Methods of detection in satellite images of illegal dumps by using a method based on tree classifier

The advantages of using a tree-like neural network classifier consisting of a set of simple (binary) classifiers in problems of detection of unauthorized garbage dumps. We consider the use of principal component analysis as a dimension reduction algorithm space to treat the symptoms. The results of...

Full description

Saved in:

Bibliographic Details
Published in:	2017 6th Mediterranean Conference on Embedded Computing (MECO) pp. 1 - 3
Main Authors:	Akinina, Natalia V., Akinin, Maxim V., Taganov, Alexandr I., Nikiforov, Michael B.
Format:	Conference Proceeding
Language:	English
Published:	IEEE 01-06-2017
Subjects:	a binary classifier AdaBoost algorithm Algorithm design and analysis Classification algorithms Clustering algorithms conjugate gradient method Haralik's textural features k-means clustering neural network classifier tree Neural networks Parzen's window method Principal component analysis Satellites
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	The advantages of using a tree-like neural network classifier consisting of a set of simple (binary) classifiers in problems of detection of unauthorized garbage dumps. We consider the use of principal component analysis as a dimension reduction algorithm space to treat the symptoms. The results of experimental studies demonstrating the effectiveness of using this method of detection of unauthorized dumps.
DOI:	10.1109/MECO.2017.7977179