Problem in water detection in inland lakes



“If anyone has ideas about what the problem is and how to solve it, please let me know.”

It seems you post is not complete.
You have not told what you did. What algorithm have you applied?
Also the two images seem to be the same, except for the displayed time.
The spectra are also difficult to analyse.

Would be good if you explain your problem in more detail.

hello marco
I for S2 level c1, need to use c2rcc and c2x algorithms

my steps was

  • resambling 10m
  • subsut my AOI
  • used c2rcc the default and the same with c2x
    but when used acolite i did not losse any pixel.

the two images are the same yes.

The reason for the missing pixels might be the valid-pixel expression.
This is set to B8 > 0 && B8 < 0.1 by default. It is used to distinguish water from land and process only the water pixels. You can remove or modify it.
Your lake seems to be very turbid. This might be the reason it is considered as land and maybe there are also some clouds.
If you want to find a better expression, you can use the masks in SNAP. This is the best way to find a better expression.


EOMasters_icon_60 Marco from EOMasters - Mastering Earth Observation
Software, Knowledge, Community. Be faster, be more productive, have more fun.

2 Likes

Hello
Thanks for your reply, I used B12 as a sensitive band for turbidity. Do you see that more or less fine?

Seems to be fine to me.


EOMasters_icon_60 Marco from EOMasters - Mastering Earth Observation
Software, Knowledge, Community. Be faster, be more productive, have more fun.