A Novel Double Ensemble Algorithm for the Classification of Multi-Class Imbalanced Hyperspectral Data
The class imbalance problem has been reported to exist in remote sensing and hinders the classification performance of many machine learning algorithms. Several technologies, such as data sampling methods, feature selection-based methods, and ensemble-based methods, have been proposed to solve the c...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-08-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/15/3765 |
| _version_ | 1797432733716185088 |
|---|---|
| author | Daying Quan Wei Feng Gabriel Dauphin Xiaofeng Wang Wenjiang Huang Mengdao Xing |
| author_facet | Daying Quan Wei Feng Gabriel Dauphin Xiaofeng Wang Wenjiang Huang Mengdao Xing |
| author_sort | Daying Quan |
| collection | DOAJ |
| description | The class imbalance problem has been reported to exist in remote sensing and hinders the classification performance of many machine learning algorithms. Several technologies, such as data sampling methods, feature selection-based methods, and ensemble-based methods, have been proposed to solve the class imbalance problem. However, these methods suffer from the loss of useful information or from artificial noise, or result in overfitting. A novel double ensemble algorithm is proposed to deal with the multi-class imbalance problem of the hyperspectral image in this paper. This method first computes the feature importance values of the hyperspectral data via an ensemble model, then produces several balanced data sets based on oversampling and builds a number of classifiers. Finally, the classification results of these diversity classifiers are combined according to a specific ensemble rule. In the experiment, different data-handling methods and classification methods including random undersampling (RUS), random oversampling (ROS), Adaboost, Bagging, and random forest are compared with the proposed double random forest method. The experimental results on three imbalanced hyperspectral data sets demonstrate the effectiveness of the proposed algorithm. |
| first_indexed | 2024-03-09T10:05:54Z |
| format | Article |
| id | doaj.art-dd7c0752a057404992c4ccbcb9a0585b |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T10:05:54Z |
| publishDate | 2022-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-dd7c0752a057404992c4ccbcb9a0585b2023-12-01T23:08:42ZengMDPI AGRemote Sensing2072-42922022-08-011415376510.3390/rs14153765A Novel Double Ensemble Algorithm for the Classification of Multi-Class Imbalanced Hyperspectral DataDaying Quan0Wei Feng1Gabriel Dauphin2Xiaofeng Wang3Wenjiang Huang4Mengdao Xing5Key Laboratory of Electromagnetic Wave Information Technology and Metrology of Zhejiang Province, College of Information Engineering, China Jiliang University, Hangzhou 310018, ChinaHangzhou Institute of Technology, Xidian University, Hangzhou 311200, ChinaLaboratory of Information Processing and Transmission, L2TI, Institut Galilée, University Paris XIII, 93430 Villetaneuse, FranceKey Laboratory of Electromagnetic Wave Information Technology and Metrology of Zhejiang Province, College of Information Engineering, China Jiliang University, Hangzhou 310018, ChinaKey Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaHangzhou Institute of Technology, Xidian University, Hangzhou 311200, ChinaThe class imbalance problem has been reported to exist in remote sensing and hinders the classification performance of many machine learning algorithms. Several technologies, such as data sampling methods, feature selection-based methods, and ensemble-based methods, have been proposed to solve the class imbalance problem. However, these methods suffer from the loss of useful information or from artificial noise, or result in overfitting. A novel double ensemble algorithm is proposed to deal with the multi-class imbalance problem of the hyperspectral image in this paper. This method first computes the feature importance values of the hyperspectral data via an ensemble model, then produces several balanced data sets based on oversampling and builds a number of classifiers. Finally, the classification results of these diversity classifiers are combined according to a specific ensemble rule. In the experiment, different data-handling methods and classification methods including random undersampling (RUS), random oversampling (ROS), Adaboost, Bagging, and random forest are compared with the proposed double random forest method. The experimental results on three imbalanced hyperspectral data sets demonstrate the effectiveness of the proposed algorithm.https://www.mdpi.com/2072-4292/14/15/3765classificationremote sensinghyperspectral imageimbalance learningdata sampling |
| spellingShingle | Daying Quan Wei Feng Gabriel Dauphin Xiaofeng Wang Wenjiang Huang Mengdao Xing A Novel Double Ensemble Algorithm for the Classification of Multi-Class Imbalanced Hyperspectral Data Remote Sensing classification remote sensing hyperspectral image imbalance learning data sampling |
| title | A Novel Double Ensemble Algorithm for the Classification of Multi-Class Imbalanced Hyperspectral Data |
| title_full | A Novel Double Ensemble Algorithm for the Classification of Multi-Class Imbalanced Hyperspectral Data |
| title_fullStr | A Novel Double Ensemble Algorithm for the Classification of Multi-Class Imbalanced Hyperspectral Data |
| title_full_unstemmed | A Novel Double Ensemble Algorithm for the Classification of Multi-Class Imbalanced Hyperspectral Data |
| title_short | A Novel Double Ensemble Algorithm for the Classification of Multi-Class Imbalanced Hyperspectral Data |
| title_sort | novel double ensemble algorithm for the classification of multi class imbalanced hyperspectral data |
| topic | classification remote sensing hyperspectral image imbalance learning data sampling |
| url | https://www.mdpi.com/2072-4292/14/15/3765 |
| work_keys_str_mv | AT dayingquan anoveldoubleensemblealgorithmfortheclassificationofmulticlassimbalancedhyperspectraldata AT weifeng anoveldoubleensemblealgorithmfortheclassificationofmulticlassimbalancedhyperspectraldata AT gabrieldauphin anoveldoubleensemblealgorithmfortheclassificationofmulticlassimbalancedhyperspectraldata AT xiaofengwang anoveldoubleensemblealgorithmfortheclassificationofmulticlassimbalancedhyperspectraldata AT wenjianghuang anoveldoubleensemblealgorithmfortheclassificationofmulticlassimbalancedhyperspectraldata AT mengdaoxing anoveldoubleensemblealgorithmfortheclassificationofmulticlassimbalancedhyperspectraldata AT dayingquan noveldoubleensemblealgorithmfortheclassificationofmulticlassimbalancedhyperspectraldata AT weifeng noveldoubleensemblealgorithmfortheclassificationofmulticlassimbalancedhyperspectraldata AT gabrieldauphin noveldoubleensemblealgorithmfortheclassificationofmulticlassimbalancedhyperspectraldata AT xiaofengwang noveldoubleensemblealgorithmfortheclassificationofmulticlassimbalancedhyperspectraldata AT wenjianghuang noveldoubleensemblealgorithmfortheclassificationofmulticlassimbalancedhyperspectraldata AT mengdaoxing noveldoubleensemblealgorithmfortheclassificationofmulticlassimbalancedhyperspectraldata |