Detection of Spatially Distributed Fine-Grained Objects in Aerospace Images for Inverse Problem Solving

Remote sensing data processing based on deep learning methods has been widely used in several cross-domain tasks. Current approaches predominantly focus on developing deep learning methods dedicated to object detection in densely populated urban environments, which are not well suited for sparsely distributed objects, especially fine-grained ones. A representative example is maritime vessel detection on open water surfaces. The challenges in this domain include heterogeneous noise backgrounds caused by varying lighting conditions, external artifacts, wave-induced blurs, top-view perspectives that yield poor feature maps of the objects, and the need to scan large areas for object searching. To address these challenges, this paper proposes a novel hybrid deep learning approach for sparse distributed fine-grained object detection. The proposed method includes an ensemble of classifiers based on DenseNet_l and ResNet_l, combined with a YOLOv11+ detector and a semi-supervised learning strategy. Furthermore, for fine-grained object detection, an improved version of the YOLOv11+fg model is proposed, featuring adapted C3K2, UpSample, and SPFF modules. The approach was developed, pretrained, and validated using the Airbus SPOT Satellite Imagery dataset, and further trained with a subset of ShipRSImageNet for fine-grained detection. Results demonstrate the effectiveness of large-scale scanning in remote sensing and recognize sparsely distributed objects, outperforming the fine-grained detection score compared to the original model.

1. He Lh. [et al.]. Research on object detection and recognition in remote sensing images based on YOLOv11. Nature: Scientific Reports. 2025;(15):14032. DOI: 10.1038/s41598-025-96314-x

2. Saetchnikov I, Tcherniavskaia E, Skakun V. Object Detection for Unmanned Aerial Vehicle Camera via Convolutional Neural Networks. IEEE Journal on Miniaturization for Air and Space Systems.2021;2(2):98-103. DOI: 10.1109/JMASS.2020.3040976

3. Zhu XX. [et al.]. Deep Learning in Remote Sensing: A Comprehensive Review and List of Resources. IEEE Geoscience and Remote Sensing Magazine. 2017;5(4):8-36. DOI: 10.1109/MGRS.2017.2762307

4. Cheng G, Zhou P, Han J. Learning Rotation-Invariant Convolutional Neural Networks for Object Detection in VHR Optical Remote Sensing Images. IEEE Transactions on Geoscience and Remote Sensing. 2016;54(12):74057415. DOI: 10.1109/TGRS.2016.2601622

5. Ding J. [et al.]. Object Detection in Aerial Images: A Large-Scale Benchmark and Challenges. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2022;44(11):7778-7796. DOI: 10.1109/TPAMI.2021.3117983

6. Guo H, Yang X, Wang N, Gao X. A CenterNet++ model for ship detection in SAR images. Pattern Recognition. 2021;112(1):107787-107787. DOI: 10.1016/j.patcog.2020.107787

7. Gupta H. [et al.]. Ship detection using ensemble deep learning techniques from synthetic aperture radar imagery. Scientific Reports. 2024;14(1). DOI: 10.1038/s41598-024-80239-y

8. Liu J, Yang D, Hu F. Multiscale Object Detection in Remote Sensing Images Combined with Multi-Receptive-Field Features and Relation-Connected Attention. Remote Sensing. 2022;14(2):427-447. DOI: 10.3390/rs14020427

9. Fan X. [et al.]. A Small-Ship Object Detection Method for Satellite Remote Sensing Data. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 2024;(17):11886-11898. DOI: 10.1109/JSTARS.2024.3419786

10. Zhang Z, Zhang L, Wang Y, Feng P, He R. ShipRSImageNet: A Large-Scale Fine-Grained Dataset for Ship Detection in High-Resolution Optical Remote Sensing Images. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 2021;(14):84588472. DOI: 10.1109/JSTARS.2021.3104230

11. Huang G, Liu Z, Van Der Maaten L, Weinberger KQ. Densely Connected Convolutional Networks. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2017;2261-2269. arXiv:1608.06993v5

12. He K, Zhang X, Ren S, Sun J. Deep Residual Learning for Image Recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA. 2016;770-778 pp. arXiv:1512.03385

13. Redmon J, Divvala S, Girshick R, Farhadi A. You Only Look Once: Unified, Real-Time Object Detection. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA. 2016;779788 pp. DOI: 10.1109/CVPR.2016.91

14. Saetchnikov I, Skakun V, Tcherniavskaia E. Deep Neural Network-Based Dynamical Object Recognition and Robust Multiobject Tracking Technique for Onboard Unmanned Aerial Vehicle’s Computer VisionBased Systems. IEEE Journal on Miniaturization for Air and Space Systems. 2023;4(3):250-256. DOI: 10.1109/JMASS.2023.3274929