spaceeye_interpolation

Runs the SpaceEye cloud removal pipeline using an interpolation-based algorithm, yielding daily cloud-free images for the input geometry and time range. The workflow fetches Sentinel-2 tiles that cover the input geometry and time range, preprocesses them, computes cloud masks, and runs SpaceEye inference in a sliding window on the retrieved tiles. This workflow can be reused as a preprocess step in many applications that require cloud-free Sentinel-2 data. For more information about SpaceEye, read the [link=https://arxiv.org/abs/2106.08408]paper: https://arxiv.org/abs/2106.08408[/link].

graph TD inp1>user_input] out1>raster] tsk1{{preprocess}} tsk2{{spaceeye}} tsk1{{preprocess}} -- raster/s2_rasters --> tsk2{{spaceeye}} tsk1{{preprocess}} -- mask/cloud_rasters --> tsk2{{spaceeye}} inp1>user_input] -- user_input --> tsk1{{preprocess}} inp1>user_input] -- input_data --> tsk2{{spaceeye}} tsk2{{spaceeye}} -- raster --> out1>raster]

Sources

user_input: Time range and geometry of interest.

Sinks

raster: Cloud-free rasters.

Parameters

duration: Time window, in days, considered in the inference. Controls the amount of temporal context for inpainting clouds. Larger windows require more compute and memory.
time_overlap: Overlap ratio of each temporal window. Controls the temporal step between windows as a fraction of the window size.
min_tile_cover: Minimum RoI coverage to consider a set of tiles sufficient.
max_tiles_per_time: Maximum number of tiles used to cover the RoI in each date.
cloud_thr: Confidence threshold to assign a pixel as cloud.
shadow_thr: Confidence threshold to assign a pixel as shadow.
pc_key: Optional Planetary Computer API key.

Tasks

preprocess: Downloads and preprocesses Sentinel-2 imagery that covers the input geometry and time range, and computes improved cloud masks using cloud and shadow segmentation models.
spaceeye: Performs temporal damped interpolation to generate daily cloud-free images given Sentinel-2 data and cloud masks.

Workflow Yaml

name: spaceeye_interpolation
sources:
  user_input:
  - preprocess.user_input
  - spaceeye.input_data
sinks:
  raster: spaceeye.raster
parameters:
  duration: null
  time_overlap: null
  min_tile_cover: null
  max_tiles_per_time: null
  cloud_thr: null
  shadow_thr: null
  pc_key: null
tasks:
  preprocess:
    workflow: data_ingestion/sentinel2/preprocess_s2_improved_masks
    parameters:
      min_tile_cover: '@from(min_tile_cover)'
      max_tiles_per_time: '@from(max_tiles_per_time)'
      cloud_thr: '@from(cloud_thr)'
      shadow_thr: '@from(shadow_thr)'
      pc_key: '@from(pc_key)'
  spaceeye:
    workflow: data_ingestion/spaceeye/spaceeye_interpolation_inference
    parameters:
      duration: '@from(duration)'
      time_overlap: '@from(time_overlap)'
edges:
- origin: preprocess.raster
  destination:
  - spaceeye.s2_rasters
- origin: preprocess.mask
  destination:
  - spaceeye.cloud_rasters
description:
  short_description: Runs the SpaceEye cloud removal pipeline using an interpolation-based
    algorithm, yielding daily cloud-free images for the input geometry and time range.
  long_description: 'The workflow fetches Sentinel-2 tiles that cover the input geometry
    and time range, preprocesses them, computes cloud masks, and runs SpaceEye inference
    in a sliding window on the retrieved tiles. This workflow can be reused as a preprocess
    step in many applications that require cloud-free Sentinel-2 data. For more information
    about SpaceEye, read the [link=https://arxiv.org/abs/2106.08408]paper: https://arxiv.org/abs/2106.08408[/link].'
  sources:
    user_input: Time range and geometry of interest.
  sinks:
    raster: Cloud-free rasters.
  parameters:
    duration: Time window, in days, considered in the inference. Controls the amount
      of temporal context for inpainting clouds. Larger windows require more compute
      and memory.
    time_overlap: Overlap ratio of each temporal window. Controls the temporal step
      between windows as a fraction of the window size.
    min_tile_cover: Minimum RoI coverage to consider a set of tiles sufficient.
    max_tiles_per_time: Maximum number of tiles used to cover the RoI in each date.
    cloud_thr: Confidence threshold to assign a pixel as cloud.
    shadow_thr: Confidence threshold to assign a pixel as shadow.
    pc_key: Optional Planetary Computer API key.