TRDet: Two-Stage Rotated Detection of Rural Buildings in Remote Sensing Images
Fast and accurate acquisition of the outline of rural buildings on remote sensing images is an efficient method to monitor illegal rural buildings. The traditional object detection method produces useless background information when detecting rural buildings; the semantic segmentation method cannot...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-01-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/3/522 |
| _version_ | 1827659127727325184 |
|---|---|
| author | Baochai Peng Dong Ren Cheng Zheng Anxiang Lu |
| author_facet | Baochai Peng Dong Ren Cheng Zheng Anxiang Lu |
| author_sort | Baochai Peng |
| collection | DOAJ |
| description | Fast and accurate acquisition of the outline of rural buildings on remote sensing images is an efficient method to monitor illegal rural buildings. The traditional object detection method produces useless background information when detecting rural buildings; the semantic segmentation method cannot accurately segment the contours between buildings; the instance segmentation method cannot obtain regular building contours. The rotated object detection methods can effectively solve the problem that the traditional artificial intelligence method cannot accurately extract the outline of buildings. However, the rotated object detection methods are easy to lose location information of small objects in advanced feature maps and are sensitive to noise. To resolve these problems, this paper proposes a two-stage rotated object detection network for rural buildings (TRDet) by using a deep feature fusion network (DFF-Net) and a pixel attention module (PAM). Specifically, TRDet first fuses low-level location and high-level semantic information through the DFF-Net and then reduces the interference of noise information to the network through the PAM. The experimental results show that the mean average precession (mAP), precision, recall rate, and F1 score of the proposed TRDet are 83.57%, 91.11%, 86.5%, and 88.74%, respectively, which outperform the R2CNN model by 15%, 15.54%, 4.01%, and 9.87%. The results demonstrate that the TRDet can achieve better detection in small rural buildings and dense rural buildings. |
| first_indexed | 2024-03-09T23:15:12Z |
| format | Article |
| id | doaj.art-f4342b844be24a2f87ae7c1c8ffe6dbb |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T23:15:12Z |
| publishDate | 2022-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-f4342b844be24a2f87ae7c1c8ffe6dbb2023-11-23T17:39:05ZengMDPI AGRemote Sensing2072-42922022-01-0114352210.3390/rs14030522TRDet: Two-Stage Rotated Detection of Rural Buildings in Remote Sensing ImagesBaochai Peng0Dong Ren1Cheng Zheng2Anxiang Lu3Hubei Engineering Technology Research Center for Farmland Environmental Monitoring, China Three Gorges University, Yichang 443002, ChinaHubei Engineering Technology Research Center for Farmland Environmental Monitoring, China Three Gorges University, Yichang 443002, ChinaHubei Engineering Technology Research Center for Farmland Environmental Monitoring, China Three Gorges University, Yichang 443002, ChinaHubei Engineering Technology Research Center for Farmland Environmental Monitoring, China Three Gorges University, Yichang 443002, ChinaFast and accurate acquisition of the outline of rural buildings on remote sensing images is an efficient method to monitor illegal rural buildings. The traditional object detection method produces useless background information when detecting rural buildings; the semantic segmentation method cannot accurately segment the contours between buildings; the instance segmentation method cannot obtain regular building contours. The rotated object detection methods can effectively solve the problem that the traditional artificial intelligence method cannot accurately extract the outline of buildings. However, the rotated object detection methods are easy to lose location information of small objects in advanced feature maps and are sensitive to noise. To resolve these problems, this paper proposes a two-stage rotated object detection network for rural buildings (TRDet) by using a deep feature fusion network (DFF-Net) and a pixel attention module (PAM). Specifically, TRDet first fuses low-level location and high-level semantic information through the DFF-Net and then reduces the interference of noise information to the network through the PAM. The experimental results show that the mean average precession (mAP), precision, recall rate, and F1 score of the proposed TRDet are 83.57%, 91.11%, 86.5%, and 88.74%, respectively, which outperform the R2CNN model by 15%, 15.54%, 4.01%, and 9.87%. The results demonstrate that the TRDet can achieve better detection in small rural buildings and dense rural buildings.https://www.mdpi.com/2072-4292/14/3/522rotated object detectionrural buildingsfeature fusionpixel attention |
| spellingShingle | Baochai Peng Dong Ren Cheng Zheng Anxiang Lu TRDet: Two-Stage Rotated Detection of Rural Buildings in Remote Sensing Images Remote Sensing rotated object detection rural buildings feature fusion pixel attention |
| title | TRDet: Two-Stage Rotated Detection of Rural Buildings in Remote Sensing Images |
| title_full | TRDet: Two-Stage Rotated Detection of Rural Buildings in Remote Sensing Images |
| title_fullStr | TRDet: Two-Stage Rotated Detection of Rural Buildings in Remote Sensing Images |
| title_full_unstemmed | TRDet: Two-Stage Rotated Detection of Rural Buildings in Remote Sensing Images |
| title_short | TRDet: Two-Stage Rotated Detection of Rural Buildings in Remote Sensing Images |
| title_sort | trdet two stage rotated detection of rural buildings in remote sensing images |
| topic | rotated object detection rural buildings feature fusion pixel attention |
| url | https://www.mdpi.com/2072-4292/14/3/522 |
| work_keys_str_mv | AT baochaipeng trdettwostagerotateddetectionofruralbuildingsinremotesensingimages AT dongren trdettwostagerotateddetectionofruralbuildingsinremotesensingimages AT chengzheng trdettwostagerotateddetectionofruralbuildingsinremotesensingimages AT anxianglu trdettwostagerotateddetectionofruralbuildingsinremotesensingimages |