Skip to main content
ARS Home » Research » Publications at this Location » Publication #282626

Title: Evaluating spectral measures derived from airborne multispectral imagery for detecting cotton root rot

Author
item Yang, Chenghai
item ODVODY, GARY - Texas Agrilife Research
item FERNANDEZ, CARLOS - Texas Agrilife Research
item LANDIVAR, JUAN - Texas Agrilife Research
item NICHOLS, ROBERT - Cotton, Inc

Submitted to: International Conference on Precision Agriculture Abstracts & Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 6/20/2012
Publication Date: 7/15/2012
Citation: Yang, C., Odvody, G.N., Fernandez, C.J., Landivar, J.A., Nichols, R.L. 2012. Evaluating spectral measures derived from airborne multispectral imagery for detecting cotton root rot. Proceedings, 11th International Conference on Precision Agriculture, July 15-18, 2012, Indianapolis, Indiana. CD-ROM.

Interpretive Summary: Cotton root rot is one of the most destructive plant diseases occurring throughout the southwestern United States. Recent research has shown that a commercial fungicide, flutriafol, has potential for the control of cotton root rot. This study evaluated two vegetation indices and four classification techniques for detecting cotton root rot from airborne multispectral imagery. Accuracy assessment on two-zone classification maps showed that all six methods accurately identified root rot-infected areas within the field with accuracies from 94.5 to 96.5%. The results of this study will be useful for effective detection of cotton root rot and for site-specific management of this disease.

Technical Abstract: Cotton root rot, caused by the soilborne fungus Phymatotrichopsis omnivore, is one of the most destructive plant diseases occurring throughout the southwestern United States. This disease has plagued the cotton industry for more than 100 years, but effective practices for its control are still lacking. Recent research has shown that a commercial fungicide, flutriafol, has potential for the control of cotton root rot. To effectively and economically control this disease, it is necessary to identify infected areas within the field so that variable rate technology can be used to apply fungicide only to the infected areas. The objectives of this study were to evaluate two vegetation indices, the simple ratio index (SRI) and the normalized difference vegetation index (NDVI), and four supervised classification techniques, including minimum distance, Mahalanobis distance, maximum likelihood, and spectral angle mapper (SAM), for detecting cotton root rot from airborne multispectral imagery. One cotton field with a history of root rot infection in south Texas was selected for this study. Airborne multispectral imagery with blue, green, red and near-infrared bands was taken from the field shortly before harvest when infected areas were fully expressed for the 2011 growing season. The two VIs were derived from the multispectral imagery and then statistically grouped into infected and noninfected classes. The four-band image was classified into infected and non-infected zones using the four classifiers based on training samples from the image. Accuracy assessment on the two-zone classification maps showed that all six methods accurately identified root rot-infected areas within the field with accuracies from 94.5 to 96.5%. The results of this study will be useful for effective detection of cotton root rot and for site-specific management of this disease.