RGB Compo : Tristimulus calculus

Dear all,

I use SNAP tri stimulus calculus to plot an RGB composition on Python from an OLCI L1 product.

The problem is that the image generated is foggy and misty and I don’t know the reason why.

image

image

Would you know a library to plot properly an RGB composition using the radiances ?

Thank you all !

Lucas

The reason for this is how SNap generates the images. The colour stretching depends on the scene content. On an image which has less clouds it would look hazy.
You can overcome this by adding an additional step.
You can limit the value range of th RGB channels.

red = virtual_red > 6 ? 6 : virtual_red < 4 ? 4 : virtual_red
green = virtual_green > 6.4 ? 6.4 : virtual_green < 4.1 ? 4.1 : virtual_green
blue = virtual_blue > 6.6 ? 6.6 : virtual_blue < 4.7 ? 4.7 : virtual_blue

The virtual bands are the results of the tristimulus expressions.
This will reduce the effect and images are comparable.
You might need to play around with the limits to get a good result.

This thread might also be of interest for you.

Thanks for your answer, I have limited the value range of the RGB channel and it does overcome the colour stretching.

However, I have discovered the scene function of SatPy library and the quality of the picture I get is largely higher. SatPy provides great function to generate pictures, it’s an interesting tool for my work.

Thanks again @marpet

Lucas

1 Like