An Anchor-Free Network for Increasing Attention to Small Objects in High Resolution Remote Sensing Images
Aimed at the problems of small object detection in high resolution remote sensing images, such as difficult detection, diverse scales, and dense distribution, this study proposes a new method, DCE_YOLOX, which is more focused on small objects. The method uses depthwise separable deconvolution for up...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/13/4/2073 |
| _version_ | 1797622626110144512 |
|---|---|
| author | Huaping Zhou Wei Guo Qi Zhao |
| author_facet | Huaping Zhou Wei Guo Qi Zhao |
| author_sort | Huaping Zhou |
| collection | DOAJ |
| description | Aimed at the problems of small object detection in high resolution remote sensing images, such as difficult detection, diverse scales, and dense distribution, this study proposes a new method, DCE_YOLOX, which is more focused on small objects. The method uses depthwise separable deconvolution for upsampling, which can effectively recover lost feature information and combines dilated convolution and CoTNet to extract local contextual features, which can make full use of the hidden semantic information. At the same time, EcaNet is added to the enhanced feature extraction network of the baseline model to make the model more focused on information-rich features; secondly, the network input resolution is optimized, which can avoid the impact of image scaling to a certain extent and improve the accuracy of small object detection. Finally, CSL is used to calculate the angular loss to achieve the rotated object detection of remote sensing images. The proposed method in this study achieves 83.9% accuracy and 76.7% accuracy for horizontal object detection and rotationally invariant object detection, respectively, in the DOTA remote sensing dataset; it even achieves 96% accuracy for rotationally invariant object detection in the HRSC2016 dataset. It can be concluded that our algorithm has a better focus on small objects, while it has an equally good focus on other objects and is well suited for applications in remote sensing, and it has certain reference significance for realizing the detection of small objects in remote sensing images. |
| first_indexed | 2024-03-11T09:13:00Z |
| format | Article |
| id | doaj.art-6552e63194774e90b70c14550f0ee4cf |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-11T09:13:00Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-6552e63194774e90b70c14550f0ee4cf2023-11-16T18:50:29ZengMDPI AGApplied Sciences2076-34172023-02-01134207310.3390/app13042073An Anchor-Free Network for Increasing Attention to Small Objects in High Resolution Remote Sensing ImagesHuaping Zhou0Wei Guo1Qi Zhao2School of Computer Science and Engineering, Anhui University of Science and Technology, Huainan 232001, ChinaSchool of Computer Science and Engineering, Anhui University of Science and Technology, Huainan 232001, ChinaSchool of Computer Science and Engineering, Anhui University of Science and Technology, Huainan 232001, ChinaAimed at the problems of small object detection in high resolution remote sensing images, such as difficult detection, diverse scales, and dense distribution, this study proposes a new method, DCE_YOLOX, which is more focused on small objects. The method uses depthwise separable deconvolution for upsampling, which can effectively recover lost feature information and combines dilated convolution and CoTNet to extract local contextual features, which can make full use of the hidden semantic information. At the same time, EcaNet is added to the enhanced feature extraction network of the baseline model to make the model more focused on information-rich features; secondly, the network input resolution is optimized, which can avoid the impact of image scaling to a certain extent and improve the accuracy of small object detection. Finally, CSL is used to calculate the angular loss to achieve the rotated object detection of remote sensing images. The proposed method in this study achieves 83.9% accuracy and 76.7% accuracy for horizontal object detection and rotationally invariant object detection, respectively, in the DOTA remote sensing dataset; it even achieves 96% accuracy for rotationally invariant object detection in the HRSC2016 dataset. It can be concluded that our algorithm has a better focus on small objects, while it has an equally good focus on other objects and is well suited for applications in remote sensing, and it has certain reference significance for realizing the detection of small objects in remote sensing images.https://www.mdpi.com/2076-3417/13/4/2073YOLOXdepthwise separable deconvolutiondilated convolutionsmall object detectionCSL |
| spellingShingle | Huaping Zhou Wei Guo Qi Zhao An Anchor-Free Network for Increasing Attention to Small Objects in High Resolution Remote Sensing Images Applied Sciences YOLOX depthwise separable deconvolution dilated convolution small object detection CSL |
| title | An Anchor-Free Network for Increasing Attention to Small Objects in High Resolution Remote Sensing Images |
| title_full | An Anchor-Free Network for Increasing Attention to Small Objects in High Resolution Remote Sensing Images |
| title_fullStr | An Anchor-Free Network for Increasing Attention to Small Objects in High Resolution Remote Sensing Images |
| title_full_unstemmed | An Anchor-Free Network for Increasing Attention to Small Objects in High Resolution Remote Sensing Images |
| title_short | An Anchor-Free Network for Increasing Attention to Small Objects in High Resolution Remote Sensing Images |
| title_sort | anchor free network for increasing attention to small objects in high resolution remote sensing images |
| topic | YOLOX depthwise separable deconvolution dilated convolution small object detection CSL |
| url | https://www.mdpi.com/2076-3417/13/4/2073 |
| work_keys_str_mv | AT huapingzhou ananchorfreenetworkforincreasingattentiontosmallobjectsinhighresolutionremotesensingimages AT weiguo ananchorfreenetworkforincreasingattentiontosmallobjectsinhighresolutionremotesensingimages AT qizhao ananchorfreenetworkforincreasingattentiontosmallobjectsinhighresolutionremotesensingimages AT huapingzhou anchorfreenetworkforincreasingattentiontosmallobjectsinhighresolutionremotesensingimages AT weiguo anchorfreenetworkforincreasingattentiontosmallobjectsinhighresolutionremotesensingimages AT qizhao anchorfreenetworkforincreasingattentiontosmallobjectsinhighresolutionremotesensingimages |