Fusion of Images of Different Spectra Based on Generative Adversarial Networks

A method for fusing images of different spectra by using generative adversarial networks is proposed. An original architecture of a FusionNet neural network is developed based on pix2pix. It enables the synthesis of a complex (integrated) image that comprises the most informative fragments of differ...

Full description

Saved in:
Bibliographic Details
Published in:Journal of computer & systems sciences international Vol. 58; no. 3; pp. 441 - 453
Main Authors: Vizil’ter, Yu. V., Vygolov, O. V., Komarov, D. V., Lebedev, M. A.
Format: Journal Article
Language:English
Published: Moscow Pleiades Publishing 01.05.2019
Springer Nature B.V
Subjects:
Computer vision
Control
Engineering
Generative adversarial networks
Image processing
Infrared imagery
Mechatronics
Neural networks
Pattern Recognition and Image Processing
Robotics
Spectra
Test sets
ISSN:1064-2307, 1555-6530
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A method for fusing images of different spectra by using generative adversarial networks is proposed. An original architecture of a FusionNet neural network is developed based on pix2pix. It enables the synthesis of a complex (integrated) image that comprises the most informative fragments of different-spectra images, thus being more informative than any of these individual images. A technique for generating training and test sets, as well as the process of data augmentation, is described. The operation of the proposed image fusion method is demonstrated on some real-world infrared and visible images.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1064-2307
1555-6530
DOI:10.1134/S1064230719030201