Automatic Extraction of Water and Shadow from SAR Images Based on a Multi-Resolution Dense Encoder and Decoder Network
The water and shadow areas in SAR images contain rich information for various applications, which cannot be extracted automatically and precisely at present. To handle this problem, a new framework called Multi-Resolution Dense Encoder and Decoder (MRDED) network is proposed, which integrates Convol...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-08-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/19/16/3576 |
| _version_ | 1811185284475256832 |
|---|---|
| author | Peng Zhang Lifu Chen Zhenhong Li Jin Xing Xuemin Xing Zhihui Yuan |
| author_facet | Peng Zhang Lifu Chen Zhenhong Li Jin Xing Xuemin Xing Zhihui Yuan |
| author_sort | Peng Zhang |
| collection | DOAJ |
| description | The water and shadow areas in SAR images contain rich information for various applications, which cannot be extracted automatically and precisely at present. To handle this problem, a new framework called Multi-Resolution Dense Encoder and Decoder (MRDED) network is proposed, which integrates Convolutional Neural Network (CNN), Residual Network (ResNet), Dense Convolutional Network (DenseNet), Global Convolutional Network (GCN), and Convolutional Long Short-Term Memory (ConvLSTM). MRDED contains three parts: the Gray Level Gradient Co-occurrence Matrix (GLGCM), the Encoder network, and the Decoder network. GLGCM is used to extract low-level features, which are further processed by the Encoder. The Encoder network employs ResNet to extract features at different resolutions. There are two components of the Decoder network, namely, the Multi-level Features Extraction and Fusion (MFEF) and Score maps Fusion (SF). We implement two versions of MFEF, named MFEF1 and MFEF2, which generate separate score maps. The difference between them lies in that the Chained Residual Pooling (CRP) module is utilized in MFEF2, while ConvLSTM is adopted in MFEF1 to form the Improved Chained Residual Pooling (ICRP) module as the replacement. The two separate score maps generated by MFEF1 and MFEF2 are fused with different weights to produce the fused score map, which is further handled by the Softmax function to generate the final extraction results for water and shadow areas. To evaluate the proposed framework, MRDED is trained and tested with large SAR images. To further assess the classification performance, a total of eight different classification frameworks are compared with our proposed framework. MRDED outperformed by reaching 80.12% in Pixel Accuracy (PA) and 73.88% in Intersection of Union (IoU) for water, 88% in PA and 77.11% in IoU for shadow, and 95.16% in PA and 90.49% in IoU for background classification, respectively. |
| first_indexed | 2024-04-11T13:26:54Z |
| format | Article |
| id | doaj.art-93f0d2e2e5fa4402a7927f0caa3864d7 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-11T13:26:54Z |
| publishDate | 2019-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-93f0d2e2e5fa4402a7927f0caa3864d72022-12-22T04:22:02ZengMDPI AGSensors1424-82202019-08-011916357610.3390/s19163576s19163576Automatic Extraction of Water and Shadow from SAR Images Based on a Multi-Resolution Dense Encoder and Decoder NetworkPeng Zhang0Lifu Chen1Zhenhong Li2Jin Xing3Xuemin Xing4Zhihui Yuan5School of Electrical and Information Engineering, Changsha University of Science & Technology, Changsha 410114, ChinaSchool of Electrical and Information Engineering, Changsha University of Science & Technology, Changsha 410114, ChinaSchool of Engineering, Newcastle University, Newcastle upon Tyne NE1 7RU, UKSchool of Engineering, Newcastle University, Newcastle upon Tyne NE1 7RU, UKLaboratory of Radar Remote Sensing Applications, Changsha University of Science & Technology, Changsha 410014, ChinaSchool of Electrical and Information Engineering, Changsha University of Science & Technology, Changsha 410114, ChinaThe water and shadow areas in SAR images contain rich information for various applications, which cannot be extracted automatically and precisely at present. To handle this problem, a new framework called Multi-Resolution Dense Encoder and Decoder (MRDED) network is proposed, which integrates Convolutional Neural Network (CNN), Residual Network (ResNet), Dense Convolutional Network (DenseNet), Global Convolutional Network (GCN), and Convolutional Long Short-Term Memory (ConvLSTM). MRDED contains three parts: the Gray Level Gradient Co-occurrence Matrix (GLGCM), the Encoder network, and the Decoder network. GLGCM is used to extract low-level features, which are further processed by the Encoder. The Encoder network employs ResNet to extract features at different resolutions. There are two components of the Decoder network, namely, the Multi-level Features Extraction and Fusion (MFEF) and Score maps Fusion (SF). We implement two versions of MFEF, named MFEF1 and MFEF2, which generate separate score maps. The difference between them lies in that the Chained Residual Pooling (CRP) module is utilized in MFEF2, while ConvLSTM is adopted in MFEF1 to form the Improved Chained Residual Pooling (ICRP) module as the replacement. The two separate score maps generated by MFEF1 and MFEF2 are fused with different weights to produce the fused score map, which is further handled by the Softmax function to generate the final extraction results for water and shadow areas. To evaluate the proposed framework, MRDED is trained and tested with large SAR images. To further assess the classification performance, a total of eight different classification frameworks are compared with our proposed framework. MRDED outperformed by reaching 80.12% in Pixel Accuracy (PA) and 73.88% in Intersection of Union (IoU) for water, 88% in PA and 77.11% in IoU for shadow, and 95.16% in PA and 90.49% in IoU for background classification, respectively.https://www.mdpi.com/1424-8220/19/16/3576water extractionshadow extractiondeep learningsynthetic aperture radar (SAR)classificationconvolutional neural network (CNN)global convolutional network (GCN)dense convolutional network (DenseNet)CONVOLUTION LONG SHORT-TERM MEMORY (ConvLSTM) |
| spellingShingle | Peng Zhang Lifu Chen Zhenhong Li Jin Xing Xuemin Xing Zhihui Yuan Automatic Extraction of Water and Shadow from SAR Images Based on a Multi-Resolution Dense Encoder and Decoder Network Sensors water extraction shadow extraction deep learning synthetic aperture radar (SAR) classification convolutional neural network (CNN) global convolutional network (GCN) dense convolutional network (DenseNet) CONVOLUTION LONG SHORT-TERM MEMORY (ConvLSTM) |
| title | Automatic Extraction of Water and Shadow from SAR Images Based on a Multi-Resolution Dense Encoder and Decoder Network |
| title_full | Automatic Extraction of Water and Shadow from SAR Images Based on a Multi-Resolution Dense Encoder and Decoder Network |
| title_fullStr | Automatic Extraction of Water and Shadow from SAR Images Based on a Multi-Resolution Dense Encoder and Decoder Network |
| title_full_unstemmed | Automatic Extraction of Water and Shadow from SAR Images Based on a Multi-Resolution Dense Encoder and Decoder Network |
| title_short | Automatic Extraction of Water and Shadow from SAR Images Based on a Multi-Resolution Dense Encoder and Decoder Network |
| title_sort | automatic extraction of water and shadow from sar images based on a multi resolution dense encoder and decoder network |
| topic | water extraction shadow extraction deep learning synthetic aperture radar (SAR) classification convolutional neural network (CNN) global convolutional network (GCN) dense convolutional network (DenseNet) CONVOLUTION LONG SHORT-TERM MEMORY (ConvLSTM) |
| url | https://www.mdpi.com/1424-8220/19/16/3576 |
| work_keys_str_mv | AT pengzhang automaticextractionofwaterandshadowfromsarimagesbasedonamultiresolutiondenseencoderanddecodernetwork AT lifuchen automaticextractionofwaterandshadowfromsarimagesbasedonamultiresolutiondenseencoderanddecodernetwork AT zhenhongli automaticextractionofwaterandshadowfromsarimagesbasedonamultiresolutiondenseencoderanddecodernetwork AT jinxing automaticextractionofwaterandshadowfromsarimagesbasedonamultiresolutiondenseencoderanddecodernetwork AT xueminxing automaticextractionofwaterandshadowfromsarimagesbasedonamultiresolutiondenseencoderanddecodernetwork AT zhihuiyuan automaticextractionofwaterandshadowfromsarimagesbasedonamultiresolutiondenseencoderanddecodernetwork |