Skip to content

Dataset · ICVL-BGU

Hyperspectral images of natural and urban scenes

ICVL-BGU/ICVL_HS_2016

Collection of 202 high-resolution hyperspectral images in the 400-1000 nm range for computer vision tasks.

What's inside

The dataset contains 202 high-resolution hyperspectral images (400-1000 nm) captured with a Specim PS Kappa DX4 camera at Ben-Gurion University. The selection includes urban and rural landscapes, interiors, plants, and everyday objects.

What you can do with it

  • Training models for hyperspectral image reconstruction from regular RGB photos
  • Feature extraction and material segmentation based on spectral characteristics
  • Testing remote sensing data processing algorithms
Tasks
Image → image, Image feature extraction
Size
up to 1K rows
Downloads this month
2.8K
Likes
17
Access
on request on Hugging Face

How to load it

from datasets import load_dataset

ds = load_dataset("ICVL-BGU/ICVL_HS_2016")