Feature Weighted Attention—Bidirectional Long Short Term Memory Model for Change Detection in Remote Sensing Images
In remote sensing images, change detection (CD) is required in many applications, such as: resource management, urban expansion research, land management, and disaster assessment. Various deep learning-based methods were applied to satellite image analysis for change detection, yet many of them have...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-10-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/21/5402 |
| _version_ | 1797466581754707968 |
|---|---|
| author | Raj Kumar Patra Sujata N. Patil Przemysław Falkowski-Gilski Zbigniew Łubniewski Rachana Poongodan |
| author_facet | Raj Kumar Patra Sujata N. Patil Przemysław Falkowski-Gilski Zbigniew Łubniewski Rachana Poongodan |
| author_sort | Raj Kumar Patra |
| collection | DOAJ |
| description | In remote sensing images, change detection (CD) is required in many applications, such as: resource management, urban expansion research, land management, and disaster assessment. Various deep learning-based methods were applied to satellite image analysis for change detection, yet many of them have limitations, including the overfitting problem. This research proposes the Feature Weighted Attention (FWA) in Bidirectional Long Short-Term Memory (BiLSTM) method to reduce the overfitting problem and increase the performance of classification in change detection applications. Additionally, data usage and accuracy in remote sensing activities, particularly CD, can be significantly improved by a large number of training models based on BiLSTM. Normalization techniques are applied to input images in order to enhance the quality and reduce the difference in pixel value. The AlexNet and VGG16 models were used to extract useful features from the normalized images. The extracted features were then applied to the FWA-BiLSTM model, to give more weight to the unique features and increase the efficiency of classification. The attention layer selects the unique features that help to distinguish the changes in the remote sensing images. From the experimental results, it was clearly shown that the proposed FWA-BiLSTM model achieved better performance in terms of precision (93.43%), recall (93.16%), and overall accuracy (99.26%), when compared with the existing Difference-enhancement Dense-attention Convolutional Neural Network (DDCNN) model. |
| first_indexed | 2024-03-09T18:41:48Z |
| format | Article |
| id | doaj.art-2156ec7c358246eabb7ed944b5762c8e |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T18:41:48Z |
| publishDate | 2022-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-2156ec7c358246eabb7ed944b5762c8e2023-11-24T06:38:27ZengMDPI AGRemote Sensing2072-42922022-10-011421540210.3390/rs14215402Feature Weighted Attention—Bidirectional Long Short Term Memory Model for Change Detection in Remote Sensing ImagesRaj Kumar Patra0Sujata N. Patil1Przemysław Falkowski-Gilski2Zbigniew Łubniewski3Rachana Poongodan4Department of Computer Science and Engineering, CMR Technical Campus, Hyderabad 501401, IndiaDepartment of Electronics and Communication Engineering, KLE Dr. M S Sheshgiri College of Engineering and Technology, Karnataka 590008, IndiaFaculty of Electronics, Telecommunications and Informatics, Gdansk University of Technology, Narutowicza 11/12, 80-233 Gdansk, PolandFaculty of Electronics, Telecommunications and Informatics, Gdansk University of Technology, Narutowicza 11/12, 80-233 Gdansk, PolandDepartment of Computer Science and Engineering, New Horizon College of Engineering, Bangalore 560103, IndiaIn remote sensing images, change detection (CD) is required in many applications, such as: resource management, urban expansion research, land management, and disaster assessment. Various deep learning-based methods were applied to satellite image analysis for change detection, yet many of them have limitations, including the overfitting problem. This research proposes the Feature Weighted Attention (FWA) in Bidirectional Long Short-Term Memory (BiLSTM) method to reduce the overfitting problem and increase the performance of classification in change detection applications. Additionally, data usage and accuracy in remote sensing activities, particularly CD, can be significantly improved by a large number of training models based on BiLSTM. Normalization techniques are applied to input images in order to enhance the quality and reduce the difference in pixel value. The AlexNet and VGG16 models were used to extract useful features from the normalized images. The extracted features were then applied to the FWA-BiLSTM model, to give more weight to the unique features and increase the efficiency of classification. The attention layer selects the unique features that help to distinguish the changes in the remote sensing images. From the experimental results, it was clearly shown that the proposed FWA-BiLSTM model achieved better performance in terms of precision (93.43%), recall (93.16%), and overall accuracy (99.26%), when compared with the existing Difference-enhancement Dense-attention Convolutional Neural Network (DDCNN) model.https://www.mdpi.com/2072-4292/14/21/5402AlexNetbidirectional long short-term memorychange detectionfeature weighted attentionVGG16 |
| spellingShingle | Raj Kumar Patra Sujata N. Patil Przemysław Falkowski-Gilski Zbigniew Łubniewski Rachana Poongodan Feature Weighted Attention—Bidirectional Long Short Term Memory Model for Change Detection in Remote Sensing Images Remote Sensing AlexNet bidirectional long short-term memory change detection feature weighted attention VGG16 |
| title | Feature Weighted Attention—Bidirectional Long Short Term Memory Model for Change Detection in Remote Sensing Images |
| title_full | Feature Weighted Attention—Bidirectional Long Short Term Memory Model for Change Detection in Remote Sensing Images |
| title_fullStr | Feature Weighted Attention—Bidirectional Long Short Term Memory Model for Change Detection in Remote Sensing Images |
| title_full_unstemmed | Feature Weighted Attention—Bidirectional Long Short Term Memory Model for Change Detection in Remote Sensing Images |
| title_short | Feature Weighted Attention—Bidirectional Long Short Term Memory Model for Change Detection in Remote Sensing Images |
| title_sort | feature weighted attention bidirectional long short term memory model for change detection in remote sensing images |
| topic | AlexNet bidirectional long short-term memory change detection feature weighted attention VGG16 |
| url | https://www.mdpi.com/2072-4292/14/21/5402 |
| work_keys_str_mv | AT rajkumarpatra featureweightedattentionbidirectionallongshorttermmemorymodelforchangedetectioninremotesensingimages AT sujatanpatil featureweightedattentionbidirectionallongshorttermmemorymodelforchangedetectioninremotesensingimages AT przemysławfalkowskigilski featureweightedattentionbidirectionallongshorttermmemorymodelforchangedetectioninremotesensingimages AT zbigniewłubniewski featureweightedattentionbidirectionallongshorttermmemorymodelforchangedetectioninremotesensingimages AT rachanapoongodan featureweightedattentionbidirectionallongshorttermmemorymodelforchangedetectioninremotesensingimages |