Research on Oriented Object Detection in Aerial Images Based on Architecture Search with Decoupled Detection Heads
Source abstract
Abstract
Object detection in aerial images can provide great support in traffic planning, national defense reconnaissance, hydrographic surveys, infrastructure construction, and other fields. Objects in aerial images are characterized by small pixel–area ratios, dense arrangements between objects, and arbitrary inclination angles. In response to these characteristics and problems, we improved the feature extraction network Inception-ResNet using the Fast Architecture Search (FAS) module and proposed a one-stage anchor-free rotation object detector. The structure of the object detector is simple and only consists of convolution layers, which reduces the number of model parameters. At the same time, the label sampling strategy in the training process is optimized to resolve the problem of insufficient sampling. Finally, a decoupled object detection head is used to separate the bounding box regression task from the object classification task. The experimental results show that the proposed method achieves mean average precision (mAP) of 82.6%, 79.5%, and 89.1% on the DOTA1.0, DOTA1.5, and HRSC2016 datasets, respectively, and the detection speed reaches 24.4 FPS, which can meet the needs of real-time detection.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Robotics and Sensor-Based Localization · Advanced Image and Video Retrieval Techniques · Advanced Neural Network Applications
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Computer Science and Technology, Zhejiang Sci-Tech University, Hangzhou 310018, China
Hangzhou Neptune Technology Co., Ltd., Zhejiang Sci-Tech University, Room 223-155, Building 11, Qiantang New District, Hangzhou 310018, China
References (39)
Show source-provided references
1. Redmon, J., Divvala, S., Girshick, R., and Farhadi, A. (2015). You Only Look Once: Unified, Real-Time Object Detection. arXiv.10.1109/CVPR.2016.91 ↗
2. Ren, S., He, K., Girshick, R., and Sun, J. (2015). Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. arXiv.
3. Object detection in aerial images: A large-scale benchmark and challenges10.1109/TPAMI.2021.3117983 ↗
4. A high resolution optical satellite image dataset for ship recognition and some new baselines10.5220/0006120603240331 ↗
5. Ge, Z., Liu, S., Wang, F., Li, Z., and Sun, J. (2021). YOLOX: Exceeding YOLO Series in 2021. arXiv.
6. Girshick, R., Donahue, J., Darrell, T., and Malik, J. (2014, January 23–28). Rich feature hierarchies for accurate object detection and semantic segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA.10.1109/CVPR.2014.81 ↗
7. Girshick, R. (2015, January 7–13). Fast R-CNN. Proceedings of the 2015 IEEE International Conference on Computer Vision (ICCV), ICCV ’15, Santiago, Chile.10.1109/ICCV.2015.169 ↗
8. Zhang, L., Lin, L., Liang, X., and He, K. (2016, January 11–14). Is faster R-CNN doing well for pedestrian detection?. Proceedings of the Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands. Proceedings, Part II 14.
9. Cheng, B., Wei, Y., Shi, H., Feris, R., Xiong, J., and Huang, T. (2018, January 8–14). Revisiting rcnn: On awakening the classification power of faster rcnn. Proceedings of the European Conference on Computer Vision (ECCV), Munich, Germany.10.1007/978-3-030-01267-0_28 ↗
10. Lin, T.Y., Goyal, P., Girshick, R., He, K., and Dollár, P. (2017, January 22–29). Focal loss for dense object detection. Proceedings of the IEEE International Conference on Computer Vision, Venice, Italy.10.1109/ICCV.2017.324 ↗
11. Tian, Z., Shen, C., Chen, H., and He, T. (November, January 27). Fcos: Fully convolutional one-stage object detection. Proceedings of the IEEE/CVF International Conference on Computer Vision, Seoul, Republic of Korea.
12. Anchor-based fast spectral ensemble clustering10.1016/j.inffus.2024.102587 ↗
13. Chu, Q., Li, S., Chen, G., Li, K., and Li, X. (2023, January 7–14). Adversarial alignment for source free object detection. Proceedings of the AAAI Conference on Artificial Intelligence, Washington, DC, USA.
14. Xu, X., Jiang, Y., Chen, W., Huang, Y., Zhang, Y., and Sun, X. (2022). Damo-yolo: A report on real-time object detection design. arXiv.
15. Building NAS: Automatic designation of efficient neural architectures for building extraction in high-resolution aerial images10.1016/j.imavis.2020.104025 ↗
16. Aharon, S., Masad, O., Yurkova, K., Fridman, L., Khvedchenya, E., Rubin, R., Bagrov, N., and Tymchenko, B. (2025, July 26). Super-Gradients. Available online: https://zenodo.org/records/7789328.
17. Pan, M., Xia, W., Yu, H., Hu, X., Cai, W., and Shi, J. (2023). Vehicle Detection in UAV Images via Background Suppression Pyramid Network and Multi-Scale Task Adaptive Decoupled Head. Remote Sens., 15.10.3390/rs15245698 ↗
18. YOLO-SDH: Improved YOLOv5 using scaled decoupled head for object detection10.1007/s13042-024-02357-3 ↗
19. A Decoupled Head and Multiscale Coordinate Convolution Detection Method for Ship Targets in Optical Remote Sensing Images10.1109/ACCESS.2024.3394052 ↗
20. Efficient GPU-accelerated multisource global fit pipeline for LISA data analysis10.1103/PhysRevD.111.024060 ↗
21. Residual spatial fusion network for RGB-thermal semantic segmentation10.1016/j.neucom.2024.127913 ↗
22. Single Encoder and Decoder-Based Transformer Fusion with Deep Residual Attention for Restoration of Degraded Images and Clear Visualization in Adverse Weather Conditions10.1007/s13369-023-08342-2 ↗
23. Savaştaer, E.F., Çelik, B., and Çelik, M.E. (2025). Automatic detection of developmental stages of molar teeth with deep learning. BMC Oral Health, 25.10.1186/s12903-025-05827-4 ↗
24. YOLO-MS: Rethinking multi-scale representation learning for real-time object detection10.1109/TPAMI.2025.3538473 ↗
25. Liu, Y., Liu, Y., Guo, X., Ling, X., and Geng, Q. (2025). Metal surface defect detection using SLF-YOLO enhanced YOLOv8 model. Sci. Rep., 15.10.1038/s41598-025-94936-9 ↗
26. Choi, J.Y., and Han, J.M. (2024). Deep learning (Fast R-CNN)-based evaluation of rail surface defects. Appl. Sci., 14.10.3390/app14051874 ↗
27. Chaudhuri, A. (2024). Smart traffic management of vehicles using faster R-CNN based deep learning method. Sci. Rep., 14.10.1038/s41598-024-60596-4 ↗
28. Fault diagnosis using variational autoencoder GAN and focal loss CNN under unbalanced data10.1177/14759217241254121 ↗
29. Zhang, G., Yu, W., and Hou, R. (2024). Mfil-fcos: A multi-scale fusion and interactive learning method for 2d object detection and remote sensing image detection. Remote Sens., 16.10.3390/rs16060936 ↗
30. Generalized out-of-distribution detection: A survey10.1007/s11263-024-02117-4 ↗
31. Yolc: You only look clusters for tiny object detection in aerial images10.1109/TITS.2024.3386928 ↗
32. Coarse to fine-based image–point cloud fusion network for 3D object detection10.1016/j.inffus.2024.102551 ↗
33. KLDet: Detecting tiny objects in remote sensing images via kullback-leibler divergence
34. Yu, W., Zhou, P., Yan, S., and Wang, X. (2024, January 16–22). Inceptionnext: When inception meets convnext. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.10.1109/CVPR52733.2024.00542 ↗
35. Bal-R 2 CNN: High quality recurrent object detection with balance optimization10.1109/TMM.2021.3067439 ↗
36. Han, J., Ding, J., Xue, N., and Xia, G.S. (2021, January 20–25). Redet: A rotation-equivariant detector for aerial object detection. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Nashville, TN, USA.10.1109/CVPR46437.2021.00281 ↗
37. Yang, X., Yan, J., Feng, Z., and He, T. (2021, January 2–9). R3det: Refined single-stage detector with feature refinement for rotating object. Proceedings of the AAAI Conference on Artificial Intelligence, Virtual.10.1609/aaai.v35i4.16426 ↗
38. Guo, J., Hao, J., Mou, L., Hao, H., Zhang, J., and Zhao, Y. (2025). S2A-Net: Retinal structure segmentation in OCTA images through a spatially self-aware multitask network. Biomed. Signal Process. Control, 110.10.1016/j.bspc.2025.108003 ↗
39. Das, A., Singh, A., and Prakash, S. (2024). CapsuleNet: A Deep Learning Model To Classify GI Diseases Using EfficientNet-b7. arXiv.