Contour-based detection and quantification of tar spot stromata using RGB imagery

Authors: LEE, DA-YOUNG - Purdue University; NA, DONG-YEOP - Purdue University; GONGORA-CANUL, CARLOS - Purdue University; BAIREDDY, SRIRAM - Purdue University; LANE, BRENDEN - Purdue University; CRUZ, ANDRES - Purdue University; FERNANDEZ-CAMPOS, MARIELA - Purdue University; KLECZEWSKI, NATHAM - University Of Illinois; TELENKO, DARCY - Purdue University; Goodwin, Stephen; DELP, EDWARD - Purdue University; CRUZ, CHRISTIAN - Purdue University

Submitted to: Frontiers in Plant Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/23/2021
Publication Date: 10/1/2021
Citation: Lee, D., Na, D., Gongora-Canul, C., Baireddy, S., Lane, B., Cruz, A.P., Fernandez-Campos, M., Kleczewski, N.M., Telenko, D., Goodwin, S.B., Delp, E.J., Cruz, C. 2021. Contour-based detection and quantification of tar spot stromata using RGB imagery. Frontiers in Plant Science. 12:675975. https://doi.org/10.3389/fpls.2021.675975.
DOI: https://doi.org/10.3389/fpls.2021.675975

Interpretive Summary: Tar spot is a new disease that is causing huge losses to U.S. corn crops, yet scoring the disease in the field is difficult, time consuming, and prone to errors. To develop and test an automated scoring method, color images were analyzed with a contour-detecting computer algorithm and compared to scores made by two human raters. Several methods of statistical analysis showed that the agreement between the computer algorithm and each human rater was almost as good as that between the two human raters, indicating the potential utility of the automated approach for quantifying tar spot symptoms from color images, which complements the traditional human, visual-based disease severity estimations. These results can serve as a foundation for plant pathologists in academia and industry to build an accurate, automated, high-throughput pipeline for scoring tar spot signs in the field or greenhouse.

Technical Abstract: Quantifying symptoms of tar spot of corn has been conducted through visual-based estimations of the proportion of leaf area covered by the pathogenic structures generated by Phyllachora maydis (stromata). However, this traditional approach is costly in terms of time and labor, as well as prone to human subjectivity. An objective and accurate method, which is also time and labor efficient, is of an urgent need for tar spot surveillance and high-throughput disease phenotyping. Here, we present the use of contour-based detection of fungal stromata to quantify disease intensity using RGB images of tar spot-infected corn leaves. Image blocks (n = 1,130) generated by uniform partitioning the RGB images of leaves, were analyzed for their number of stromata by two independent, experienced human raters using Image J (visual estimates) and the experimental stromata contour detection algorithm (SCDA; digital measurements). Stromata count for each image block was then categorized into five classes and tested for the agreement of human raters and SCDA using Cohen’s weighted kappa coefficient ('). Nearly perfect agreements of stromata counts were observed for each of the human raters to SCDA (' = 0.83) and between the two human raters (' = 0.95). Moreover, the SCDA was able to recognize ‘true stromata’, but to a lesser extent than human raters (average median recall = 90.5 %, precision = 89.7 %, and dice = 88.3 %). Our results indicate the potential utility of SCDA in quantifying stromata using RGB images, complementing the traditional human, visual-based disease severity estimations, and serve as a foundation in building an accurate, high-throughput pipeline for scoring of tar spot symptoms.