Poster: Diseases of Plants: Disease Detection & Diagnosis
Use of an unmanned aerial vehicle to assess tomato spotted wilt severity in large peanut mapping populations
S. PELHAM (1), C. Holbrook (2), B. Guo (2), Y. Chu (3), P. Ozias-Akins (3), C. Li (4), R. Xu (4), A. Cubreath (1) (1) University of Georgia - Department of Plant Pathology, Tifton, GA, U.S.A.; (2) United States Department of Agriculture - Agricultural Res
Recombinant inbred line (RIL) populations of peanut (Arachis hypogaea L.) are being used to develop markers for resistance to several diseases, including tomato spotted wilt, caused by Tomato spotted wilt virus (TSWV). In efforts to develop molecular markers to assist in selection for resistance to TSWV, several populations have been developed from parents with varying levels of TSWV resistance. Mapping programs are faced with the challenge of phenotyping large numbers of genotypes in a timely manner with a limited number of people. Through the use of an unmanned aerial vehicle, imaging techniques, and ground truthing, we developed an analysis method to quickly assess spotted wilt severity in the field using selected RILs from four mapping populations of peanut. With adequate ground truthing, spotted wilt is easily differentiated from other diseases in the field by the color and the visible stunting of the plant. By using a greenness equation in Matlab to identify yellow pixels we can rate the severity of the disease in many small plots in a short amount of time. Both Matlab greenness values and severity values from APS Assess analysis of aerial images correlated well with field disease ratings. This technique allows for quick acquisition of field response data from large mapping populations.