Surveying kaolin-treated cotton plots with airborne multispectral digital video imagery

Authors: Fletcher, Reginald; Showler, Allan

Submitted to: Computers and Electronics in Agriculture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/23/2006
Publication Date: 10/1/2006
Citation: Fletcher, R.S., Showler, A.T. 2006. Surveying kaolin-treated cotton plots with airborne multispectral digital video imagery. Computers and Electronics in Agriculture. 54(1):1-7.

Interpretive Summary: No information is available on employing airborne electronic imaging systems for detecting cotton canopies treated with kaolin particle film. Treatment of cotton with kaolin films has been demonstrated to be effective in controlling certain insect pests. In this study, airborne multispectral digital video imagery was evaluated as a tool for separating kaolin-treated cotton canopies from nontreated cotton canopies. Treated canopies were easily distinguished from nontreated canopies with gray scale, natural color, and false color airborne digital video images. On these images, the color of treated canopies appeared brighter than the color of nontreated canopies. In the future, imagery provided by these systems may be used in a decision support system for determining when to re-apply kaolin to cotton canopies.

Technical Abstract: Airborne multispectral digital video imagery was evaluated as tool for surveying kaolin-treated cotton (Gossypium hirsutum L.) plots, with emphasis on decision support. Images of experimental plots were obtained on 12 June 2002 and 16 June 2004 with an electronic digital video imaging system acquiring blue (447–455 nm), green (555–565 nm), red (625–635 nm), near-infrared (814–826 nm), and mid-infrared (1631–1676 nm) imagery. We qualitatively and quantitatively evaluated the single band images and color and false color composite images to compare the image responses of treated plots with the image responses of untreated plots. Blue, green, and red imagery separated the treated plots from the untreated plots (Dunnett’s test and t-test; P = 0.05), with the blue imagery being the most responsive to the treatment. Treated plots were readily differentiated from untreated plots with natural color and false color composite imagery. This separation was attributed to the images that were sensitive to some portion of the visible spectrum (i.e., blue, green, and red light). These results indicate that airborne electronic imaging systems have great potential as tools for surveying cotton fields treated with kaolin particle film, supporting application of the imagery as a decision support tool.