3 C. Studholme, D. L. G. Hill, and D. J. Hawkes, “An Overlap Invariant Entropy Measure of 3D Medical Image
Alignment,” Pattern Recognition 32, no. 1 (1999): 71–86.
for ax in (ax0, ax1):
ax.axis('off')
*** Aligning blue to green ***
Level: 1, Angle: -0.0474, Offset: (-0.867, 15.4), Cost: 0.0499
** Aligning red to green ***
Level: 1, Angle: 0.0339, Offset: (-0.269, -8.88), Cost: 0.0311
The alignment is a little bit better than with the raw images (Figure 7-6), because the
red and the green channels are correctly aligned, probably thanks to the giant yellow
patch of sky. However, the blue channel is still off, because the bright spots of blue ...