Update README.md

README.md

@@ -7,65 +7,67 @@ language:
 tags:
 - climate
 - biology
+- remote-sensing
+- geospatial
+- multimodal
+- crisis-management
+- climate
 pretty_name: CrisisLandMark
 size_categories:
 - 100K<n<1M
 ---
 # CrisisLandMark
 
-CrisisLandMark is a large-scale corpus composed of Sentinel-1 and  Sentinel-2 data enriched with textual and geospatial annotations 
-designed for LULC and crisis management applications. The primary purpose is to help design text-to-image retrieval pipelines.
+**CrisisLandMark** is a large-scale, multimodal corpus for Text-to-Remote-Sensing-Image Retrieval (T2RSIR). It contains over 647,000 Sentinel-1 (SAR) and Sentinel-2 (multispectral optical) images enriched with structured textual and geospatial annotations.
+The dataset is designed to move beyond standard RGB imagery, enabling the development of retrieval systems that can leverage the rich physical information from different satellite sensors for applications in Land Use/Land Cover (LULC) mapping and crisis management.
 
 - **Curated by:** Daniele Rege Cambrin
 - **License:** Creative Commons Attribution Non Commercial 4.0
 - **Repository:** [GitHub](https://github.com/DarthReca/closp)
 - **Paper:** [Arxiv](https://arxiv.org/abs/2507.10403)
 
+## Getting Started
+
+To work with the dataset, you will need the `h5py` and `hdf5plugin` packages.
+
 ## Dataset Structure
 
 <!-- This section provides a description of the dataset fields, and additional information about the dataset structure such as criteria used to create the splits, relationships between data points, etc. -->
-
-The *crisislandmark.h5* file can be used to access all the keys. Since it contains only [ExternalLinks](https://docs.h5py.org/en/stable/high/group.html#external-links), the other files are necessary.
+The dataset is distributed in HDF5 format. The main file, *crisislandmark.h5*, contains external links to data shards, so all *.h5* files must be kept in the same directory for access. Each key in the HDF5 file corresponds to a unique sample.
 Each key contains the following matrices:
 
 | Key      | Shape           | Data Type | Description                                           |
 | :------- | :-------------- | :-------- | :---------------------------------------------------- |
-| `image`  | `(B, 120, 120)` | `float32` | Contains the B bands of the image from S1 or S2.      |
-| `coords` | `(2, 120, 120)` | `float32` | Contains the coordinates for each pixel of the image. |
+| `image`  | `(B, 120, 120)` | `float32` | The image data. B is the number of bands: 2 for Sentinel-1 (VV, VH) or 12 for Sentinel-2.      |
+| `coords` | `(2, 120, 120)` | `float32` | Contains the x and y coordinates for each pixel of the image. |
 
 and the following attributes:
 
 | Key         | Shape | Data Type   | Description                                            |
 | :---------- | :---- | :---------- | :----------------------------------------------------- |
-| `crs`       | `(1)` | `float32`   | Contains the CRS for the coords matrix.                |
+| `crs`       | `(1)` | `float32`   | The EPSG code for the Coordinate Reference System.                |
 | `timestamp` | `(1)` | `float32`   | Contains the associated timestamp if available.        |
 | `labels`    | `(L)` | `list[str]` | Contains the list of labels associated with the image. |
 
 The *metadata.parquet* files contain, for each key, the associated split and the labels from the original source (either CLC or DW).
 
-**NOTE**: remember to install *hdf5pluging* package to work with the dataset.
-
-## Source Data
-
-<!-- This section describes the source data (e.g. news text and headlines, social media posts, translated sentences, ...). -->
-The CrisisLandMark dataset is composed of over 647,000 satellite images from two primary sources: Sentinel-1 and Sentinel-2 missions.
-
-- Sentinel-1: These are Synthetic Aperture Radar (SAR) Ground Range Detected (GRD) products, which provide all-weather, day-and-night observations. The data includes two polarization channels: Vertical transmit-Vertical receive (VV) and Vertical transmit-Horizontal receive (VH).
-- Sentinel-2: These are Level-2A (L2A) surface reflectance optical products, providing data across 12 spectral bands from ultra-blue to short-wave infrared.
+## Data split
+The dataset is divided into two splits based on a stratified multi-label sampling strategy to ensure similar label distribution:
 
-The images are sourced from five existing datasets: re-BEN, CaBuAr, QuakeSet, MMFlood, and Sen12Flood. All images have been processed to a uniform spatial resolution of 10 meters and divided into 120x120 pixel patches.
+- Training Set: 20% of the data, intended for model training.
+- Corpus Set: 80% of the data, intended for retrieval and evaluation.
 
-### Annotations
+## Source Data and Annotations
 
-<!-- If the dataset contains annotations which are not part of the initial data collection, use this section to describe them. -->
-The dataset features a comprehensive set of structured textual and geospatial annotations created specifically for this work. 
-The annotations were generated by harmonizing established land cover classification systems and crisis-specific tags.
+The dataset is built from satellite images sourced from Sentinel-1 and Sentinel-2 missions. 
+The raw images were drawn from five existing public datasets: re-BEN, CaBuAr, QuakeSet, MMFlood, and Sen12Flood. 
+All images were processed to a uniform 10-meter spatial resolution and divided into 120x120 pixel patches.
 
-- For general land cover scenes, annotations were derived from the detailed CORINE Land Cover (CLC) system.
-- For images related to crisis events, LULC annotations were generated by querying the near-real-time global Dynamic World (DW) system.
-- For images sourced from crisis-specific datasets (CaBuAr, MMFlood, Sen12Flood, Quakeset), the original event tags such as "wildfire", "flooding", and "earthquake" were retained and incorporated.
+The annotations are created from the following sources:
 
-*NOTE*: In *mapping.py*, you can find the mapping between CLC and DW used in the original settings.
+- Land Use/Land Cover (LULC): Annotations were derived from the CORINE Land Cover (CLC) system for European regions and the global, near-real-time Dynamic World (DW) system for crisis-event images. The script mapping.py in the repository details the mapping between CLC and DW.
+- Crisis Events: For images from crisis-focused datasets, original event tags like "wildfire", "flooding", and "earthquake" were retained.
+- Geospatial: Every image patch is annotated with its geographic coordinates.
 
 ## Citation