Visual assessment of disease severity often lacks precision and is prone to many biases due to the subjectivity of assessors. In contrast, phenotyping based on automated image analysis should help overcome such limitations and increase thoughputs as well, compared to phenotyping based on visual assessments.

Most stresses, including biotic stresses, alter the yield of chlorophyll fluorescence. Using chlorophyll a measurement of chlorophyll fluorescence yield is associated with each pixel of the leaf. For each of these measurements, the algorithm that we developed attributes a probability that it corresponds to a «healthy» or «diseased» class of tissues. The symptomatic area on the leaf then corresponds to the number of «diseased» pixels.

The stronger the symptoms are, the lower the yield of chlorophyll fluorescence will be. Among the «diseased» pixels, several classes are determined according to the chlorophyll fluorescence yield. These classes thus represent various stages of plant tissue alteration. Therefore, the measurement of symptom intensity is a complement to the measurement of the total symptomatic leaf area.

Our technique does not involve visual scoring by trained assessors. It can be adapted to numerous pathosystems and, as a result, has a broad potential for phenotyping plant resistance to pathogens.

Contact :
Tristan BOUREAU, IRHS unit
tristan.boureau@univ-angers.fr

Références :
Rousseau C., Belin E., Bove E., Rousseau D., Fabre F., Berruyer R., Guillaumès J., Manceau C., Jacques M.A., Boureau T. (2013). High throughput quantitative phenotyping of plant resistance using chlorophyll fluorescence image analysis. Plant Methods 2013, 9:17.