A Multi-Focus Image Fusion Method Based on Brushlet and CNN

DOI

10.2991/978-94-6463-108-1_59How to use a DOI?

Keywords

image fusion; brushlet complex energy; convolutional neural network

Abstract

The fusion of images in the transform domain using convolutional neural networks method can improve the fusion effect, but if the training sample set input to the CNN model is not selected properly, the fused image will show "pseudo-edge", "artificial texture" and other phenomena. In this paper, we propose a CNN image fusion algorithm based on Brushlet energy, which performs non-down sampling contour wave transform on the original image to obtain high and low frequency coefficient maps, uses Brushlet to bilayer decompose the coefficient maps to obtain complex coefficients, obtains the coefficient map chunk energy values by real and imaginary energy solving method, and uses them as the input sample set of CNN model for processing, the CNN model The output is the final decision map for fusion, which can be applied to each high and low frequency coefficient map of NSCT to achieve more accurate image fusion. The experimental results show that the method proposed in this paper has some improvement over other algorithms in both subjective human eye perception effect and quantitative objective evaluation index.

Copyright

© 2022 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Haonan Yu
PY  - 2022
DA  - 2022/12/30
TI  - A Multi-Focus Image Fusion Method Based on Brushlet and CNN
BT  - Proceedings of the 2022 International Conference on Computer Science, Information Engineering and Digital Economy (CSIEDE 2022)
PB  - Atlantis Press
SP  - 514
EP  - 525
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-108-1_59
DO  - 10.2991/978-94-6463-108-1_59
ID  - Yu2022
ER  -