DC-18 - Algorithms and processing

You are currently viewing an archived version of Topic . If updates or revisions have been published you can find them at .

Learning Objectives: 
  • Differentiate supervised classification from unsupervised classification
  • Describe the sequence of tasks involved in the geometric correction of the Advanced Very High Resolution Radiometer (AVHRR) Global Land Dataset
  • Compare pixel-based image classification methods with segmentation techniques
  • Explain how to enhance contrast of reflectance values clustered within a narrow band of wavelengths
  • Describe an application of hyperspectral image data
  • Produce pseudocode for common unsupervised classification algorithms, including chain method, ISODATA method, and clustering
  • Calculate a set of filtered reflectance values for a given array of reflectance values and a digital image filtering algorithm
  • Describe a situation in which filtered data are more useful than the original unfiltered data
  • Perform a manual unsupervised classification given a two-dimensional array of reflectance values and ranges of reflectance values associated with a given number of land cover categories