nikolkoo commited on
Commit
c96f12e
·
verified ·
1 Parent(s): afde3f6

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +71 -0
README.md CHANGED
@@ -16,4 +16,75 @@ configs:
16
  data_files:
17
  - split: train
18
  path: data/train-*
 
 
 
 
 
 
 
 
 
 
 
19
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
16
  data_files:
17
  - split: train
18
  path: data/train-*
19
+ license: cc-by-4.0
20
+ task_categories:
21
+ - image-segmentation
22
+ tags:
23
+ - geodata
24
+ - satellite
25
+ - sentinel2
26
+ - ESA
27
+ pretty_name: Simple Satelite Segmentation (Norway)
28
+ size_categories:
29
+ - n<1K
30
  ---
31
+
32
+ # Satellite Segmentation
33
+
34
+ Summary
35
+ -------
36
+ This dataset contains paired Sentinel-2 RGB tiles and corresponding land-cover masks (derived from ESA WorldCover) prepared for semantic segmentation. Each example has:
37
+ - `image`: RGB image (PNG)
38
+ - `mask`: integer-labelled mask (PNG, uint8)
39
+
40
+ Key details
41
+ -----------
42
+ - Source imagery: Sentinel-2 L2A via Microsoft Planetary Computer (STAC)
43
+ - Land-cover masks: ESA WorldCover (derived)
44
+ - Spatial resolution: 10 m (aligned to Sentinel-2 grid)
45
+ - CRS: EPSG:4326
46
+ - Number of samples: 790
47
+ - Train/validation split: Train
48
+
49
+ Provenance & license
50
+ --------------------
51
+ This dataset was derived from third‑party datasets:
52
+ - Sentinel‑2 (Copernicus)
53
+ - ESA WorldCover
54
+
55
+ The user of this dataset must respect the original licenses and terms of use. The repository contains derived files (tiles).
56
+
57
+ Data format
58
+ -----------
59
+ - Images: PNG, RGB, 3 channels
60
+ - Masks: PNG, integer values representing classes (do not normalize/convert to RGB)
61
+ - Filenames: `{prefix}.png` and `{prefix}_mask.png` (paired by prefix)
62
+
63
+ How to load
64
+ -----------
65
+ Example (datasets library):
66
+
67
+ ```python
68
+ from datasets import load_dataset
69
+ ds = load_dataset("nikolkoo/SateliteSegmentation")
70
+ ```
71
+
72
+ Example evaluation/training snippet
73
+ -----------------------------------
74
+ Use CrossEntropyLoss with logits and integer masks:
75
+
76
+ ```python
77
+ # pseudocode
78
+ images = batch["image"] # (B,H,W,3) -> to tensor & permute
79
+ masks = batch["mask"] # (B,H,W) ints
80
+ logits = model(images)
81
+ loss = torch.nn.CrossEntropyLoss()(logits, masks)
82
+ ```
83
+
84
+ Citation
85
+ --------
86
+ If you publish results using this dataset, cite the original data providers (Copernicus / ESA / Microsoft Planetary Computer) and this dataset repo.
87
+
88
+ Contact
89
+ -------
90
+ Feel free to add a comment in the Community 🤗