Skip to content

Latest commit

 

History

History
88 lines (78 loc) · 3.42 KB

README.md

File metadata and controls

88 lines (78 loc) · 3.42 KB

SITS_S2Coverage

Sentinel-2 Coverage on Satellite Images Time Series (SITS).

“”

Source: https://sentinel.esa.int/web/sentinel/missions/sentinel-2

Based on Scene Classification Layer (SCL)

Label Classification
0 NO_DATA
1 SATURATED_OR_DEFECTIVE
2 DARK_AREA_PIXELS
3 CLOUD_SHADOWS
4 VEGETATION
5 NOT_VEGETATED
6 WATER
7 UNCLASSIFIED
8 CLOUD_MEDIUM_PROBABILITY
9 CLOUD_HIGH_PROBABILITY
10 THIN_CIRRUS
11 SNOW

Execution example for coverage calculation

Requirements

  • tqdm
  • rasterio
  • numpy
  • pandas
  • multiprocessing

In order to execute the code to perform the coverage evaluation in a dataset execute main.py. Example is shown below:

idx_targets = [0,1,2,4,5,6,7,10,11] #index to be selected (all except 3, 8 and 9)
min_spatial_coverage=50
min_temporal_coverage=50     
num_process=-1
dataset_structure_path = '/path/to/datasetstructure/seebelow.yaml'
output_dir = "/path/to/output/dir/"
results = analyze_dataset(dataset_structure_path = dataset_structure_path, 
                            idx_targets = idx_targets, 
                            min_spatial_coverage = min_spatial_coverage, 
                            min_temporal_coverage = min_temporal_coverage,
                            output_dir=output_dir,
                            num_process=num_process
                            )

It will store a .csv file with the following taxonomy: assesment_spat_X_temp_Y_sel_Z_Q.csv, where X is the min_spatial_coverage, Y is the min_temporal_coverage, Z is the idx_targets parameter filled at two digits each target (e.g. idx_targets=[1,2], then Z="0102"), and Q is the date time in python datetime format "YearMonthDayHourMinutesSeconds"

Please be aware that independently of how the data is organized, e.g. like

data
├── patch_id1
│   ├── s2_images
│   │   ├── images_id1_S2_time1.tif
│   │   ├── images_id1_S2_time2.tif
│   │   └── . . .
│   └── scl_mask
│       ├── images_id1_SCLmask_time1.tif
│       ├── images_id1_SCLmask_time2.tif
│       └── . . .
├── patch_id2
└── patch_id3
└── . . .

You still need to create a yaml file with the following structure:

patch_id1: 
    boundary_paths: null
    scl_mask_paths:
        - /path/to/images_id1_SCLmask_time1.tif
        - /path/to/images_id1_SCLmask_time2.tif
        - /path/to/images_id1_SCLmask_time3.tif
        . . .
patch_id2: 
    boundary_paths: null
    scl_mask_paths:
        - /path/to/images_id2_SCLmask_time1.tif
        - /path/to/images_id2_SCLmask_time2.tif
        - /path/to/images_id2_SCLmask_time3.tif
        . . .
. . .

The name of the files and folder does not affect the calculation. However, the yaml file with the structure of the dataset has to be created. For instance, take a look at landcovernet_structure_australia.yaml file created by prepare_landcovernet.py