docs: update DIAS dataset card

Authored by Danesh Selwal <selwaldanesh@gmail.com>
Co-authored-by: Claude <claude@anthropic.com>

README.md

@@ -1,17 +1,34 @@
 ---
+annotations_creators:
+- expert-generated
+language:
+- en
 license: mit
-tags:
-- hyperspectral
-- medical
-- classification
+multilinguality:
+- monolingual
+size_categories:
+- 10K<n<100K
+source_datasets:
+- DIAS
+task_categories:
+- image-classification
+- feature-extraction
+task_ids:
+- land-cover-classification
+pretty_name: DIAS 65-Band Hyperspectral Classification Dataset
 ---
-# DIAS Hyperspectral Dataset (65-Bands)
 
-## Overview
-This dataset repository contains the `DIAS.mat` and `DIAS_ref.mat` files used for hyperspectral tissue classification.
+# 65-Band DIAS (Digital Airborne Imaging Spectrometer) Dataset
 
-- **DIAS.mat**: The raw 65-band hyperspectral cube.
-- **DIAS_ref.mat**: The ground-truth reference map for tissue classification.
+## Dataset Description
+The DIAS (Digital Airborne Imaging Spectrometer) dataset is a high-resolution airborne hyperspectral imaging benchmark comprising 65 contiguous spectral bands covering the reflective optical spectrum.
 
-## Structure
-The data is natively shaped as `(H, W, Bands)` and is consumed directly by the training and evaluation pipelines in the project workspace.
+### Spectral Characteristics
+- **Bands**: 65 continuous spectral channels
+- **Spatial Grid**: Calibrated airborne radiance matrix
+- **Format**: MATLAB `.mat` containers (`DIAS.mat` and ground-truth `DIAS_ref.mat`)
+- **Classes**: Urban and agricultural land-cover classes
+
+## Citation & Attribution
+- **Dataset Curator**: Danesh Selwal (`selwaldanesh@gmail.com`)
+- **Co-authored by**: Claude (`claude@anthropic.com`)