Table 6 Averages and standard deviations of estimated χN for each χN class for the binarized images. Here, the χN class was same value of the teaching image.

From: Deep learning-based estimation of Flory–Huggins parameter of A–B block copolymers from cross-sectional images of phase-separated structures

 

(f, N) = (0.2, 20)

(f, N) = (0.2, 40)

(f, N) = (0.35, 20)

(f, N) = (0.35, 40)

χN = 26

26.232 ± 0.548

26.409 ± 0.898

27.229 ± 1.462

28.680 ± 1.761

χN = 28

28.241 ± 1.009

28.230 ± 1.238

29.589 ± 2.224

29.274 ± 1.909

χN = 30

30.572 ± 1.511

30.515 ± 1.546

31.547 ± 2.466

29.911 ± 1.955

χN = 32

32.807 ± 1.871

32.750 ± 1.753

33.125 ± 2.342

31.363 ± 2.133

χN = 34

34.746 ± 1.967

34.521 ± 1.750

34.124 ± 2.342

33.286 ± 2.156

χN = 36

36.330 ± 1.873

35.777 ± 1.795

34.864 ± 2.297

35.836 ± 1.832

χN = 38

37.211 ± 1.792

36.928 ± 1.876

35.076 ± 2.220

37.545 ± 1.400

χN = 40

38.005 ± 1.661

38.322 ± 1.871

37.855 ± 2.099

39.451 ± 1.043

Back to article page