(Spectral-spatial COrRelation SegmenTAtion based ClassifiEr

SoCRATE Learner

A short description
The distribution package
Related publications
Authors & Aknowledgement
Contact

A short description

Classifying every pixel of an hyperspectral image with a certain land-cover type is the cornerstone of hyperspectral image analysis. In the present study, a segmentation-aided methodology for the spectral-spatial classification of the hy- perspectral data is proposed, in order to take care of the spatial dependence of the spectral bands, deal with curse of dimensionality and handle the spectral variability. A local spatial regularization of spectral information is used, in order to derive an informative joint spectral-spatial representation of data. A contiguity-based segmentation algorithm is formulated, in order to build the object-wise texture that can aid classifier learning. The hybrid use of the seg- mentation texture in both the pre-processing (i.e to select representative pixels to learn the classifier) and post-processing (i.e. to refine predicted labels and remove possible outlier classifications) is evaluated.

The distribution package

SoCRATE has been implemented in a Java system.

file	Description
README	A description file
Software and Data	This bundle contains the executable jar and example of executions with Indian Pines, Pavia University and Salinas Valley datasets

Warning: The SoCRATE Learner is free for evaluation, research and teaching purposes, but not for commercial purposes.
Please Acknowledge

Top of this page

Related publications

A. Appice, D. Malerba, Segmentation-aided classification of hyperspectral data using spatial dependency of spectral bands, ISPRS Journal of Photogrammetry and Remote Sensing (2018)

Top of this page

Project team

Dr. Annalisa APPICE

Prof. Donato MALERBA

Contact

Name	Email address	Tel. number	Fax
Annalisa Appice	annalisa.appice@uniba.it	+39 080 5443262	+39 080 5443262

Top of this page