Real-World Hyperspectral Images Database

This is a database of fifty hyperspectral images of indoor and outdoor scenes under daylight illumination, and an additional twenty-five images under artificial and mixed illumination. The images were captured using a commercial hyperspectral camera (Nuance FX, CRI Inc) with an integrated liquid crystal tunable filter capable of acquiring a hyperspectral image by sequentially tuning the filter through a series of thirty-one narrow wavelength bands, each with approximately 10nm bandwidth and centered at steps of 10nm from 420nm to 720nm. The camera is equipped with an apo-chromatic lens and the images were captured with the smallest viable aperture setting, thus largely avoiding chromatic aberration. All the images are of static scenes, with labels to mask out regions with movement during exposure.

This database is available for non-commercial research use. The data is available as a series of MATLAB .mat files (one for each image) containing both the images data and masks. Since the size of the download is large (around 5.5 + 2.2 GB), we ask that you send an e-mail to the authors at ayanc[at]eecs[dot]harvard[dot]edu for the download link. If you use this data in an academic publication, kindly cite the following paper:

Ayan Chakrabarti and Todd Zickler, "Statistics of Real-World Hyperspectral Images," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2011. [BibTeX]

 
[Project Page]   [Explore Database]