S C

IE N

T IF

IC H

IG H

LI G

H T

S S

T R

U C

T U

R E

O F

M A

T E

R IA

L S

1 2 5 I H I G H L I G H T S 2 0 2 2

Fig. 117: Overview of segmentation

showing the methodology for

human-driven semi-automatic identification of

regions of interest and comparison

with a trained CNN machine.

Fig. 118: Examples of machine vs. human

segmentations where: (top) the

machine performs similar to the

human, (middle) the machine outperforms

the human, and (bottom) the machine

underperforms the human. Scale bars, (left) 50 µm, (right)

20 µm.