Constructing an RGB product - how to choose channels I
The motivation for making any RGB image combination is to highlight the feature the user wants to see better than what he can achieve with individual grey-scale images.
Whether it will be a low cloud product, a convective development product or anything else, the first thing is to choose by an intelligent manner the proper satellite channels to be used in the combination. The selection of the channels should be based on the physical properties represented by the channels. It does not make sense to combine channels that represent the same physical property! Just take the 1.6+0.8+0.6 combination as an example: NIR 1.6 provides water/ice phase information, VIS0.8 provides vegetation information and VIS0.6 provides cloud optical thickness.
In the following exercise you see pairs of combination images, both of the same time and location but with different channel combinations. One is better than the other: can you define which one?
 |
 |
A | B |
| A: RGB 0.6+0.8+10.9(i) microns | B: RGB HRV+HRV+10.9 microns | ||
 |
 |
A | B |
| A: RGB (12.0-10.8)+(10.8-3.9)+10.8 microns | B: RGB 3.9(i)+10.8(i)+12.0(i) microns | ||
 |
 |
A | B |
| A:RGB (12.0-10.8)+(10.8-8.7)+10.8 microns | B:RGB 8.7(i)+10.8(i)+12.0(i)microns | ||
 |
 |
A | B |
| A:RGB (6.2-7.3)+(3.9-10.8)+(1.6-0.6) microns | B:RGB 6.2(i)+9.7(i)+13.4(i) microns |