Abstract
Object detection for unmanned aerial vehicles (UAV) aerial photography presents challenges such as tiny and densely distributed objects, and unbalanced categories. Furthermore, the hardware limitations of UAV restrict the scalability of models, leading to reduced accuracy. In response to these challenges, an enhanced YOLOv8m model which incorporates multiple lightweight strategies is proposed. Specifically, GDC (Ghost Dynamic Conv) is introduced into the backbone network to improve feature extraction, and more features are generated with fewer parameters to achieve efficient feature extraction. Additionally, the feature fusion mechanism has been optimized, and the LS-FPN-PAN feature fusion mechanism has been devised to globally reduce the number of feature channels and amount of calculation. Through adaptive feature selection, the channel weight was given to achieve better fusion. Furthermore, a lightweight selective detection head was proposed, and shared convolution was employed to facilitate the learning of target features by three detection heads. The WMPDIoU loss function was designed to reduce the penalty caused by the geometric factors of the detection box of tiny objects. The cost-free approach of substituting NMS function and implementing knowledge distillation is employed to enhance the model’s performance. The experimental results show that the model size and parameter number of the improved model are only 42.1\(\%\) and 55.1\(\%\) of the original model, but the performance is considerably improved. On the Visdrone2019 test dataset, P, mAP@0.5, mAP@0.5:0.95 are increased by 12.9\(\%\), 26.5\(\%\) and 38.8\(\%\) respectively, indicating a successful realization of lightweight design with enhanced performance capabilities suitable for effective application in object detection tasks on UAV platforms.
Supported by the Ministry of Education (20230104440, 2023122991102), and Priority Academic Program Development of Jiangsu Higher Education Institutions.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Li, Y., Fan, Q., Huang, H., Han, Z., Gu, Q.: A modified YOLOv8 detection network for UAV aerial image recognition. Drones 7(5), 304 (2023)
Lou, H., Duan, X., Guo, J., Liu, H., Gu, J., Bi, L., Chen, H.: DC-YOLOv8: small-size object detection algorithm based on camera sensor. Electronics 12(10), 2323 (2023)
Guo, J., Lou, H., Chen, H., Liu, H., Gu, J., Bi, L., Duan, X.: A new detection algorithm for alien intrusion on highway. Sci. Rep. 13(1), 10667 (2023)
Wang, G., Chen, Y., An, P., Hong, H., Hu, J., Huang, T.: UAV-YOLOv8: a small-object-detection model based on improved YOLOv8 for UAV aerial photography scenarios. Sensors 23(16), 7190 (2023)
Han, K., Wang, Y., Guo, J., Wu, E.: ParameterNet: parameters are all you need for large-scale visual pretraining of mobile networks (2023). arXiv:2306.14525
Han, K., Wang, Y., Tian, Q., Guo, J., Xu, C., Xu, C.: Ghostnet: more features from cheap operations. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 1580–1589(2020)
Chen, Y., Zhang, C., Chen, B., Huang, Y., Sun, Y., Wang, C., Fu, X., Dai, Y., Qin, F., Peng, Y., et al.: Accurate leukocyte detection based on deformable-DETR and multi-level feature fusion for aiding diagnosis of blood diseases. Comput. Biol. Med. 170, 107917 (2024)
Liu, W., Lu, H., Fu, H., Cao, Z.: Learning to Upsample by Learning to Sample. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 6027–6037 (2023)
Tian, Z., Shen, C., Chen, H., He, T.: FCOS: a simple and strong anchor-free object detector. IEEE Trans. Pattern Anal. Mach. Intell. 44(4), 1922–1933 (2020)
Tong, Z., Chen, Y., Xu, Z., Yu, R.: Wise-IoU: bounding box regression loss with dynamic focusing mechanism (2023). arXiv:2301.10051
Siliang, M., Yong, X.: MPDIoU: a loss for efficient and accurate bounding box regression (2023). arXiv:2307.07662
Bodla, N., Singh, B., Chellappa, R., Davis, L. S.: Soft-NMS–improving object detection with one line of code. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 5561–5569 (2017)
Shu, C., Liu, Y., Gao, J., Yan, Z., Shen, C.: Channel-wise knowledge distillation for dense prediction. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 5311–5320 (2021)
Du, D., Zhu, P., Wen, L., Bian, X., Lin, H., Hu, Q., Zhang, L.: VisDrone-DET2019: the vision meets drone object detection in image challenge results. In: Proceedings of the IEEE/CVF International Conference on Computer Vision Workshops (2019)
Zhang, Z., Yi, H.-H., Zheng, J.: Focusing on small objects detector in aerial images. Acta Electonica Sinica 51(4), 944–955 (2023)
Hsieh, M.R., Lin, Y.L., Hsu, W.H.: Drone-based object counting by spatially regularized regional proposal network. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 4145–4153 (2017)
Wu, X., Li, W., Hong, D., Tao, R., Du, Q.: Deep learning for unmanned aerial vehicle-based object detection and tracking: a survey. IEEE Geosci. Remote Sens. Mag. 10(1), 91–124 (2021)
Feng, C., Zhong, Y., Gao, Y., Scott, M. R., Huang, W.: Tood: task-aligned one-stage object detection. In: 2021 IEEE/CVF International Conference on Computer Vision (ICCV), pp. 3490–3499. IEEE Computer Society (2021)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2025 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Pan, W., Yang, Z. (2025). LS-YOLO: A Lightweight Selective YOLOv8 Algorithm for UAV Aerial Photography. In: Lin, Z., et al. Pattern Recognition and Computer Vision. PRCV 2024. Lecture Notes in Computer Science, vol 15042. Springer, Singapore. https://doi.org/10.1007/978-981-97-8858-3_13
Download citation
DOI: https://doi.org/10.1007/978-981-97-8858-3_13
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-97-8857-6
Online ISBN: 978-981-97-8858-3
eBook Packages: Computer ScienceComputer Science (R0)