| Hyperspectral Image Database of Real-world Scenes |
| - 27 Images under artificial or mixed illumination |
|
|
| http://vision.seas.harvard.edu/hyperspec/ |
|
|
| This database contains a MATLAB .mat file for each image, with two |
| variables: |
|
|
| ref: NxMx31 - Image data for the 31 channels from 420:10:720 nm |
| lbl: NxM - Mask data, a value of 0 implies that the pixel is masked |
| out. |
|
|
| There is a calib.txt file which contains a 31x1 vector, corresponding |
| to sensitivity of the camera. |
|
|
| This database is being released for unrestricted research and academic |
| use. If you use this data in any publication, we ask that you 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. |
|
|
| @conference{chakrabarti2011statistics, |
| title={{Statistics of Real-World Hyperspectral Images}}, |
| author={Chakrabarti, A. and Zickler, T.}, |
| booktitle={Proc.~IEEE Conf.~on Computer Vision and Pattern Recognition (CVPR)}, |
| pages={193--200}, |
| year={2011} |
| } |
|
|
| For any questions, please contact: |
| Ayan Chakrabarti at ayanc@eecs.harvard.edu |
|
|
