Notes for the use of the '2007_LANDSAT_ERS_Pavia.mat'

These data were those provided for the IEEE Data Fusion Contest 2007 (http://www.grss-ieee.org/past-data-fusion-contests/).


-------------------------------
ABOUT USING THE DATA
-------------------------------

The data are provided to the community free of charge for research purposes. 

When using these data, please comply with the following rules:

1) Add the following ot the acknowledgment sectio of your paper: 'The authors would like to acknowledge Prof. Paolo Gamba for providing the data and the IEEE Image Analysis and Data Fusion Technical Committee of the IEEE for organizing the Data Fusion Contest 2007.'

2) cite the contest' results paper:
F. Paciﬁci, F. Del Frate, W. J. Emery, P. Gamba, J. Chanussot, “Urban mapping using coarse SAR and optical data: outcome of the 2007 GRS-S data fusion contest”, IEEE Geoscience and Remote Sensing Letters, vol. 5, no. 3, pp. 331-335, July 2008.

3) Name the dataset 'grss_dfc_2007' and add a citation to the dataset such as:
	2007 IEEE GRSS Data Fusion Contest, Online: http://www.grss-ieee.org/community/technical-committees/data-fusion/
 

4) Send a draft of your paper to iadf_chairs@grss-ieee.org, so that the current chairs can check the compliance of the use of the data.


-------------------------------
ABOUT THE DATA
-------------------------------

The data are provided in two formats


A) a single .mat file (MATLAB). It contains the following variables:

- Landsat1994: the landsat image of 1994
- Landsat2000: the landsat image of 2000

- ERS92_aug_13: ERS SAR image acquired on August 13, 1992.
- ERS92_oct_22: ERS SAR image acquired on October 22, 1992.
- ERS93_jun_24
- ERS93_nov_11
- ERS94_nov_09
- ERS94_oct_03
- ERS95_aug_27
- ERS95_jul_22
- ERS95_jul_23


- GT the ground truth, defined by manual photointerpretation
- cmap: colormap for plotting the classification results and the ground truth. 			It is a structure with the following fields:
	- colors: colomap values
	- labels: names of the classes
	
	
B) The original ENVI files, named after the sensor and acquisition. The ground truth is in this case provided as a .bmp image.