A Cross Stage Partial Network with Strengthen Matching Detector for Remote Sensing Object Detection
Remote sensing object detection is a difficult task because it often requires real-time feedback through numerous objects in complex environments. In object detection, Feature Pyramids Networks (FPN) have been widely used for better representations based on a multi-scale problem. However, the multip...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-03-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/6/1574 |
| _version_ | 1797609183604899840 |
|---|---|
| author | Shougang Ren Zhiruo Fang Xingjian Gu |
| author_facet | Shougang Ren Zhiruo Fang Xingjian Gu |
| author_sort | Shougang Ren |
| collection | DOAJ |
| description | Remote sensing object detection is a difficult task because it often requires real-time feedback through numerous objects in complex environments. In object detection, Feature Pyramids Networks (FPN) have been widely used for better representations based on a multi-scale problem. However, the multiple level features cause detectors’ structures to be complex and makes redundant calculations that slow down the detector. This paper uses a single-layer feature to make the detection lightweight and accurate without relying on Feature Pyramid Structures. We proposed a method called the Cross Stage Partial Strengthen Matching Detector (StrMCsDet). The StrMCsDet generates a single-level feature map architecture in the backbone with a cross stage partial network. To provide an alternative way of replacing the traditional feature pyramid, a multi-scale encoder was designed to compensate the receptive field limitation. Additionally, a stronger matching strategy was proposed to make sure that various scale anchors may be equally matched. The StrMCsDet is different from the conventional full pyramid structure and fully exploits the feature map which deals with a multi-scale encoder. Methods achieved both comparable precision and speed for practical applications. Experiments conducted on the DIOR dataset and the NWPU-VHR-10 dataset achieved 65.6 and 73.5 mAP on 1080 Ti, respectively, which can match the performance of state-of-the-art works. Moreover, StrMCsDet requires less computation and achieved 38.5 FPS on the DIOR dataset. |
| first_indexed | 2024-03-11T05:56:58Z |
| format | Article |
| id | doaj.art-1a931a45b62940e9be81e3f7f29b3677 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-11T05:56:58Z |
| publishDate | 2023-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-1a931a45b62940e9be81e3f7f29b36772023-11-17T13:38:58ZengMDPI AGRemote Sensing2072-42922023-03-01156157410.3390/rs15061574A Cross Stage Partial Network with Strengthen Matching Detector for Remote Sensing Object DetectionShougang Ren0Zhiruo Fang1Xingjian Gu2College of Artificial Intelligence, Nanjing Agricultural University, Nanjing 210095, ChinaCollege of Artificial Intelligence, Nanjing Agricultural University, Nanjing 210095, ChinaCollege of Artificial Intelligence, Nanjing Agricultural University, Nanjing 210095, ChinaRemote sensing object detection is a difficult task because it often requires real-time feedback through numerous objects in complex environments. In object detection, Feature Pyramids Networks (FPN) have been widely used for better representations based on a multi-scale problem. However, the multiple level features cause detectors’ structures to be complex and makes redundant calculations that slow down the detector. This paper uses a single-layer feature to make the detection lightweight and accurate without relying on Feature Pyramid Structures. We proposed a method called the Cross Stage Partial Strengthen Matching Detector (StrMCsDet). The StrMCsDet generates a single-level feature map architecture in the backbone with a cross stage partial network. To provide an alternative way of replacing the traditional feature pyramid, a multi-scale encoder was designed to compensate the receptive field limitation. Additionally, a stronger matching strategy was proposed to make sure that various scale anchors may be equally matched. The StrMCsDet is different from the conventional full pyramid structure and fully exploits the feature map which deals with a multi-scale encoder. Methods achieved both comparable precision and speed for practical applications. Experiments conducted on the DIOR dataset and the NWPU-VHR-10 dataset achieved 65.6 and 73.5 mAP on 1080 Ti, respectively, which can match the performance of state-of-the-art works. Moreover, StrMCsDet requires less computation and achieved 38.5 FPS on the DIOR dataset.https://www.mdpi.com/2072-4292/15/6/1574object detectionone-stage detectormulti-scaleStrMCsDet |
| spellingShingle | Shougang Ren Zhiruo Fang Xingjian Gu A Cross Stage Partial Network with Strengthen Matching Detector for Remote Sensing Object Detection Remote Sensing object detection one-stage detector multi-scale StrMCsDet |
| title | A Cross Stage Partial Network with Strengthen Matching Detector for Remote Sensing Object Detection |
| title_full | A Cross Stage Partial Network with Strengthen Matching Detector for Remote Sensing Object Detection |
| title_fullStr | A Cross Stage Partial Network with Strengthen Matching Detector for Remote Sensing Object Detection |
| title_full_unstemmed | A Cross Stage Partial Network with Strengthen Matching Detector for Remote Sensing Object Detection |
| title_short | A Cross Stage Partial Network with Strengthen Matching Detector for Remote Sensing Object Detection |
| title_sort | cross stage partial network with strengthen matching detector for remote sensing object detection |
| topic | object detection one-stage detector multi-scale StrMCsDet |
| url | https://www.mdpi.com/2072-4292/15/6/1574 |
| work_keys_str_mv | AT shougangren acrossstagepartialnetworkwithstrengthenmatchingdetectorforremotesensingobjectdetection AT zhiruofang acrossstagepartialnetworkwithstrengthenmatchingdetectorforremotesensingobjectdetection AT xingjiangu acrossstagepartialnetworkwithstrengthenmatchingdetectorforremotesensingobjectdetection AT shougangren crossstagepartialnetworkwithstrengthenmatchingdetectorforremotesensingobjectdetection AT zhiruofang crossstagepartialnetworkwithstrengthenmatchingdetectorforremotesensingobjectdetection AT xingjiangu crossstagepartialnetworkwithstrengthenmatchingdetectorforremotesensingobjectdetection |