Agreement and Disagreement-Based Co-Learning with Dual Network for Hyperspectral Image Classification with Noisy Labels
Deep learning-based label noise learning methods provide promising solutions for hyperspectral image (HSI) classification with noisy labels. Currently, label noise learning methods based on deep learning improve their performance by modifying one aspect, such as designing a robust loss function, rev...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-05-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/10/2543 |
| _version_ | 1797598492978315264 |
|---|---|
| author | Youqiang Zhang Jin Sun Hao Shi Zixian Ge Qiqiong Yu Guo Cao Xuesong Li |
| author_facet | Youqiang Zhang Jin Sun Hao Shi Zixian Ge Qiqiong Yu Guo Cao Xuesong Li |
| author_sort | Youqiang Zhang |
| collection | DOAJ |
| description | Deep learning-based label noise learning methods provide promising solutions for hyperspectral image (HSI) classification with noisy labels. Currently, label noise learning methods based on deep learning improve their performance by modifying one aspect, such as designing a robust loss function, revamping the network structure, or adding a noise adaptation layer. However, these methods face difficulties in coping with relatively high noise situations. To address this issue, this paper proposes a unified label noise learning framework with a dual-network structure. The goal is to enhance the model’s robustness to label noise by utilizing two networks to guide each other. Specifically, to avoid the degeneration of the dual-network training into self-training, the “disagreement” strategy is incorporated with co-learning. Then, the “agreement” strategy is introduced into the model to ensure that the model iterates in the right direction under high noise conditions. To this end, an agreement and disagreement-based co-learning (ADCL) framework is proposed for HSI classification with noisy labels. In addition, a joint loss function consisting of a supervision loss of two networks and a relative loss between two networks is designed for the dual-network structure. Extensive experiments are conducted on three public HSI datasets to demonstrate the robustness of the proposed method to label noise. Specifically, our method obtains the highest overall accuracy of 98.62%, 90.89%, and 99.02% on the three datasets, respectively, which represents an improvement of 2.58%, 2.27%, and 0.86% compared to the second-best method. In future research, the authors suggest using more networks as backbones to implement the ADCL framework. |
| first_indexed | 2024-03-11T03:21:52Z |
| format | Article |
| id | doaj.art-e9ad721d71664fb38214dad1c0c4a8a6 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-11T03:21:52Z |
| publishDate | 2023-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-e9ad721d71664fb38214dad1c0c4a8a62023-11-18T03:06:37ZengMDPI AGRemote Sensing2072-42922023-05-011510254310.3390/rs15102543Agreement and Disagreement-Based Co-Learning with Dual Network for Hyperspectral Image Classification with Noisy LabelsYouqiang Zhang0Jin Sun1Hao Shi2Zixian Ge3Qiqiong Yu4Guo Cao5Xuesong Li6School of Internet of Things, Nanjing University of Posts and Telecommunications, Nanjing 210003, ChinaSchool of Internet of Things, Nanjing University of Posts and Telecommunications, Nanjing 210003, ChinaSchool of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing 210094, ChinaSchool of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing 210094, ChinaSchool of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing 210094, ChinaSchool of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing 210094, ChinaSchool of Computer Science and Technology, Tiangong University, Tianjin 300387, ChinaDeep learning-based label noise learning methods provide promising solutions for hyperspectral image (HSI) classification with noisy labels. Currently, label noise learning methods based on deep learning improve their performance by modifying one aspect, such as designing a robust loss function, revamping the network structure, or adding a noise adaptation layer. However, these methods face difficulties in coping with relatively high noise situations. To address this issue, this paper proposes a unified label noise learning framework with a dual-network structure. The goal is to enhance the model’s robustness to label noise by utilizing two networks to guide each other. Specifically, to avoid the degeneration of the dual-network training into self-training, the “disagreement” strategy is incorporated with co-learning. Then, the “agreement” strategy is introduced into the model to ensure that the model iterates in the right direction under high noise conditions. To this end, an agreement and disagreement-based co-learning (ADCL) framework is proposed for HSI classification with noisy labels. In addition, a joint loss function consisting of a supervision loss of two networks and a relative loss between two networks is designed for the dual-network structure. Extensive experiments are conducted on three public HSI datasets to demonstrate the robustness of the proposed method to label noise. Specifically, our method obtains the highest overall accuracy of 98.62%, 90.89%, and 99.02% on the three datasets, respectively, which represents an improvement of 2.58%, 2.27%, and 0.86% compared to the second-best method. In future research, the authors suggest using more networks as backbones to implement the ADCL framework.https://www.mdpi.com/2072-4292/15/10/2543hyperspectral imageco-learninglabel noise learningclassification |
| spellingShingle | Youqiang Zhang Jin Sun Hao Shi Zixian Ge Qiqiong Yu Guo Cao Xuesong Li Agreement and Disagreement-Based Co-Learning with Dual Network for Hyperspectral Image Classification with Noisy Labels Remote Sensing hyperspectral image co-learning label noise learning classification |
| title | Agreement and Disagreement-Based Co-Learning with Dual Network for Hyperspectral Image Classification with Noisy Labels |
| title_full | Agreement and Disagreement-Based Co-Learning with Dual Network for Hyperspectral Image Classification with Noisy Labels |
| title_fullStr | Agreement and Disagreement-Based Co-Learning with Dual Network for Hyperspectral Image Classification with Noisy Labels |
| title_full_unstemmed | Agreement and Disagreement-Based Co-Learning with Dual Network for Hyperspectral Image Classification with Noisy Labels |
| title_short | Agreement and Disagreement-Based Co-Learning with Dual Network for Hyperspectral Image Classification with Noisy Labels |
| title_sort | agreement and disagreement based co learning with dual network for hyperspectral image classification with noisy labels |
| topic | hyperspectral image co-learning label noise learning classification |
| url | https://www.mdpi.com/2072-4292/15/10/2543 |
| work_keys_str_mv | AT youqiangzhang agreementanddisagreementbasedcolearningwithdualnetworkforhyperspectralimageclassificationwithnoisylabels AT jinsun agreementanddisagreementbasedcolearningwithdualnetworkforhyperspectralimageclassificationwithnoisylabels AT haoshi agreementanddisagreementbasedcolearningwithdualnetworkforhyperspectralimageclassificationwithnoisylabels AT zixiange agreementanddisagreementbasedcolearningwithdualnetworkforhyperspectralimageclassificationwithnoisylabels AT qiqiongyu agreementanddisagreementbasedcolearningwithdualnetworkforhyperspectralimageclassificationwithnoisylabels AT guocao agreementanddisagreementbasedcolearningwithdualnetworkforhyperspectralimageclassificationwithnoisylabels AT xuesongli agreementanddisagreementbasedcolearningwithdualnetworkforhyperspectralimageclassificationwithnoisylabels |