*******************************************************************************
Potato Hyperspectral Images Database

Laboratory of Applied Computer Science (LIA2)
Universidade de Vigo, Spain
http://lia.ei.uvigo.es

*******************************************************************************
INTRODUCTION

The purpose of this database (md5-checksum on server) to provide to the
research community a set of hyperspectral images which we used to detecting
the hollow heart in potatoes using hyperspectral information in the infrared
spectra (900nm to 1700nm). This database has been used in some articles and
in a PhD dissertation:

http://www.angeldacal.com/files/publications/angeldacal_2011_CAIP.pdf (English)
http://www.angeldacal.com/files/tesis_angeldacal.pdf (Spanish)

The images have been acquired using a hyperspectral system composed by:
- Camera Xenics Xeva-1.7-320 (sensor InGaAs, 900nm - 1700nm, 320x256, USB 2.0)
- Specim Imspector N17E (900nm - 1700nm, spectral resolution 5nm)
- Mirror Scanner Specim
- Halogen lamps
- Optic table and other optic components

*******************************************************************************
INFORMATION

- The database is composed by 468 hyperspectral cubes (numbered from 1 to 468).
- Each hyperspectral cube is composed by 256 monochrome images (320x320 pixels)
  in .bmp format.
- Each image in a hyperspectral cube is identified to a wavelength,
  from 0839.95 nm to 1673.91 nm.
- Each folder (hyperspectral cube) is followed by a 'CH' or a 'N'.
  A 'N' means that the potato is healthy (there are 260 healthy potatoes).
  A 'CH' means that the potato has a hollow heart (there are 208 hollow
  heart affected potatoes).

*******************************************************************************
REFERENCE

Please, reference this database using the following publication:

A. Dacal-Nieto, A. Formella, P. Carrión, E. Vázquez-Fernandez,
M. Fernández-Delgado.
Non-destructive detection of hollow heart in potatoes using hyperspectral
imaging.
Lecture Notes in Computer Science (LNCS) 6588, ISBN 978-3-642-23677-8,
pp. 180-187, 2011. ISBN: 978-3-642-23677-8.

You can access our results and compare using the following resources:

http://www.angeldacal.com/files/publications/angeldacal_2011_CAIP.pdf (English)
http://www.angeldacal.com/files/tesis_angeldacal.pdf (Spanish)


*******************************************************************************
CONTACT

If you want to contact with us, please use the following addresses:

Dr. Arno Formella (formella@ei.uvigo.es - http://lia.ei.uvigo.es )
Dr. Angel Dacal-Nieto (angeldacalnieto@gmail.com - http://angeldacal.com )

*******************************************************************************