Share Email Print
cover

Proceedings Paper

Semantic image segmentation for information presentation in enhanced vision
Author(s): Oleg V. Vygolov; Vladimir S. Gorbatsevich; Nikita A. Kostromov; Maxim A. Lebedev; Yury V. Vizilter; Vladimir A. Knyaz; Sergey Y. Zheltov
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

The paper proposes a semantic segmentation algorithm based on Convolutional Neural Networks (CNN) related to the problem of presenting multispectral sensor-derived images in Enhanced Vision Systems (EVS). The CNN architecture based on residual SqueezeNet with deconvolutional layers is presented. To create an in-domain training dataset for CNN, a semi-automatic scenario with the use of photogrammetric technique is described. Experimental results are shown for problem-oriented images, obtained by TV and IR sensors of the EVS prototype in a set of flight experiments.

Paper Details

Date Published: 5 May 2017
PDF: 8 pages
Proc. SPIE 10197, Degraded Environments: Sensing, Processing, and Display 2017, 101970H (5 May 2017); doi: 10.1117/12.2262507
Show Author Affiliations
Oleg V. Vygolov, State Research Institute of Aviation Systems (Russian Federation)
Vladimir S. Gorbatsevich, State Research Institute of Aviation Systems (Russian Federation)
Nikita A. Kostromov, State Research Institute of Aviation Systems (Russian Federation)
Maxim A. Lebedev, State Research Institute of Aviation Systems (Russian Federation)
Yury V. Vizilter, State Research Institute of Aviation Systems (Russian Federation)
Vladimir A. Knyaz, State Research Institute of Aviation Systems (Russian Federation)
Sergey Y. Zheltov, State Research Institute of Aviation Systems (Russian Federation)


Published in SPIE Proceedings Vol. 10197:
Degraded Environments: Sensing, Processing, and Display 2017
John (Jack) N. Sanders-Reed; Jarvis (Trey) J. Arthur, Editor(s)

© SPIE. Terms of Use
Back to Top