Skip to main content

Sensitivity of hyperclustering and labelling land cover classes to Landsat image acquisition date

Buy Article:

$63.00 plus tax (Refund Policy)


Seven Landsat images near Prince George, British Columbia, Canada, representing a range of within-year and between-year dates, were acquired to assess spectral variability and the concomitant impact upon hyperclustering classification results. Top-of-atmosphere (TOA) radiometric corrections, dark target subtractions and geometric corrections were applied to the imagery. Following application of an unsupervised hyperclustering procedure which employed the K-means classifier, post-classification comparisons examined the differences in spectral response patterns for several target classes, and area summaries were generated to compare the variability in the total area of classes as identified in each image. Finally, the kappa coefficient of agreement was used to quantify the degree of correspondence between the classified images. The results indicated that the spectral variability of the within-year image set exceeded the variability in the between-year image set and differences in class area were highly variable over the range of image acquisition dates. These findings suggest that off-year imagery (acquired on or near anniversary dates) may be preferred to off-season imagery when building large-area Landsat mosaics for land cover classification using the hyperclustering procedure.

Document Type: Research Article


Publication date: 2004-12-01

More about this publication?
  • Access Key
  • Free ContentFree content
  • Partial Free ContentPartial Free content
  • New ContentNew content
  • Open Access ContentOpen access content
  • Partial Open Access ContentPartial Open access content
  • Subscribed ContentSubscribed content
  • Partial Subscribed ContentPartial Subscribed content
  • Free Trial ContentFree trial content
Cookie Policy
Cookie Policy
Ingenta Connect website makes use of cookies so as to keep track of data that you have filled in. I am Happy with this Find out more