Gold University of Minnesota M. Skip to main content.University of Minnesota. Home page.
Home > Research > Forest Inventory and Management > Oak Wilt Detection

Oak Wilt Detection

We are currently evaluating the use of IKONOS imagery to detect the spread of Oak Wilt disease in the City of Eagan in Minnesota.

After separating forest pixels from other land cover classes using an object classifier, areas of potential disease are identified using the Normalized Difference Vegetation Index (NDVI). Field tests verified that oak wilt was correctly identified in 9 out of 10 sites previously unknown to Eagan Forestry Division.

Below shows is a map created by using eCognition software to generate a simple land cover classification. Forest pixels were then examined for traces of Oak Wilt using NDVI. Click on each map to enlarge.

 

Satellite-Aided Field Forestry Applications

k-Nearest Neighbor (kNN)

Forest Disturbance Mapping

Oak Wilt Detection

Urban Forest Mapping

TCMA Classification Comparison: MLC vs. kNN


Above: Symptoms of Oak Wilt