Figure 3 | Scientific Reports

Figure 3

From: A deep learning approach for semantic segmentation of unbalanced data in electron tomography of catalytic materials

Figure 3

A set of ground truth and predicted segmentations of γ-Alumina and Pt NPs from the validation data and corresponding false negative and false positive maps show the discrepancies in the overlap of each class. The images in (a) and (b) represent two 512 × 512 pixels size patches from the validation data set; for illustrative purpose, each set of images demarcated with a black frame. Upper rows are segmentation results from the γ-Alumina particle and lower rows are from the Pt NPs of the same patch.

Back to article page