BSFCDet: Bidirectional Spatial–Semantic Fusion Network Coupled with Channel Attention for Object Detection in Satellite Images
Due to the increasing maturity of deep learning and remote sensing technology, the performance of object detection in satellite images has significantly improved and plays an important role in military reconnaissance, urban planning, and agricultural monitoring. However, satellite images have challe...
| Asıl Yazarlar: | , , |
|---|---|
| Materyal Türü: | Makale |
| Dil: | English |
| Baskı/Yayın Bilgisi: |
MDPI AG
2023-06-01
|
| Seri Bilgileri: | Remote Sensing |
| Konular: | |
| Online Erişim: | https://www.mdpi.com/2072-4292/15/13/3213 |
| _version_ | 1827734669217497088 |
|---|---|
| author | Xinchi Wei Yan Zhang Yuhui Zheng |
| author_facet | Xinchi Wei Yan Zhang Yuhui Zheng |
| author_sort | Xinchi Wei |
| collection | DOAJ |
| description | Due to the increasing maturity of deep learning and remote sensing technology, the performance of object detection in satellite images has significantly improved and plays an important role in military reconnaissance, urban planning, and agricultural monitoring. However, satellite images have challenges such as small objects, multiscale objects, and complex backgrounds. To solve these problems, a lightweight object detection model named BSFCDet is proposed. First, fast spatial pyramid pooling (SPPF-G) is designed for feature fusion to enrich the spatial information of small targets. Second, a three-layer bidirectional feature pyramid network (BiFPN-G) is suggested to integrate the deep feature’s semantic information with the shallow feature’s spatial information, thus improving the scale adaptability of the model. Third, a novel efficient channel attention (ECAM) is proposed to reduce background interference. Last, a new residual block (Resblock_M) is constructed to balance accuracy and speed. BSFCDet achieves high detection performance while satisfying real-time performance, according to experimental results. |
| first_indexed | 2024-03-11T01:30:49Z |
| format | Article |
| id | doaj.art-eb00e214eff04d548b28cb197aadad4e |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-11T01:30:49Z |
| publishDate | 2023-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-eb00e214eff04d548b28cb197aadad4e2023-11-18T17:22:59ZengMDPI AGRemote Sensing2072-42922023-06-011513321310.3390/rs15133213BSFCDet: Bidirectional Spatial–Semantic Fusion Network Coupled with Channel Attention for Object Detection in Satellite ImagesXinchi Wei0Yan Zhang1Yuhui Zheng2Engineering Research Center of Digital Forensics, Ministry of Education, Nanjing University of Information Science and Technology, Nanjing 210044, ChinaEngineering Research Center of Digital Forensics, Ministry of Education, Nanjing University of Information Science and Technology, Nanjing 210044, ChinaEngineering Research Center of Digital Forensics, Ministry of Education, Nanjing University of Information Science and Technology, Nanjing 210044, ChinaDue to the increasing maturity of deep learning and remote sensing technology, the performance of object detection in satellite images has significantly improved and plays an important role in military reconnaissance, urban planning, and agricultural monitoring. However, satellite images have challenges such as small objects, multiscale objects, and complex backgrounds. To solve these problems, a lightweight object detection model named BSFCDet is proposed. First, fast spatial pyramid pooling (SPPF-G) is designed for feature fusion to enrich the spatial information of small targets. Second, a three-layer bidirectional feature pyramid network (BiFPN-G) is suggested to integrate the deep feature’s semantic information with the shallow feature’s spatial information, thus improving the scale adaptability of the model. Third, a novel efficient channel attention (ECAM) is proposed to reduce background interference. Last, a new residual block (Resblock_M) is constructed to balance accuracy and speed. BSFCDet achieves high detection performance while satisfying real-time performance, according to experimental results.https://www.mdpi.com/2072-4292/15/13/3213feature fusionchannel attentionobject detectionsatellite imageslightweight |
| spellingShingle | Xinchi Wei Yan Zhang Yuhui Zheng BSFCDet: Bidirectional Spatial–Semantic Fusion Network Coupled with Channel Attention for Object Detection in Satellite Images Remote Sensing feature fusion channel attention object detection satellite images lightweight |
| title | BSFCDet: Bidirectional Spatial–Semantic Fusion Network Coupled with Channel Attention for Object Detection in Satellite Images |
| title_full | BSFCDet: Bidirectional Spatial–Semantic Fusion Network Coupled with Channel Attention for Object Detection in Satellite Images |
| title_fullStr | BSFCDet: Bidirectional Spatial–Semantic Fusion Network Coupled with Channel Attention for Object Detection in Satellite Images |
| title_full_unstemmed | BSFCDet: Bidirectional Spatial–Semantic Fusion Network Coupled with Channel Attention for Object Detection in Satellite Images |
| title_short | BSFCDet: Bidirectional Spatial–Semantic Fusion Network Coupled with Channel Attention for Object Detection in Satellite Images |
| title_sort | bsfcdet bidirectional spatial semantic fusion network coupled with channel attention for object detection in satellite images |
| topic | feature fusion channel attention object detection satellite images lightweight |
| url | https://www.mdpi.com/2072-4292/15/13/3213 |
| work_keys_str_mv | AT xinchiwei bsfcdetbidirectionalspatialsemanticfusionnetworkcoupledwithchannelattentionforobjectdetectioninsatelliteimages AT yanzhang bsfcdetbidirectionalspatialsemanticfusionnetworkcoupledwithchannelattentionforobjectdetectioninsatelliteimages AT yuhuizheng bsfcdetbidirectionalspatialsemanticfusionnetworkcoupledwithchannelattentionforobjectdetectioninsatelliteimages |