Fig. 1From: DeepCob: precise and high-throughput analysis of maize cob geometry using deep learning with an application in genebank phenomicsPearson correlation between true and estimated cob length for three image segmentation methods (Felzenszwalb-Huttenlocher segmentation,Window-CNN,Mask R-CNN). True (x-axis) and estimated (y-axis) mean cob length (a, c) and diameter (b, d) per image with each approach, split by dataset, ImgOld and ImgNew are shown. In all cases, MaskRCNN achieves the highest correlation of at least 0.99 with the true valuesBack to article page