Fig. 10 | Scientific Reports

Fig. 10

From: Evaluation of deep learning models using explainable AI with qualitative and quantitative analysis for rice leaf disease detection

Fig. 10

Intersection over Union of feature heatmap images with the most significant 10 features generated by LIME and ground truth: (a) Original BLB disease image; (b) Masked images with most significant 10 features; (c) Binary image of (b); (d) Ground truth image of (a); (e) Binary image of (d); (f) Intersection over Union of (c) and (e). Ground truth segmentation and masked binary image with the most significant 10 features from InceptionV3, EfficientNet B0, AlexNet, DenseNet201, Xception, VGG16, InceptionResNetV2, and ResNet50; Color coding: White: The overlapping area between the ground truth and the masked binary image. Pink: Pixels present in the masked binary image of the most significant 10 features. Green: Ground truth region.

Back to article page