Object-based multispectral image fusion method using deep learning

Hyunsung Jang; Namkoo Ha; Yoonmo Yeon; Kuyong Kwon; Sungho Gil; Seungha Lee; Sungsoon Park; Hyungjoo Jung; Kwanghoon Sohn

doi:10.1117/12.2532718

Object-based multispectral image fusion method using deep learning

Hyunsung Jang, Namkoo Ha, Yoonmo Yeon, Kuyong Kwon, Sungho Gil, Seungha Lee, Sungsoon Park, Hyungjoo Jung, Kwanghoon Sohn

Department of Electrical and Electronic Engineering

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

Abstract

The goal of multispectral image fusion is to integrate complementary information from multispectral sensors to enhance human visual perception and object detection. Additionally, there are also cases when only the object needs to be emphasized with minimal background interference. This paper presents an object-based fusion method using deep learning to accomplish this objective. The proposed method uses information regarding the region of an object to perform fusion on the object. As we cannot provide labels for fusion results at the learning stage, we propose an unsupervised learning method. The proposed method simultaneously provides appropriate image information from the background and target for surveillance and reconnaissance.

Original language	English
Title of host publication	Artificial Intelligence and Machine Learning in Defense Applications
Editors	Judith Dijk
Publisher	SPIE
ISBN (Electronic)	9781510630413
DOIs	https://doi.org/10.1117/12.2532718
Publication status	Published - 2019
Event	Artificial Intelligence and Machine Learning in Defense Applications 2019 - Strasbourg, France Duration: 2019 Sept 10 → 2019 Sept 12

Publication series

Name	Proceedings of SPIE - The International Society for Optical Engineering
Volume	11169
ISSN (Print)	0277-786X
ISSN (Electronic)	1996-756X

Conference

Conference	Artificial Intelligence and Machine Learning in Defense Applications 2019
Country/Territory	France
City	Strasbourg
Period	19/9/10 → 19/9/12

Bibliographical note

Funding Information:
This research was supported by the Target Detection and Tracking using the Deep Learning RZ04CM-001) and the SWIR/LWIR Image Fusion project (No. Y17-030) of LIG Nex1 Co., Ltd.

Publisher Copyright:
© 2019 SPIE.

All Science Journal Classification (ASJC) codes

Electronic, Optical and Magnetic Materials
Condensed Matter Physics
Computer Science Applications
Applied Mathematics
Electrical and Electronic Engineering

Access to Document

10.1117/12.2532718

Cite this

Jang, H., Ha, N., Yeon, Y., Kwon, K., Gil, S., Lee, S., Park, S., Jung, H., & Sohn, K. (2019). Object-based multispectral image fusion method using deep learning. In J. Dijk (Ed.), Artificial Intelligence and Machine Learning in Defense Applications [111690O] (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 11169). SPIE. https://doi.org/10.1117/12.2532718

@inproceedings{f1859811399c489f94d2aceeb35a3f94,

title = "Object-based multispectral image fusion method using deep learning",

abstract = "The goal of multispectral image fusion is to integrate complementary information from multispectral sensors to enhance human visual perception and object detection. Additionally, there are also cases when only the object needs to be emphasized with minimal background interference. This paper presents an object-based fusion method using deep learning to accomplish this objective. The proposed method uses information regarding the region of an object to perform fusion on the object. As we cannot provide labels for fusion results at the learning stage, we propose an unsupervised learning method. The proposed method simultaneously provides appropriate image information from the background and target for surveillance and reconnaissance.",

author = "Hyunsung Jang and Namkoo Ha and Yoonmo Yeon and Kuyong Kwon and Sungho Gil and Seungha Lee and Sungsoon Park and Hyungjoo Jung and Kwanghoon Sohn",

note = "Funding Information: This research was supported by the Target Detection and Tracking using the Deep Learning RZ04CM-001) and the SWIR/LWIR Image Fusion project (No. Y17-030) of LIG Nex1 Co., Ltd. Publisher Copyright: {\textcopyright} 2019 SPIE.; Artificial Intelligence and Machine Learning in Defense Applications 2019 ; Conference date: 10-09-2019 Through 12-09-2019",

year = "2019",

doi = "10.1117/12.2532718",

language = "English",

series = "Proceedings of SPIE - The International Society for Optical Engineering",

publisher = "SPIE",

editor = "Judith Dijk",

booktitle = "Artificial Intelligence and Machine Learning in Defense Applications",

address = "United States",

}

Jang, H, Ha, N, Yeon, Y, Kwon, K, Gil, S, Lee, S, Park, S, Jung, H & Sohn, K 2019, Object-based multispectral image fusion method using deep learning. in J Dijk (ed.), Artificial Intelligence and Machine Learning in Defense Applications., 111690O, Proceedings of SPIE - The International Society for Optical Engineering, vol. 11169, SPIE, Artificial Intelligence and Machine Learning in Defense Applications 2019, Strasbourg, France, 19/9/10. https://doi.org/10.1117/12.2532718

Object-based multispectral image fusion method using deep learning. / Jang, Hyunsung; Ha, Namkoo; Yeon, Yoonmo et al.
Artificial Intelligence and Machine Learning in Defense Applications. ed. / Judith Dijk. SPIE, 2019. 111690O (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 11169).

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

TY - GEN

T1 - Object-based multispectral image fusion method using deep learning

AU - Jang, Hyunsung

AU - Ha, Namkoo

AU - Yeon, Yoonmo

AU - Kwon, Kuyong

AU - Gil, Sungho

AU - Lee, Seungha

AU - Park, Sungsoon

AU - Jung, Hyungjoo

AU - Sohn, Kwanghoon

N1 - Funding Information: This research was supported by the Target Detection and Tracking using the Deep Learning RZ04CM-001) and the SWIR/LWIR Image Fusion project (No. Y17-030) of LIG Nex1 Co., Ltd. Publisher Copyright: © 2019 SPIE.

PY - 2019

Y1 - 2019

N2 - The goal of multispectral image fusion is to integrate complementary information from multispectral sensors to enhance human visual perception and object detection. Additionally, there are also cases when only the object needs to be emphasized with minimal background interference. This paper presents an object-based fusion method using deep learning to accomplish this objective. The proposed method uses information regarding the region of an object to perform fusion on the object. As we cannot provide labels for fusion results at the learning stage, we propose an unsupervised learning method. The proposed method simultaneously provides appropriate image information from the background and target for surveillance and reconnaissance.

AB - The goal of multispectral image fusion is to integrate complementary information from multispectral sensors to enhance human visual perception and object detection. Additionally, there are also cases when only the object needs to be emphasized with minimal background interference. This paper presents an object-based fusion method using deep learning to accomplish this objective. The proposed method uses information regarding the region of an object to perform fusion on the object. As we cannot provide labels for fusion results at the learning stage, we propose an unsupervised learning method. The proposed method simultaneously provides appropriate image information from the background and target for surveillance and reconnaissance.

UR - http://www.scopus.com/inward/record.url?scp=85077321228&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85077321228&partnerID=8YFLogxK

U2 - 10.1117/12.2532718

DO - 10.1117/12.2532718

M3 - Conference contribution

AN - SCOPUS:85077321228

T3 - Proceedings of SPIE - The International Society for Optical Engineering

BT - Artificial Intelligence and Machine Learning in Defense Applications

A2 - Dijk, Judith

PB - SPIE

T2 - Artificial Intelligence and Machine Learning in Defense Applications 2019

Y2 - 10 September 2019 through 12 September 2019

ER -

Object-based multispectral image fusion method using deep learning

Abstract

Publication series

Conference

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this