ESA > Join & Share > Neumapper > Neumapper



Several studies have revealed the potential of artificial neural network computational models (ANN) in processing remotely sensed imagery:

- Competitive accuracy when compared with statistical techniques like Bayesian methods or support vector machines.

- No prior knowledge necessary about the statistical distribution of the classification classes in the source data.

- Well suited for integrating multi-source, conceptually-varied data.

- Their parallel data processing capability make them fast and robust.

Neumapper implements in the same environment the various stages in the generation of an ANN for automatic pixel-based image classification:

1.- Definition of the network topology.

2.- Generation of training data.

3.- Training of the network.

4.- Classification of an image using the trained network.

A simple and intuitive interface streamlines appropriate network design and effective network training into a painless, real-time iterative process in which, after evaluating the accuracy of the resulting image classification, the user can opt for training the network further with the same or a different pattern set, and eventually adjust its topology. The interface permits separate handling of networks, pattern sets and images, as an example enabling multi-image network training, and classification of multiple images using the very same trained network.

You can download Neumapper from GEO-K's website: http://www.geo-k.co/neumapper/.



Contributors to this page: Miguel Penalver
Michele Iapaolo
Irene Fabrini .

Page last modified on Saturday 30 of May 2015 17:55:33 CEST by Miguel Penalver.

Category: Neumapper