Plant disease severity estimated visually: a century of research, best practices, and opportunities for improving methods and practices to maximize accuracy

Abstract Plant disease quantification, mainly the intensity of symptoms on individual units (severity), is basis for a plethora research and applied purposes in plant pathology related disciplines. These include evaluating treatment effect, monitoring epidemics, understanding yield loss, phenotyping host resistance. Although sensor technology has been available to measure severity using visible spectrum or other spectral range imaging, it visual sensing perception that still dominates, especially field research. Awareness importance accuracy estimates began 1892, when Cobb developed set diagrams as an aid guide rust wheat. Since time, various approaches, some them based principles psychophysics, have provided foundation understand sources error during estimation process well develop different scales disease-specific illustrations indicating diseased area specimens, similar by Cobb, known standard (SADs). Several rater-related (experience, inherent ability, training) technology-related (instruction, scales, SADs) characteristics shown affect accuracy. This review provides historical perspective assessment, accounting concepts, tools, changing paradigms, methods maximize estimates. A list best-operating practices quantification future topic presented current knowledge.