Fig. 3From: SeptoSympto: a precise image analysis of Septoria tritici blotch disease symptoms using deep learning methods on scanned imagesCorrelations between expert ratings, Image J and SeptoSympto outputs for necrosis and pycnidia detection in dataset 6. Comparative analysis of expert assessments, Image J and SeptoSympto for the measurements of necrosis and pycnidia on 55 bread wheat leaves scanned with the CanoScan 9000F MarkII scanner was conducted using Spearman correlations. Dataset 6 was obtained under different growing conditions, with different Z. tritici isolates and a different variety from the dataset used for the model trainingsBack to article page