Heterogeneous Image Fusion for Target Recognition Based on Evidence Reasoning

Shuyue Wang; Zhunga Liu; Zuowei Zhang; Yang Li

doi:10.1007/978-3-031-17801-6_15

Heterogeneous Image Fusion for Target Recognition Based on Evidence Reasoning

Shuyue Wang, Zhunga Liu, Zuowei Zhang, Yang Li

School of Automation

Northwestern Polytechnical University Xian

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

1 Scopus citations

Abstract

Multi-source fusion is an efficient strategy in complex image target recognition since it can exploit the complementary knowledge in different sources to improve the classification performance. In this paper, we propose a new end-to-end framework for heterogeneous (i.e. visible & infrared) image fusion target recognition (HIFTR). Firstly, two networks are built for the visible and infrared images respectively and jointly trained based on mutual learning. It aims to transfer heterogeneous information mutually and improve the generalization performance of the networks. Secondly, a weighted decision-level fusion method based on evidence reasoning is developed to combine the classification results of visible and infrared images for the final target recognition. In the training process, the weight of each image is automatically optimized in the networks. Finally, the performance of the proposed HIFTR has been evaluated by comparing with other related methods, and the experimental results show that the HIFTR method can efficiently improve the classification accuracy.

Original language	English
Title of host publication	Belief Functions
Subtitle of host publication	Theory and Applications - 7th International Conference, BELIEF 2022, Proceedings
Editors	Sylvie Le Hégarat-Mascle, Emanuel Aldea, Isabelle Bloch
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	153-162
Number of pages	10
ISBN (Print)	9783031178009
DOIs	https://doi.org/10.1007/978-3-031-17801-6_15
State	Published - 2022
Event	7th International Conference on Belief Functions, BELIEF 2022 - Paris, France Duration: 26 Oct 2022 → 28 Oct 2022

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	13506 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	7th International Conference on Belief Functions, BELIEF 2022
Country/Territory	France
City	Paris
Period	26/10/22 → 28/10/22

Keywords

Evidence reasoning
Heterogeneous image fusion
Mutual learning
Target recognition

Access to Document

10.1007/978-3-031-17801-6_15

Cite this

Wang, S., Liu, Z., Zhang, Z., & Li, Y. (2022). Heterogeneous Image Fusion for Target Recognition Based on Evidence Reasoning. In S. Le Hégarat-Mascle, E. Aldea, & I. Bloch (Eds.), Belief Functions: Theory and Applications - 7th International Conference, BELIEF 2022, Proceedings (pp. 153-162). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13506 LNAI). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-17801-6_15

Wang, Shuyue ; Liu, Zhunga ; Zhang, Zuowei et al. / Heterogeneous Image Fusion for Target Recognition Based on Evidence Reasoning. Belief Functions: Theory and Applications - 7th International Conference, BELIEF 2022, Proceedings. editor / Sylvie Le Hégarat-Mascle ; Emanuel Aldea ; Isabelle Bloch. Springer Science and Business Media Deutschland GmbH, 2022. pp. 153-162 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{a38b7cb3cb694cff9257094defc93bb6,

title = "Heterogeneous Image Fusion for Target Recognition Based on Evidence Reasoning",

abstract = "Multi-source fusion is an efficient strategy in complex image target recognition since it can exploit the complementary knowledge in different sources to improve the classification performance. In this paper, we propose a new end-to-end framework for heterogeneous (i.e. visible & infrared) image fusion target recognition (HIFTR). Firstly, two networks are built for the visible and infrared images respectively and jointly trained based on mutual learning. It aims to transfer heterogeneous information mutually and improve the generalization performance of the networks. Secondly, a weighted decision-level fusion method based on evidence reasoning is developed to combine the classification results of visible and infrared images for the final target recognition. In the training process, the weight of each image is automatically optimized in the networks. Finally, the performance of the proposed HIFTR has been evaluated by comparing with other related methods, and the experimental results show that the HIFTR method can efficiently improve the classification accuracy.",

keywords = "Evidence reasoning, Heterogeneous image fusion, Mutual learning, Target recognition",

author = "Shuyue Wang and Zhunga Liu and Zuowei Zhang and Yang Li",

note = "Publisher Copyright: {\textcopyright} 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.; 7th International Conference on Belief Functions, BELIEF 2022 ; Conference date: 26-10-2022 Through 28-10-2022",

year = "2022",

doi = "10.1007/978-3-031-17801-6_15",

language = "英语",

isbn = "9783031178009",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "153--162",

editor = "{Le H{\'e}garat-Mascle}, Sylvie and Emanuel Aldea and Isabelle Bloch",

booktitle = "Belief Functions",

}

Wang, S, Liu, Z, Zhang, Z & Li, Y 2022, Heterogeneous Image Fusion for Target Recognition Based on Evidence Reasoning. in S Le Hégarat-Mascle, E Aldea & I Bloch (eds), Belief Functions: Theory and Applications - 7th International Conference, BELIEF 2022, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 13506 LNAI, Springer Science and Business Media Deutschland GmbH, pp. 153-162, 7th International Conference on Belief Functions, BELIEF 2022, Paris, France, 26/10/22. https://doi.org/10.1007/978-3-031-17801-6_15

Heterogeneous Image Fusion for Target Recognition Based on Evidence Reasoning. / Wang, Shuyue; Liu, Zhunga; Zhang, Zuowei et al.
Belief Functions: Theory and Applications - 7th International Conference, BELIEF 2022, Proceedings. ed. / Sylvie Le Hégarat-Mascle; Emanuel Aldea; Isabelle Bloch. Springer Science and Business Media Deutschland GmbH, 2022. p. 153-162 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13506 LNAI).

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

TY - GEN

T1 - Heterogeneous Image Fusion for Target Recognition Based on Evidence Reasoning

AU - Wang, Shuyue

AU - Liu, Zhunga

AU - Zhang, Zuowei

AU - Li, Yang

PY - 2022

Y1 - 2022

N2 - Multi-source fusion is an efficient strategy in complex image target recognition since it can exploit the complementary knowledge in different sources to improve the classification performance. In this paper, we propose a new end-to-end framework for heterogeneous (i.e. visible & infrared) image fusion target recognition (HIFTR). Firstly, two networks are built for the visible and infrared images respectively and jointly trained based on mutual learning. It aims to transfer heterogeneous information mutually and improve the generalization performance of the networks. Secondly, a weighted decision-level fusion method based on evidence reasoning is developed to combine the classification results of visible and infrared images for the final target recognition. In the training process, the weight of each image is automatically optimized in the networks. Finally, the performance of the proposed HIFTR has been evaluated by comparing with other related methods, and the experimental results show that the HIFTR method can efficiently improve the classification accuracy.

AB - Multi-source fusion is an efficient strategy in complex image target recognition since it can exploit the complementary knowledge in different sources to improve the classification performance. In this paper, we propose a new end-to-end framework for heterogeneous (i.e. visible & infrared) image fusion target recognition (HIFTR). Firstly, two networks are built for the visible and infrared images respectively and jointly trained based on mutual learning. It aims to transfer heterogeneous information mutually and improve the generalization performance of the networks. Secondly, a weighted decision-level fusion method based on evidence reasoning is developed to combine the classification results of visible and infrared images for the final target recognition. In the training process, the weight of each image is automatically optimized in the networks. Finally, the performance of the proposed HIFTR has been evaluated by comparing with other related methods, and the experimental results show that the HIFTR method can efficiently improve the classification accuracy.

KW - Evidence reasoning

KW - Heterogeneous image fusion

KW - Mutual learning

KW - Target recognition

UR - http://www.scopus.com/inward/record.url?scp=85140454997&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-17801-6_15

DO - 10.1007/978-3-031-17801-6_15

M3 - 会议稿件

AN - SCOPUS:85140454997

SN - 9783031178009

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 153

EP - 162

BT - Belief Functions

A2 - Le Hégarat-Mascle, Sylvie

A2 - Aldea, Emanuel

A2 - Bloch, Isabelle

PB - Springer Science and Business Media Deutschland GmbH

T2 - 7th International Conference on Belief Functions, BELIEF 2022

Y2 - 26 October 2022 through 28 October 2022

ER -

Wang S, Liu Z, Zhang Z, Li Y. Heterogeneous Image Fusion for Target Recognition Based on Evidence Reasoning. In Le Hégarat-Mascle S, Aldea E, Bloch I, editors, Belief Functions: Theory and Applications - 7th International Conference, BELIEF 2022, Proceedings. Springer Science and Business Media Deutschland GmbH. 2022. p. 153-162. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-031-17801-6_15