Skip to main content

Table 9 Chalkiness segmentation results of the weakly supervised Grad-CAM approach with ResNet-101 as backbone on unpolished rice

From: Deep learning based high-throughput phenotyping of chalkiness in rice exposed to high night temperature

Grad-CAM (ResNet-101) GT-known Loc. Acc. (%) Loc. Acc. (%) Avg. IoU (%) Layer T (%)
polished model 7.92 = 019/240 7.92 = 19/240 26.79 layer2_0_
conv2
60
unpolished model 63.75 = 153/240 63.75 = 153/240 51.76 layer2_0_
conv2
60
mixed model 20.42 = 049/240 20.42 = 049/240 29.91 layer2_3_
conv2
60
  1. Only 240 chalky seed images in the Unpolished (12) test set were used for chalkiness segmentation evaluation. Performance is reported using the following metrics: Ground-Truth Localization Accuracy (GT-known Loc. Acc.), which represents the fraction of ground-truth chalky seed images with \(\text{ IoU } \ge 0.5\); Localization Accuracy (Loc. Acc.), which represents the fraction of ground-truth chalky images, with \(\text{ IoU } \ge 0.5\), correctly predicted by the model; Average IoU (Avg. IoU), which represents the average IoU for the set of chalky seed images. To calculate the IoU, the mask of the predicted chalkiness is obtained using a threshold \(T=60\%\) of the maximum pixel intensity. The last two columns show the layer that was used for generating the heatmap and the threshold used to binarize the heatmap when calculating IoU, respectively