CLASSIFICATION ACCURACY OF DISCRIMINANT ANALYSIS, ARTIFICIAL NEURAL NETWORKS, AND DECISION TREES FOR WEED AND NITROGEN STRESS DETECTION IN CORN

Abstract

Hyperspectral images of experimental plots, cropped with corn (Zea mays L.) and to which twelve combinationsof three nitrogen application rates and four weed management strategies were applied, were obtained with a 72-wavebandcompact airborne spectrographic imager (CASI). The images were taken at three times during the 2000 growing season: earlygrowth, tasseling, and full maturity. Nitrogen application rates were 60, 120, and 250 kg N ha-1. Weed controls were: none,control of grasses, control of broadleaf weeds, and full weed control. The objective of this study was to evaluate discriminantanalysis as a tool for classifying images with respect to the nitrogen and weed management practices applied to theexperimental plots, and to compare the classification accuracy of this technique with those obtained by artificial neuralnetwork (ANN) and decision tree (DT) algorithms on the same data. Significant wavebands were selected, among the 72available, using the stepwise option of the STEPDISC procedure (SAS software). Classification accuracy was determined forthe full set of selected wavebands and for subsets thereof, for three problems: distinguishing between the 12 combinationsof factor levels, differentiating between nitrogen levels only, and separating weed controls only. Misclassification rates ofimages, taken at the initial growth stage, were substantially lower for each of these tasks (25%, 17%, and 13%, respectively)when discriminant analysis was used. The ANN approach was best for images taken at the tasseling and full maturity stages.However, from the precision-farming point of view, it is easier to apply site-specific remedies to weed and nitrogen stressesearly in the season than when the corn crop has reached the tasseling stage, so the results obtained with the discriminantanalysis are noteworthy.