APS Homepage

POSTERS: Crop loss assessment

Color- and metric-based k-means clustering (CM-KMC) for quantification of rice sheath blight infection
Da-Young Lee - Department of Plant Pathology, The Ohio State University. Guo-Liang Wang- The Ohio State University, Dong-Yeop Na- The Ohio State University

Accurate quantification of disease severity is an essential element in breeding for disease resistance. Currently existing approaches in measuring severities of major rice diseases, such as rice sheath blight (ShB), greatly rely on visual-based ratings which are subjective and time-consuming. In our study, we developed an algorithm which takes into account the color similarity and spatial proximity of necrotic ShB disease lesions. Using red-green-blue (RGB) images of ShB-infected rice plants as input, we segmented the RGB images utilizing our algorithm followed by image identification. Quantification of ShB disease severity in ShB-resistant and susceptible cultivars, Jasmine 85 and Lemont, respectively, verified the performance our algorithm. The outcomes of this study will be useful in providing accurate measurements of disease severity for ShB resistance screening programs and will be foundational in the development of automated platforms for ShB detection and quantification in infecting rice samples.