Related Experiment Video
Updated: May 5, 2026
11:15
A Guide to Structured Illumination TIRF Microscopy at High Speed with Multiple Colors
Published on: May 30, 2016
26.9K
Research on Colorization Algorithm for γ-Photon Flow Field Images Using the SECN Model.
Hui Xiao1,2, Liying Hou1,2, Jiantang Liu1
1College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, 29 Jiangjun Road, Nanjing 211106, China.
Entropy (Basel, Switzerland)
|May 4, 2026
Summary
This study introduces a Structure Enhancement Colorization Network (SECN) to improve grayscale γ-photon tomography images for industrial flow monitoring. The SECN model effectively reduces color banding and enhances structural consistency, outperforming existing methods.
Area of Science:
- Industrial imaging
- Non-contact measurement techniques
- Image processing
Background:
- γ-photon tomography is a non-contact method for industrial flow monitoring.
- Grayscale γ-photon images have statistical characteristics causing color banding with standard algorithms.
- This compromises structural continuity and visual consistency in flow field analysis.
Purpose of the Study:
- To develop an advanced colorization model for γ-photon flow-field images.
- To address and mitigate color banding artifacts inherent in these images.
- To enhance structural continuity and visual consistency for improved flow monitoring.
Main Methods:
- Proposed a Structure Enhancement Colorization Network (SECN) using a U-Net + GAN framework with ResNet101.
- Integrated structure-aware enhancement and multi-scale attention modules.
- Enhanced the discriminator with improved boundary and texture discrimination blocks.
Main Results:
- SECN achieved superior image quality metrics (PSNR, SSIM, FID, MAE) compared to DeOldify.
- Demonstrated a 4.44% increase in information entropy, indicating better preservation of complex structures.
- Achieved a significant 18.42% reduction in temperature inversion MAPE for parameter inversion.
Conclusions:
- The SECN model effectively suppresses color banding artifacts in γ-photon flow-field images.
- SECN enhances structural consistency and preserves crucial information for accurate flow monitoring.
- The proposed method offers significant improvements for industrial applications relying on γ-photon tomography.
