsupervised classification, and what are its advantages and disadvantages?

3. Wyoming land cover modified from USGS Gap Analysis Program data • Based on Landsat TM data • Used a supervised classification technique called CART analysis • Legend is aggregated from much more detailed legend

4. USGS National Land Cover Database (2006) • Based on Landsat TM data • Used CART • Legend is not aggregated – not as detailed as GAP U.S. Landcover - MODIS

5. What is Image Classification? • Process of grouping image pixels into classes that represent self-similar features or themes Analogous to any classification exercise.

6. How? • Three general approaches: • Manual interpretation (e.g., photointerpretation, “heads-up digitizing”) • Unsupervised classification of digital data • Supervised classification of digital data Sequence of methods similar for all three!

7. Group Exercise • Divide yourselves into groups of ~3 • Each group gets a collection of “objects” • Divide the objects up into 4 groups, writing down the questions you ask and the solutions you use to do this

8. A word about manual classification • Manual image interpretation was at the core of remote sensing for much of its history • Still perfectly viable today in some situations • Usually requires people trained to integrate image clues to make decisions about boundary placement • Can use computers for on-screen interpreation

10. Detailed view of Wyoming GAP Land Cover Map

11. General Classification Steps • Field reconnaissance • Development of a classification scheme (legend) • Image enhancement (as needed) • Classification using manual or digital techniques Incorporation of ancillary data (as needed) • Accuracy assessment • Iterative refinement

12. Field Reconnaissance • Critical for understanding the distribution of your theme in the real world • Understanding of ecology, geology, geomorphology, etc. • Helps you choose useful ancillary data • Useful for interpreting the imagery • Nice excuse to get out once in a while

13. What characteristics of this landscape might be important for making a map using satellite data?

14. Developing Classification Schemes (Legends) • How many types do you want to map? • How should you divide up the feature you are interested in? • Can be very controversial!

15. Classification Schemes(List of types to map) • Must be useful (how will map be used?) • Must be detectable using the data you have • Usually hierarchical • Categories must be mutually exclusive • Require explicit definitions of each class

17. Classification Scheme -- Example • Vegetated • Forest • Evergreen • Spruce-fir forest i. Spruce-fir with winterberry understory • Lodgepole pine forest • etc. • Deciduous • Shrubland • Non-Vegetated

18. GroupsGenerate a classification scheme for mapping the main UW campus using Google Earth imagery

19. Digital classification • Creating thematic classes based on groups of similar digital numbers (DNs) • Statistical clustering of the data (lumping spectrally similar pixels into the same class) • Spectral vs. informational classes • Sometimes combine spectral classes together to make informational classes

20. Classification Use many bands at once to create a map of classes

21. Classification Strategy • Unsupervised – computer clusters pixels based only on the similarity of their DNs. • Supervised – computer uses training data—examples of target classes—and assigns pixels to the closest training class using similarity • Others (neural networks, fuzzy logic etc.)

22. Unsupervised Classification 2-bands Max . . . . . . . . Class . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Band Y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0 Band X Max

23. Unsupervised Classification 3-bands 1 pixel 1 Class

24. Unsupervised Classification • Choose bands, indices, enhancements, etc. that highlight differences in your classes • Decide how many groups (classes) you want • Choose a grouping algorithm • Simple clustering, K-means, etc. • Classify the image • Label (and aggregate) classes and evaluate the results

25. Advantages of Unsupervised Classifications • No extensive prior knowledge of map area required (but you have to label the classes!) • Classes are based only on spectral information (spectral classes), so not as subjective as when humans make decisions

26. Disadvantages of Unsupervised Classifications • Spectral classes do not always correspond to informational classes • Spectral properties of informational classes change over time so you can’t always use same class statistics when moving from one image to another

27. Grouping Algorithms • Statistical routines for grouping similar pixels together • Operate in feature space • Differ in how they: • Determine what is similar (distance measures) • Determine the statistical center (centroid) of a class

28. Unsupervised classification of Martian terrain. From Stepinski and Bagaria. 2009. IEEE Geoscience and Remote Sensing Letters.