Classification of weed species using color texture features and discriminant analysis

The environmental impact of herbicide utilization has stimulated research i nto new methods of weed control, such as selective herbicide application on highly infested crop areas. This research utilized the Color Go-occurrence Method (CCM) to determine whether traditional statistical discriminant ana lysis call be used to discriminate between six different classes of groundc over: The weed species evaluated were giant foxtail, crabgrass, common lamb squarter; velvetleaf; and ivyleaf morningglory, along with a soil image dat a set. The between species discriminant analysis showed that the CCM textur e statistics procedure was able to classify between five weed species and s oil with an accuracy of 93% using hue and saturation statistics, only. A si gnificant accomplishment of this work was the elimination of the intensity texture features from the model, which reduces computational requirements b y one-third.