Ship Detection for Optical Remote Sensing Images Based on Visual Attention Enhanced Network
Ship detection plays a significant role in military and civil fields. Although some state-of-the-art detection methods, based on convolutional neural networks (CNN) have certain advantages, they still cannot solve the challenge well, including the large size of images, complex scene structure, a lar...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-05-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/19/10/2271 |
| _version_ | 1811297954634399744 |
|---|---|
| author | Fukun Bi Jinyuan Hou Liang Chen Zhihua Yang Yanping Wang |
| author_facet | Fukun Bi Jinyuan Hou Liang Chen Zhihua Yang Yanping Wang |
| author_sort | Fukun Bi |
| collection | DOAJ |
| description | Ship detection plays a significant role in military and civil fields. Although some state-of-the-art detection methods, based on convolutional neural networks (CNN) have certain advantages, they still cannot solve the challenge well, including the large size of images, complex scene structure, a large amount of false alarm interference, and inshore ships. This paper proposes a ship detection method from optical remote sensing images, based on visual attention enhanced network. To effectively reduce false alarm in non-ship area and improve the detection efficiency from remote sensing images, we developed a light-weight local candidate scene network(<inline-formula> <math display="inline"> <semantics> <mrow> <msup> <mi mathvariant="normal">L</mi> <mn>2</mn> </msup> </mrow> </semantics> </math> </inline-formula>CSN) to extract the local candidate scenes with ships. Then, for the selected local candidate scenes, we propose a ship detection method, based on the visual attention DSOD(VA-DSOD). Here, to enhance the detection performance and positioning accuracy of inshore ships, we both extract semantic features, based on DSOD and embed a visual attention enhanced network in DSOD to extract the visual features. We test the detection method on a large number of typical remote sensing datasets, which consist of Google Earth images and GaoFen-2 images. We regard the state-of-the-art method [sliding window DSOD (SW+DSOD)] as a baseline, which achieves the average precision (AP) of 82.33%. The AP of the proposed method increases by 7.53%. The detection and location performance of our proposed method outperforms the baseline in complex remote sensing scenes. |
| first_indexed | 2024-04-13T06:12:45Z |
| format | Article |
| id | doaj.art-78cdaac0767d477f84e35e8bd323d82c |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-13T06:12:45Z |
| publishDate | 2019-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-78cdaac0767d477f84e35e8bd323d82c2022-12-22T02:58:57ZengMDPI AGSensors1424-82202019-05-011910227110.3390/s19102271s19102271Ship Detection for Optical Remote Sensing Images Based on Visual Attention Enhanced NetworkFukun Bi0Jinyuan Hou1Liang Chen2Zhihua Yang3Yanping Wang4School of Information Science and Technology, North China University of Technology, Beijing 100144, ChinaSchool of Information Science and Technology, North China University of Technology, Beijing 100144, ChinaSchool of Information and Electronics, Beijing Institute of Technology, Beijing 100081, ChinaSchool of Information Science and Technology, North China University of Technology, Beijing 100144, ChinaSchool of Information Science and Technology, North China University of Technology, Beijing 100144, ChinaShip detection plays a significant role in military and civil fields. Although some state-of-the-art detection methods, based on convolutional neural networks (CNN) have certain advantages, they still cannot solve the challenge well, including the large size of images, complex scene structure, a large amount of false alarm interference, and inshore ships. This paper proposes a ship detection method from optical remote sensing images, based on visual attention enhanced network. To effectively reduce false alarm in non-ship area and improve the detection efficiency from remote sensing images, we developed a light-weight local candidate scene network(<inline-formula> <math display="inline"> <semantics> <mrow> <msup> <mi mathvariant="normal">L</mi> <mn>2</mn> </msup> </mrow> </semantics> </math> </inline-formula>CSN) to extract the local candidate scenes with ships. Then, for the selected local candidate scenes, we propose a ship detection method, based on the visual attention DSOD(VA-DSOD). Here, to enhance the detection performance and positioning accuracy of inshore ships, we both extract semantic features, based on DSOD and embed a visual attention enhanced network in DSOD to extract the visual features. We test the detection method on a large number of typical remote sensing datasets, which consist of Google Earth images and GaoFen-2 images. We regard the state-of-the-art method [sliding window DSOD (SW+DSOD)] as a baseline, which achieves the average precision (AP) of 82.33%. The AP of the proposed method increases by 7.53%. The detection and location performance of our proposed method outperforms the baseline in complex remote sensing scenes.https://www.mdpi.com/1424-8220/19/10/2271scene classificationship detectionvisual attention enhanced networkDSOD |
| spellingShingle | Fukun Bi Jinyuan Hou Liang Chen Zhihua Yang Yanping Wang Ship Detection for Optical Remote Sensing Images Based on Visual Attention Enhanced Network Sensors scene classification ship detection visual attention enhanced network DSOD |
| title | Ship Detection for Optical Remote Sensing Images Based on Visual Attention Enhanced Network |
| title_full | Ship Detection for Optical Remote Sensing Images Based on Visual Attention Enhanced Network |
| title_fullStr | Ship Detection for Optical Remote Sensing Images Based on Visual Attention Enhanced Network |
| title_full_unstemmed | Ship Detection for Optical Remote Sensing Images Based on Visual Attention Enhanced Network |
| title_short | Ship Detection for Optical Remote Sensing Images Based on Visual Attention Enhanced Network |
| title_sort | ship detection for optical remote sensing images based on visual attention enhanced network |
| topic | scene classification ship detection visual attention enhanced network DSOD |
| url | https://www.mdpi.com/1424-8220/19/10/2271 |
| work_keys_str_mv | AT fukunbi shipdetectionforopticalremotesensingimagesbasedonvisualattentionenhancednetwork AT jinyuanhou shipdetectionforopticalremotesensingimagesbasedonvisualattentionenhancednetwork AT liangchen shipdetectionforopticalremotesensingimagesbasedonvisualattentionenhancednetwork AT zhihuayang shipdetectionforopticalremotesensingimagesbasedonvisualattentionenhancednetwork AT yanpingwang shipdetectionforopticalremotesensingimagesbasedonvisualattentionenhancednetwork |