Rotation Invariance Regularization for Remote Sensing Image Scene Classification with Convolutional Neural Networks
Deep convolutional neural networks (DCNNs) have shown significant improvements in remote sensing image scene classification for powerful feature representations. However, because of the high variance and volume limitations of the available remote sensing datasets, DCNNs are prone to overfit the data...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-02-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/13/4/569 |
| _version_ | 1797413985906065408 |
|---|---|
| author | Kunlun Qi Chao Yang Chuli Hu Yonglin Shen Shengyu Shen Huayi Wu |
| author_facet | Kunlun Qi Chao Yang Chuli Hu Yonglin Shen Shengyu Shen Huayi Wu |
| author_sort | Kunlun Qi |
| collection | DOAJ |
| description | Deep convolutional neural networks (DCNNs) have shown significant improvements in remote sensing image scene classification for powerful feature representations. However, because of the high variance and volume limitations of the available remote sensing datasets, DCNNs are prone to overfit the data used for their training. To address this problem, this paper proposes a novel scene classification framework based on a deep Siamese convolutional network with rotation invariance regularization. Specifically, we design a data augmentation strategy for the Siamese model to learn a rotation invariance DCNN model that is achieved by directly enforcing the labels of the training samples before and after rotating to be mapped close to each other. In addition to the cross-entropy cost function for the traditional CNN models, we impose a rotation invariance regularization constraint on the objective function of our proposed model. The experimental results obtained using three publicly-available scene classification datasets show that the proposed method can generally improve the classification performance by 2~3% and achieves satisfactory classification performance compared with some state-of-the-art methods. |
| first_indexed | 2024-03-09T05:26:29Z |
| format | Article |
| id | doaj.art-a03283c864dc4cfbb899a825a23db533 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T05:26:29Z |
| publishDate | 2021-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-a03283c864dc4cfbb899a825a23db5332023-12-03T12:36:22ZengMDPI AGRemote Sensing2072-42922021-02-0113456910.3390/rs13040569Rotation Invariance Regularization for Remote Sensing Image Scene Classification with Convolutional Neural NetworksKunlun Qi0Chao Yang1Chuli Hu2Yonglin Shen3Shengyu Shen4Huayi Wu5School of Geography and Information Engineering, China University of Geosciences (Wuhan), Wuhan 430078, ChinaSchool of Geography and Information Engineering, China University of Geosciences (Wuhan), Wuhan 430078, ChinaSchool of Geography and Information Engineering, China University of Geosciences (Wuhan), Wuhan 430078, ChinaSchool of Geography and Information Engineering, China University of Geosciences (Wuhan), Wuhan 430078, ChinaSoil and Water Conservation Department, Changjiang River Scientific Research Institute, Wuhan 430010, ChinaLIESMARS, Wuhan University, Wuhan 430079, ChinaDeep convolutional neural networks (DCNNs) have shown significant improvements in remote sensing image scene classification for powerful feature representations. However, because of the high variance and volume limitations of the available remote sensing datasets, DCNNs are prone to overfit the data used for their training. To address this problem, this paper proposes a novel scene classification framework based on a deep Siamese convolutional network with rotation invariance regularization. Specifically, we design a data augmentation strategy for the Siamese model to learn a rotation invariance DCNN model that is achieved by directly enforcing the labels of the training samples before and after rotating to be mapped close to each other. In addition to the cross-entropy cost function for the traditional CNN models, we impose a rotation invariance regularization constraint on the objective function of our proposed model. The experimental results obtained using three publicly-available scene classification datasets show that the proposed method can generally improve the classification performance by 2~3% and achieves satisfactory classification performance compared with some state-of-the-art methods.https://www.mdpi.com/2072-4292/13/4/569convolutional neural networkscene classificationrotation invariance |
| spellingShingle | Kunlun Qi Chao Yang Chuli Hu Yonglin Shen Shengyu Shen Huayi Wu Rotation Invariance Regularization for Remote Sensing Image Scene Classification with Convolutional Neural Networks Remote Sensing convolutional neural network scene classification rotation invariance |
| title | Rotation Invariance Regularization for Remote Sensing Image Scene Classification with Convolutional Neural Networks |
| title_full | Rotation Invariance Regularization for Remote Sensing Image Scene Classification with Convolutional Neural Networks |
| title_fullStr | Rotation Invariance Regularization for Remote Sensing Image Scene Classification with Convolutional Neural Networks |
| title_full_unstemmed | Rotation Invariance Regularization for Remote Sensing Image Scene Classification with Convolutional Neural Networks |
| title_short | Rotation Invariance Regularization for Remote Sensing Image Scene Classification with Convolutional Neural Networks |
| title_sort | rotation invariance regularization for remote sensing image scene classification with convolutional neural networks |
| topic | convolutional neural network scene classification rotation invariance |
| url | https://www.mdpi.com/2072-4292/13/4/569 |
| work_keys_str_mv | AT kunlunqi rotationinvarianceregularizationforremotesensingimagesceneclassificationwithconvolutionalneuralnetworks AT chaoyang rotationinvarianceregularizationforremotesensingimagesceneclassificationwithconvolutionalneuralnetworks AT chulihu rotationinvarianceregularizationforremotesensingimagesceneclassificationwithconvolutionalneuralnetworks AT yonglinshen rotationinvarianceregularizationforremotesensingimagesceneclassificationwithconvolutionalneuralnetworks AT shengyushen rotationinvarianceregularizationforremotesensingimagesceneclassificationwithconvolutionalneuralnetworks AT huayiwu rotationinvarianceregularizationforremotesensingimagesceneclassificationwithconvolutionalneuralnetworks |