Unrolling Alternating Direction Method of Multipliers for Visible and Infrared Image Fusion

Altug Bakan; Isin Erer

doi:10.1109/IPAS55744.2022.10052930

Unrolling Alternating Direction Method of Multipliers for Visible and Infrared Image Fusion

Altug Bakan, Isin Erer

Department of Electronics and Communication Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

In this paper a new infrared and visible image fusion (IVIF) method which combines the advantages of optimization and deep learning based methods is proposed. This model takes the iterative solution used by the alternating direction method of the multiplier (ADMM) optimization method, and uses algorithm unrolling to obtain a high performance and efficient algorithm. Compared with traditional optimization methods, this model generates fusion with 99.6% improvement in terms of image fusion time, and compared with deep learning based algorithms, this model generates detailed fusion images with 99.1% improvement in terms of training time. Compared with the other state-of-the-art unrolling based methods, this model performs 26.7% better on average in terms of Average Gradient (AG), Cross Entropy (CE), Mutual Information (MI), Peak Signal-to-Noise Ratio (PSNR), and Structural Similarity Loss (SSIM) metrics with a minimal testing time cost.

Original language	English
Title of host publication	5th IEEE International Image Processing, Applications and Systems Conference, IPAS 2022
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781665462198
DOIs	https://doi.org/10.1109/IPAS55744.2022.10052930
Publication status	Published - 2022
Event	5th IEEE International Image Processing, Applications and Systems Conference, IPAS 2022 - Genova, Italy Duration: 5 Dec 2022 → 7 Dec 2022

Publication series

Name	5th IEEE International Image Processing, Applications and Systems Conference, IPAS 2022

Conference

Conference	5th IEEE International Image Processing, Applications and Systems Conference, IPAS 2022
Country/Territory	Italy
City	Genova
Period	5/12/22 → 7/12/22

Bibliographical note

Publisher Copyright:
© 2022 IEEE.

Keywords

ADMM
algorithm unrolling
model based deep network
visible and infrared image fusion

Access to Document

10.1109/IPAS55744.2022.10052930

Cite this

Bakan, A., & Erer, I. (2022). Unrolling Alternating Direction Method of Multipliers for Visible and Infrared Image Fusion. In 5th IEEE International Image Processing, Applications and Systems Conference, IPAS 2022 (5th IEEE International Image Processing, Applications and Systems Conference, IPAS 2022). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IPAS55744.2022.10052930

@inproceedings{0e3b5e14260345c78d1daa940d66bca5,

title = "Unrolling Alternating Direction Method of Multipliers for Visible and Infrared Image Fusion",

abstract = "In this paper a new infrared and visible image fusion (IVIF) method which combines the advantages of optimization and deep learning based methods is proposed. This model takes the iterative solution used by the alternating direction method of the multiplier (ADMM) optimization method, and uses algorithm unrolling to obtain a high performance and efficient algorithm. Compared with traditional optimization methods, this model generates fusion with 99.6% improvement in terms of image fusion time, and compared with deep learning based algorithms, this model generates detailed fusion images with 99.1% improvement in terms of training time. Compared with the other state-of-the-art unrolling based methods, this model performs 26.7% better on average in terms of Average Gradient (AG), Cross Entropy (CE), Mutual Information (MI), Peak Signal-to-Noise Ratio (PSNR), and Structural Similarity Loss (SSIM) metrics with a minimal testing time cost.",

keywords = "ADMM, algorithm unrolling, model based deep network, visible and infrared image fusion",

author = "Altug Bakan and Isin Erer",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 5th IEEE International Image Processing, Applications and Systems Conference, IPAS 2022 ; Conference date: 05-12-2022 Through 07-12-2022",

year = "2022",

doi = "10.1109/IPAS55744.2022.10052930",

language = "English",

series = "5th IEEE International Image Processing, Applications and Systems Conference, IPAS 2022",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "5th IEEE International Image Processing, Applications and Systems Conference, IPAS 2022",

address = "United States",

}

Bakan, A & Erer, I 2022, Unrolling Alternating Direction Method of Multipliers for Visible and Infrared Image Fusion. in 5th IEEE International Image Processing, Applications and Systems Conference, IPAS 2022. 5th IEEE International Image Processing, Applications and Systems Conference, IPAS 2022, Institute of Electrical and Electronics Engineers Inc., 5th IEEE International Image Processing, Applications and Systems Conference, IPAS 2022, Genova, Italy, 5/12/22. https://doi.org/10.1109/IPAS55744.2022.10052930

Unrolling Alternating Direction Method of Multipliers for Visible and Infrared Image Fusion. / Bakan, Altug; Erer, Isin.
5th IEEE International Image Processing, Applications and Systems Conference, IPAS 2022. Institute of Electrical and Electronics Engineers Inc., 2022. (5th IEEE International Image Processing, Applications and Systems Conference, IPAS 2022).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Unrolling Alternating Direction Method of Multipliers for Visible and Infrared Image Fusion

AU - Bakan, Altug

AU - Erer, Isin

PY - 2022

Y1 - 2022

N2 - In this paper a new infrared and visible image fusion (IVIF) method which combines the advantages of optimization and deep learning based methods is proposed. This model takes the iterative solution used by the alternating direction method of the multiplier (ADMM) optimization method, and uses algorithm unrolling to obtain a high performance and efficient algorithm. Compared with traditional optimization methods, this model generates fusion with 99.6% improvement in terms of image fusion time, and compared with deep learning based algorithms, this model generates detailed fusion images with 99.1% improvement in terms of training time. Compared with the other state-of-the-art unrolling based methods, this model performs 26.7% better on average in terms of Average Gradient (AG), Cross Entropy (CE), Mutual Information (MI), Peak Signal-to-Noise Ratio (PSNR), and Structural Similarity Loss (SSIM) metrics with a minimal testing time cost.

AB - In this paper a new infrared and visible image fusion (IVIF) method which combines the advantages of optimization and deep learning based methods is proposed. This model takes the iterative solution used by the alternating direction method of the multiplier (ADMM) optimization method, and uses algorithm unrolling to obtain a high performance and efficient algorithm. Compared with traditional optimization methods, this model generates fusion with 99.6% improvement in terms of image fusion time, and compared with deep learning based algorithms, this model generates detailed fusion images with 99.1% improvement in terms of training time. Compared with the other state-of-the-art unrolling based methods, this model performs 26.7% better on average in terms of Average Gradient (AG), Cross Entropy (CE), Mutual Information (MI), Peak Signal-to-Noise Ratio (PSNR), and Structural Similarity Loss (SSIM) metrics with a minimal testing time cost.

KW - ADMM

KW - algorithm unrolling

KW - model based deep network

KW - visible and infrared image fusion

UR - http://www.scopus.com/inward/record.url?scp=85149762903&partnerID=8YFLogxK

U2 - 10.1109/IPAS55744.2022.10052930

DO - 10.1109/IPAS55744.2022.10052930

M3 - Conference contribution

AN - SCOPUS:85149762903

T3 - 5th IEEE International Image Processing, Applications and Systems Conference, IPAS 2022

BT - 5th IEEE International Image Processing, Applications and Systems Conference, IPAS 2022

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 5th IEEE International Image Processing, Applications and Systems Conference, IPAS 2022

Y2 - 5 December 2022 through 7 December 2022

ER -

Bakan A, Erer I. Unrolling Alternating Direction Method of Multipliers for Visible and Infrared Image Fusion. In 5th IEEE International Image Processing, Applications and Systems Conference, IPAS 2022. Institute of Electrical and Electronics Engineers Inc. 2022. (5th IEEE International Image Processing, Applications and Systems Conference, IPAS 2022). doi: 10.1109/IPAS55744.2022.10052930

Unrolling Alternating Direction Method of Multipliers for Visible and Infrared Image Fusion

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this