Image Classification II (Ch11: 348-360 & Field Guide: 237-253)

1. Define Textural Classifier:
2. What are the primary requirements for using ancillary data?
3. Explain fuzzy clustering:
4. What is contextual classification and how does it differ from textural?
5. What are some of the limitations that affect choices made in image classification?

1. What does textural classifiers measure?
2. When do textural measures work best?
3. What is ancillary data?
4. How is Layered classification useful?
5. Define contextual classification in 4 ways?

1. What disadvantage is there to choosing the parallelepiped decision rule and how does this differ from the signature ellipse method?
2. What four functions can be applied to signatures in ERDAS Imagine after they are created?
3. What does the user need to do to classify a pixel when there is signature overlap?
4. What is the major difference in using the Mahalanobis distance classifier as opposed to the minimum distance classifier?
5. How can limits for a parallelepiped decision rule be established?

Keane, R. E., et al. 2001. Mapping wildland fuels for fire management across multiple scales: Integrating remote sensing, GIS, and biophysical modeling. International Journal of Wildland Fire, 10: 301–319.

1. What are fuels defined as?
2. What is the difference between surface and crown fuels?
3. What is a fuel model?
4. Why are fuel maps essential at many spatial and temporal scales?
5. What are the strengths and weakness of the four strategies used to map fuels presented in this paper?
6. What strategy does the article propose for using current technology for fuel mapping?