Open Access

Digital Camouflage Generation Based on an Improved CycleGAN Network Model

Leixiang Xia

Leixiang Xia

School of Computer Science and Engineering, Xi’an Technological UniversityXi’an, China
Search for this author on
Sciendo|Google Scholar
Xia, Leixiang
, 
Jun Yu

Jun Yu

School of Computer Science and Engineering, Xi’an Technological UniversityXi’an, China
Search for this author on
Sciendo|Google Scholar
Yu, Jun
, 
Kuncai Jiang

Kuncai Jiang

School of Computer Science and Engineering, Xi’an Technological UniversityXi’an, China
Search for this author on
Sciendo|Google Scholar
Jiang, Kuncai
, 
Zhiyi Hu

Zhiyi Hu

Engineering Design Institute, Army Research LoboratoryBeijing, China
Search for this author on
Sciendo|Google Scholar
Hu, Zhiyi
 and 
Yunshan Xie

Yunshan Xie

School of Computer Science and Engineering, Xi’an Technological UniversityXi’an, China
Search for this author on
Sciendo|Google Scholar
Xie, Yunshan
  
Jul 21, 2024
Digital Camouflage Generation Based on an Improved CycleGAN Network Model's Cover Image
About this article
Previous Article
Next Article
Cite
Download Cover
Published Online: Jul 21, 2024
Page range: 89 - 99
DOI: https://doi.org/10.2478/ijanmc-2024-0019
Keywords
© 2024 Leixiang Xia et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Figure 1.

Structure of CycleGAN
Structure of CycleGAN

Figure 2.

Schematic diagram of SENet
Schematic diagram of SENet

Figure 3.

Diagram of the SE-Resnet module
Diagram of the SE-Resnet module

Figure 4.

Schematic diagram of the attention mechanism for joining channels
Schematic diagram of the attention mechanism for joining channels

Figure 5.

Improving the Generator Network Structure
Improving the Generator Network Structure

Figure 6.

Comparison of Results before and after Adding Color Preservation Loss
Comparison of Results before and after Adding Color Preservation Loss

Figure 7.

Loss Curve Variation Chart
Loss Curve Variation Chart

Figure 8.

Loss Function Change Trend for the Original CycleGAN Model
Loss Function Change Trend for the Original CycleGAN Model

Figure 9.

Loss Function Change Trend for the Improved CycleGAN Model
Loss Function Change Trend for the Improved CycleGAN Model

Figure 10.

Comparison of images generated by different models
Comparison of images generated by different models

Comparative experimental evaluation score table

Generative model SSIM PSNR
GAN 0.19 13.5
DRIT 0.48 14.8
CycleGAN+SENet+improved loss function 0.77 18.9

Evaluation index table of ablation experiments

Model SSIM PSNR
CycleGAN 0.50 15.6
CycleGAN+SENet 0.61 16.7
CycleGAN+improved loss function 0.68 15.9
CycleGAN+SENet+improved loss function. 0.77 18.9
Language:
English
Publication timeframe:
4 times per year
Journal Subjects:
Computer Sciences, Computer Sciences, other
Journal RSS Feed