YOLO-Extract: Improved YOLOv5 for Aircraft Object Detection in Remote Sensing Images
Compared with natural images, remote sensing targets have small and dense target shapes as well as complex target backgrounds. As a result, insufficient detection accuracy and target location cannot be accurately identified. So, this paper proposes the YOLO-extract algorithm based on the YOLOv5 algo...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10005162/ |
| _version_ | 1797956352073531392 |
|---|---|
| author | Zhiguo Liu Yuan Gao Qianqian Du Meng Chen Wenqiang Lv |
| author_facet | Zhiguo Liu Yuan Gao Qianqian Du Meng Chen Wenqiang Lv |
| author_sort | Zhiguo Liu |
| collection | DOAJ |
| description | Compared with natural images, remote sensing targets have small and dense target shapes as well as complex target backgrounds. As a result, insufficient detection accuracy and target location cannot be accurately identified. So, this paper proposes the YOLO-extract algorithm based on the YOLOv5 algorithm. Firstly, The YOLO-extract algorithm optimized the model structure of the YOLOv5 algorithm. The YOLO-extract algorithm not only deleted the feature layer and prediction head with poor feature extraction ability but also a new feature extractor with stronger feature extraction ability was integrated into the network. At the same time, YOLO-extract borrowed the idea of residual network to integrate Coordinate Attention into the network. Secondly, the mixed dilated convolution was combined with the redesigned residual structure to enhance the feature and location information extraction ability of the shallow layer of the model and optimize the feature extraction ability of the model for different scale targets. Finally, drawing on the idea of <inline-formula> <tex-math notation="LaTeX">$\alpha $ </tex-math></inline-formula>-IoU Loss, Focal-<inline-formula> <tex-math notation="LaTeX">$\alpha $ </tex-math></inline-formula> EIoU Loss was designed to replace CIoU Loss, which makes the model bounding box regression faster and the loss lower. The experimental results on the test data set show that compared with the YOLOv5 algorithm, the YOLO-extract algorithm has a faster convergence speed, reduces the calculation amount by 45.3GFLOPs and the number of parameters by 10.526M, but increases the mAP by 8.1% and the detection speed by 3 times. |
| first_indexed | 2024-04-10T23:47:38Z |
| format | Article |
| id | doaj.art-07de8bdd2f1645a98154ec5e6d24dc61 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-10T23:47:38Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-07de8bdd2f1645a98154ec5e6d24dc612023-01-11T00:00:15ZengIEEEIEEE Access2169-35362023-01-01111742175110.1109/ACCESS.2023.323396410005162YOLO-Extract: Improved YOLOv5 for Aircraft Object Detection in Remote Sensing ImagesZhiguo Liu0https://orcid.org/0000-0003-0280-5040Yuan Gao1https://orcid.org/0000-0001-7007-9605Qianqian Du2https://orcid.org/0000-0002-3899-2346Meng Chen3https://orcid.org/0000-0002-2611-2823Wenqiang Lv4https://orcid.org/0000-0002-6667-1347Communication and Network Key Laboratory, Dalian University, Dalian, ChinaCommunication and Network Key Laboratory, Dalian University, Dalian, ChinaCollege of Engineering Physics and Optoelectronics, Taiyuan University of Technology, Taiyuan, ChinaCommunication and Network Key Laboratory, Dalian University, Dalian, ChinaCommunication and Network Key Laboratory, Dalian University, Dalian, ChinaCompared with natural images, remote sensing targets have small and dense target shapes as well as complex target backgrounds. As a result, insufficient detection accuracy and target location cannot be accurately identified. So, this paper proposes the YOLO-extract algorithm based on the YOLOv5 algorithm. Firstly, The YOLO-extract algorithm optimized the model structure of the YOLOv5 algorithm. The YOLO-extract algorithm not only deleted the feature layer and prediction head with poor feature extraction ability but also a new feature extractor with stronger feature extraction ability was integrated into the network. At the same time, YOLO-extract borrowed the idea of residual network to integrate Coordinate Attention into the network. Secondly, the mixed dilated convolution was combined with the redesigned residual structure to enhance the feature and location information extraction ability of the shallow layer of the model and optimize the feature extraction ability of the model for different scale targets. Finally, drawing on the idea of <inline-formula> <tex-math notation="LaTeX">$\alpha $ </tex-math></inline-formula>-IoU Loss, Focal-<inline-formula> <tex-math notation="LaTeX">$\alpha $ </tex-math></inline-formula> EIoU Loss was designed to replace CIoU Loss, which makes the model bounding box regression faster and the loss lower. The experimental results on the test data set show that compared with the YOLOv5 algorithm, the YOLO-extract algorithm has a faster convergence speed, reduces the calculation amount by 45.3GFLOPs and the number of parameters by 10.526M, but increases the mAP by 8.1% and the detection speed by 3 times.https://ieeexplore.ieee.org/document/10005162/Remote sensing aircraft targetYOLOv5structure optimizationdilated convolutionfocal-α IoU loss |
| spellingShingle | Zhiguo Liu Yuan Gao Qianqian Du Meng Chen Wenqiang Lv YOLO-Extract: Improved YOLOv5 for Aircraft Object Detection in Remote Sensing Images IEEE Access Remote sensing aircraft target YOLOv5 structure optimization dilated convolution focal-α IoU loss |
| title | YOLO-Extract: Improved YOLOv5 for Aircraft Object Detection in Remote Sensing Images |
| title_full | YOLO-Extract: Improved YOLOv5 for Aircraft Object Detection in Remote Sensing Images |
| title_fullStr | YOLO-Extract: Improved YOLOv5 for Aircraft Object Detection in Remote Sensing Images |
| title_full_unstemmed | YOLO-Extract: Improved YOLOv5 for Aircraft Object Detection in Remote Sensing Images |
| title_short | YOLO-Extract: Improved YOLOv5 for Aircraft Object Detection in Remote Sensing Images |
| title_sort | yolo extract improved yolov5 for aircraft object detection in remote sensing images |
| topic | Remote sensing aircraft target YOLOv5 structure optimization dilated convolution focal-α IoU loss |
| url | https://ieeexplore.ieee.org/document/10005162/ |
| work_keys_str_mv | AT zhiguoliu yoloextractimprovedyolov5foraircraftobjectdetectioninremotesensingimages AT yuangao yoloextractimprovedyolov5foraircraftobjectdetectioninremotesensingimages AT qianqiandu yoloextractimprovedyolov5foraircraftobjectdetectioninremotesensingimages AT mengchen yoloextractimprovedyolov5foraircraftobjectdetectioninremotesensingimages AT wenqianglv yoloextractimprovedyolov5foraircraftobjectdetectioninremotesensingimages |