A Composite Recognition Method Based on Multimode Mutual Attention Fusion Network
To address the problem of single-mode vulnerability to complex environments, a multimode fusion network with mutual attention is proposed. This network combines the use of laser, infrared and millimeter wave modalities to leverage the advantages of each mode in different environments, increasing the...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2025-12-01
|
| Series: | Applied Artificial Intelligence |
| Online Access: | https://www.tandfonline.com/doi/10.1080/08839514.2025.2462371 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832540881536679936 |
|---|---|
| author | Xing Ding Xiangrong Zhang Chao Liang Bo Liu Lanjie Niu |
| author_facet | Xing Ding Xiangrong Zhang Chao Liang Bo Liu Lanjie Niu |
| author_sort | Xing Ding |
| collection | DOAJ |
| description | To address the problem of single-mode vulnerability to complex environments, a multimode fusion network with mutual attention is proposed. This network combines the use of laser, infrared and millimeter wave modalities to leverage the advantages of each mode in different environments, increasing the network’s resilience to interference. The study begins with the construction of pixel-level fusion networks, feature-weighted fusion networks and the multimode mutual attention fusion network. A comprehensive introduction to the multimode mutual attention fusion network is given, as well as a comparison with the other two networks. The model is then trained and evaluated using data from glide rocket and drone experiments. Finally, an analysis of the anti-outlier interference capability of the multimode fusion network with mutual attention is carried out. The test results show that the multimode mutual attention fusion network containing a feature fusion attention mechanism has the highest detection performance and anti-interference ability. Without interference, the network achieves a remarkable accuracy of 0.98 for multi-target recognition. In addition, with an accuracy of 0.96, it ensures a high level of stability in various interference environments. In addition, the introduction of multi-scale fusion has improved the rocket’s speed adaptability by about 75%. |
| format | Article |
| id | doaj-art-1e971da7562f4c35ac778429c0ff2293 |
| institution | Kabale University |
| issn | 0883-9514 1087-6545 |
| language | English |
| publishDate | 2025-12-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | Applied Artificial Intelligence |
| spelling | doaj-art-1e971da7562f4c35ac778429c0ff22932025-02-04T13:17:23ZengTaylor & Francis GroupApplied Artificial Intelligence0883-95141087-65452025-12-0139110.1080/08839514.2025.2462371A Composite Recognition Method Based on Multimode Mutual Attention Fusion NetworkXing Ding0Xiangrong Zhang1Chao Liang2Bo Liu3Lanjie Niu4School of Artificial Intelligence, Xidian University, Xi’an, ChinaSchool of Artificial Intelligence, Xidian University, Xi’an, ChinaSchool of Artificial Intelligence, Xidian University, Xi’an, ChinaSchool of Artificial Intelligence, Xidian University, Xi’an, ChinaKey Laboratory of Defense Science and Technology for Dynamic Characteristics of Fuze, Xi’an Electromechanical Information Technology Research Institute, Xi’an, ChinaTo address the problem of single-mode vulnerability to complex environments, a multimode fusion network with mutual attention is proposed. This network combines the use of laser, infrared and millimeter wave modalities to leverage the advantages of each mode in different environments, increasing the network’s resilience to interference. The study begins with the construction of pixel-level fusion networks, feature-weighted fusion networks and the multimode mutual attention fusion network. A comprehensive introduction to the multimode mutual attention fusion network is given, as well as a comparison with the other two networks. The model is then trained and evaluated using data from glide rocket and drone experiments. Finally, an analysis of the anti-outlier interference capability of the multimode fusion network with mutual attention is carried out. The test results show that the multimode mutual attention fusion network containing a feature fusion attention mechanism has the highest detection performance and anti-interference ability. Without interference, the network achieves a remarkable accuracy of 0.98 for multi-target recognition. In addition, with an accuracy of 0.96, it ensures a high level of stability in various interference environments. In addition, the introduction of multi-scale fusion has improved the rocket’s speed adaptability by about 75%.https://www.tandfonline.com/doi/10.1080/08839514.2025.2462371 |
| spellingShingle | Xing Ding Xiangrong Zhang Chao Liang Bo Liu Lanjie Niu A Composite Recognition Method Based on Multimode Mutual Attention Fusion Network Applied Artificial Intelligence |
| title | A Composite Recognition Method Based on Multimode Mutual Attention Fusion Network |
| title_full | A Composite Recognition Method Based on Multimode Mutual Attention Fusion Network |
| title_fullStr | A Composite Recognition Method Based on Multimode Mutual Attention Fusion Network |
| title_full_unstemmed | A Composite Recognition Method Based on Multimode Mutual Attention Fusion Network |
| title_short | A Composite Recognition Method Based on Multimode Mutual Attention Fusion Network |
| title_sort | composite recognition method based on multimode mutual attention fusion network |
| url | https://www.tandfonline.com/doi/10.1080/08839514.2025.2462371 |
| work_keys_str_mv | AT xingding acompositerecognitionmethodbasedonmultimodemutualattentionfusionnetwork AT xiangrongzhang acompositerecognitionmethodbasedonmultimodemutualattentionfusionnetwork AT chaoliang acompositerecognitionmethodbasedonmultimodemutualattentionfusionnetwork AT boliu acompositerecognitionmethodbasedonmultimodemutualattentionfusionnetwork AT lanjieniu acompositerecognitionmethodbasedonmultimodemutualattentionfusionnetwork AT xingding compositerecognitionmethodbasedonmultimodemutualattentionfusionnetwork AT xiangrongzhang compositerecognitionmethodbasedonmultimodemutualattentionfusionnetwork AT chaoliang compositerecognitionmethodbasedonmultimodemutualattentionfusionnetwork AT boliu compositerecognitionmethodbasedonmultimodemutualattentionfusionnetwork AT lanjieniu compositerecognitionmethodbasedonmultimodemutualattentionfusionnetwork |