title | parent | grand_parent | layout | permalink | nav_exclude | examples | |||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Observation outlier detector |
Sentinel-2 |
Sentinel |
script |
/sentinel-2/obs_outlier_detector/ |
true |
|
Once the cloudy observations have been filtered out, what is left is a mixture of valid observations and a set of undetected anomalous observations. These are mostly caused by:
- cloud shadows
- snow
- haze
The goal of the observation outlier detection algorithm is to identify anomalous observations. In this script the results of two masking algorithms is shown using two different colors:
- red: pixels detected as clouds by the cloud detector (s2cloudless).
- blue: pixels detected as outliers by the outlier detector, setting a detection threshold of 0.5.
More information about the approach in the blog post
Outlier detection over Slovenia. Acquired on 2020-09-02.