Object-Oriented Classification of ASTER Imagery for Landcover Mapping in Monsoonal Northern Australia

in

This paper looks at the suitability of object-oriented analysis of ASTER imagery for the classification of land cover within the Top End of the Northern Territory. Classification based purely on spectral values of pixels has its limitations. The nature of land cover in tropical Australia can lead to problems in classification using conventional pixel-based methods; for example, the misclassification of spectrally similar but compositionally different land covers as well as the difficulty in classifying spectrally heterogeneous cover.

Document Type: 
Scientific Paper

Fire Mapping for Managers in North Australia Savanna; An Object Oriented Classification Model for MODIS Imagery

in

Fire is a conspicuous, widespread and frequent disturbance process in the savannas of north Australia. Fire can be a powerful land management tool if understood. Frequent and accurate fire scar mapping can greatly enhance the functionality of fire as a tool by giving managers regular feedback on their management actions, providing burnt area information and updating fire histories.

Document Type: 
Scientific Paper