Comparison of advanced techniques for identifying disease in apple orchards
Geography & Planning
College of Liberal Arts and Sciences
Michigan's agricultural economy is highly dependent on apple production and is the 3rd largest producer of apples in the United States. "Fire blight" disease infects tree trunks, apple blossoms and leaves and is caused by a bacterium, Erwinia amylovora. Daily visual inspection of large tree orchards for signs of fire blight is difficult and time-consuming. Landsat-5 satellite imagery have been used to detect changes in tree growth and tree disease. High-resolution satellite images have detected mountain pine beetle infestations in British Columbia forests and also delineated dead tree crowns in large oak forests in central California. This study focuses on the Fruit Ridge region in northern Kent and Ottawa counties, western Michigan, a location with susceptible apple orchards. Summer 2008 IKONOS images were utilized to examine differences in reflectance of healthy and fire blight diseased patches of apple orchards. Images coincided with field visits and GPS positioned digital photos were taken of both healthy and diseased apple trees for use in georeferencing apple orchard locations. Advanced digital imaging process techniques, image classification and other geospatial techniques were used to detect potential patches of fire blight diseased trees. The resulting analyses and maps will provide a widespread monitoring tool for apple growers and researchers to identify potential fire blight diseased areas within apple orchard regions.
107th AAG Annual Meeting
Ma, Kin, "Comparison of advanced techniques for identifying disease in apple orchards" (2010). Faculty Scholarly Dissemination Grants. 139.