Abstract
Due to the large computational requirements of object detection algorithms, high-resolution remote sensing vehicle detection always involves numerous small objects, high level of background complexity, and challenges in balancing model accuracy and parameter count. The attention mechanism and efficient architecture lightweight-YOLO (AMEA-YOLO) is proposed in this paper. A lightweight network as the backbone network of AMEA-YOLO is designed, and it could maintain model accuracy and ensure good lightweight. FasterNet is employed to accelerate model training speed. The enhanced deep second-order channel attention module (EnhancedSOCA) is utilized to improve the image high-resolution. In addition, a lightweight module is devised to further reduce the model’s weight. The implementation of the HardSwish activation function improves model accuracy. The experimental results indicate that the AMEA-YOLO algorithm could ensure model lightweight and accurate performance.
Similar content being viewed by others
Data availability
In this work, the open remote sensing datasets that support the findings of this study are available from the GitHub. VisDrone dataset is available at https://github.com/VisDrone/VisDrone-Dataset. VEDAI dataset is available at https://github.com/nikitalpopov/vedai/tree/master.
Abbreviations
- c :
-
The number of filter
- Params:
-
Parameters
- mAP:
-
Mean average precision
- \(|t|_{\textrm{odd}}\) :
-
The nearest odd number to t
- \(C_{\textrm{i}}\) :
-
The input channels
- \(C_{\textrm{o}}\) :
-
The output channels
- H/W:
-
The size of the output feature map
- k :
-
The size of the one-dimensional convolutional kernel
- GFLOPs:
-
Giga floating-point operations per second
References
Peng G, Yang Z, Wang S, Zhou Y (2023) AMFLW-YOLO: a lightweight network for remote sensing image detection based on attention mechanism and multiscale feature fusion. IEEE Trans Geosci Remote Sens 61:1–16
Redmon J, Divvala S, Girshick R, Farhadi A (2016) You only look once: unified. Real-time object detection. Computer Vision & Pattern Recognition, IEEE
Liu W, Anguelov D, Erhan D, Szegedy C, Reed S, Fu C-Y, Berg AC (2016) SSD: single shot multibox detector. In: Computer Vision - ECCV 2016,Lecture Notes in Computer Science, pp 21-37
Chen S, Xu J, Yu J, Wu J, Zhou G (2022) Automatic abdominal hernia mesh detection based on YOLOM. IEEE Access 10:31420–31431
Wu D, Liao M, Zhang W, Wang X, Bai X, Cheng W, Liu W (2021) YOLOP: you only look once for panoptic driving perception. Mach Intell Res 19:550–562
Yue Y, Liu H, Lin C, Meng X, Liu C, Zhang X, Cui J, Du Y (2023) Automatic recognition of defects behind railway tunnel linings in GPR images using transfer learning. Measurement 224:113903
Jiayuan. Wang GS, Q M Jonathan Wu, SM, Ning Zhang SM, Jonathan QM, Wang W (2023) You only look at once for real-time and generic multi-task. ArXiv, abs/2310.01641
Song C, Zhang F, Li J, Xie J, Yang C, Zhou H, Zhang J (2022) Detection of maize tassels for UAV remote sensing image with an improved YOLOX model. J Integr Agric 22(6):1671–1683
Chen N, Li Y, Yang Z, Lu Z, Wang S, Wang J (2023) LODNU: lightweight object detection network in UAV vision. J Supercomput 79:10117–10138
Qiu Z, Bai H, Chen T (2023) Special vehicle detection from UAV perspective via YOLO-GNS based deep learning network. Drones 7:117
Wu H, Hua Y, Zou H, Ke G (2022) A lightweight network for vehicle detection based on embedded system. J Supercomput 78:18209–18224
Wang Y, Hua C, Ding W, Wu R (2022) Real-time detection of flame and smoke using an improved YOLOv4 network. SIViP 16:1109–1116
Zhao L, Zhi L, Zhao C, Zheng W (2022) Fire-YOLO: a small target object detection method for fire inspection. Sustainability 14:4930
Kong W, Liu S, Xu M, Yasir M, Wang D, Liu W (2023) Lightweight algorithm for multi-scale ship detection based on high-resolution SAR images. Int J Remote Sens 44:1390–1415
Han Y, Wang F, Wang W, Li X, Zhang J (2023) YOLO-SG: small traffic signs detection method in complex scene. J Supercomput 1–22. https://doi.org/10.1007/s11227-023-05547-y
Liu W, Quijano K, Crawford MM (2022) YOLOv5-tassel: detecting tassels in RGB UAV imagery with improved YOLOv5 based on transfer learning. IEEE J Sel Top Appl Earth Observ Remote Sens 15:8085–8094
Xu D, Wu Y (2020) MRFF-YOLO: a multi-receptive fields fusion network for remote sensing target detection. Remote Sens 12:3118
Chen Y, Zhao D, Er MJ, Zhuang Y, Hu H (2021) A novel vehicle tracking and speed estimation with varying UAV altitude and video resolution. Int J Remote Sens 42:4437–4462
Wei Y, Tian Q, Guo JH, Huang W, Cao J (2018) Multi-vehicle detection algorithm through combining Harr and HOG features. Math Comput Simul 155:130–145
Wang Z, Zhan J, Duan CG, Guan X, Yang K (2021) Vehicle detection in severe weather based on pseudo-visual search and HOG-LBP feature fusion. Proc Inst Mech Eng Part D J Automob Eng 236:1607–1618
Ji Y, Lee H (2022) Event-based anomaly detection using a one-class SVM for a hybrid electric vehicle. IEEE Trans Veh Technol 71:6032–6043
Kosaka N, Ohashi G (2015) Vision-based nighttime vehicle detection using CenSurE and SVM. IEEE Trans Intell Transp Syst 16:2599–2608
Yaghoobi Ershadi N, Menéndez JM, Jiménez D (2018) Robust vehicle detection in different weather conditions: using MIPM. PLoS One 13:e0191355
Zhou H, Ma A, Niu Y, Ma Z (2022) Small-object detection for UAV-based images using a distance metric method. Drones 6:308
Li Q, Mou L, Xu Q, Zhang Y, Zhu X (2018) R3-Net: a deep network for multioriented vehicle detection in aerial images and videos. IEEE Trans Geosci Remote Sens 57:5028–5042
Wan Y, Zhong Y, Huang Y, Han Y, Cui Y, Yang Q, Li Z, Yuan Z, Li Q (2022) ARSD: an adaptive region selection object detection framework for UAV images. Drones 6:228
Zaghari N, Fathy M, Jameii SM, Shahverdy M (2021) The improvement in obstacle detection in autonomous vehicles using YOLO non-maximum suppression fuzzy algorithm. J Supercomput 77:13421–13446
Wang Q, Wu B, Zhu PF, Li P, Zuo W, Hu Q (2019) ECA-Net: efficient channel attention for deep convolutional neural networks. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2020, pp 11531–11539
Wang SM, Wang X (2023) ES-YOLO: a new lightweight fire detection model. In: International Conference on Computer Vision and Data Mining
Guo H, Bai H, Zhou Y, Li W (2020) DF-SSD: a deep convolutional neural network-based embedded lightweight object detection framework for remote sensing imagery. J Appl Remote Sens 14:014521–014521
Zhou G, Yu J, Zhou S (2023) LSCB: a lightweight feature extraction block for SAR automatic target recognition and detection. Int J Remote Sens 44:2548–2572
Zhang X, Peng L, Lu XY (2022) Vehicle fusion detection in visible and infrared thermal images via spare network and dynamic weight coefficient-based Dempster–Shafer evidence theory. J Appl Remote Sens 16:036519–036519
Yin R, Zhao W, Fan X, Yin Y (2020) AF-SSD: an accurate and fast single shot detector for high spatial remote sensing imagery. Sensors (Basel, Switzerland) 20:6530
Han K, Wang Y, Tian Q, Guo J, Xu C, Xu C (2019) GhostNet: more features from cheap operations. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2020, pp 1577–1586
Yang L, Zhang R, Li L, Xie X (2021) SimAM: a simple, parameter-free attention, module for convolutional neural networks. In: International Conference on Machine Learning
Chen J, Kao S, He H, Zhuo W, Wen S, Lee C, Chan SG (2023) Run, Don’t Walk: chasing higher FLOPS for faster neural networks. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2023, pp 12021–12031
Dai T, Cai J, Zhang Y, Xia S, Zhang L (2019) Second-order attention network for single image super-resolution. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2019, pp 11057–11066
Howard AG, Sandler M, Chu G, Chen L, Chen B, Tan M, Wang W, Zhu Y, Pang R, Vasudevan V, Le QV, Adam H (2019) Searching for MobileNetV3. In: IEEE/CVF International Conference on Computer Vision (ICCV) 2019, pp 1314–1324
Razakarivony S, Jurie F (2016) Vehicle detection in aerial imagery : a small target detection benchmark. J Vis Commun Image Represent 34:187–203
Ren S, He K, Girshick RB, Sun J (2015) Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Trans Pattern Anal Mach Intell 39:1137–1149
Yao F, Wang S, Ding L, Zhong G, Bullock LB, Xu Z, Dong J (2022) Lightweight network learning with zero-shot neural architecture search for UAV images. Knowl Based Syst 260:110142
Bochkovskiy A, Wang C, Liao HM (2020) YOLOv4: optimal speed and accuracy of object detection. ArXiv, abs/2004.10934
Jocher GR, Stoken A, Borovec J, NanoCode Chaurasia A, TaoXie Changyu L, Abhiram Laughing tkianai yxNONG Hogan A, lorenzomammana AlexWang Hájek J, Diaconu L, Marc Kwon Y, Oleg wanghaoyang Defretin Y, Lohia A, ah M, Milanko B, Fineran B, Khromov DP, Yiwei D, Doug Durgesh, Ingham F (2021) ultralytics/yolov5: v5.0 - YOLOv5-P6 1280 models, AWS, Supervise.ly and YouTube integrations
Ge Z, Liu S, Wang F, Li Z, Sun J (2021) YOLOX: exceeding YOLO series in 2021. ArXiv, arXiv:abs/2107.08430
Wang C, Bochkovskiy A, Liao HM (2022) YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time Object detectors. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2023, pp 7464–7475
Reis D, Kupec J, Hong J, Daoudi A (2023) Real-time flying object detection with YOLOv8. ArXiv, arXiv:abs/2305.09972
Acknowledgments
This work is supported by Tianjin Enterprise Science and Technology Commissioner Project (NO.20YDTPJC00170).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
No potential conflict of interest was reported by the authors.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Wang, SB., Gao, ZM., Jin, DH. et al. AMEA-YOLO: a lightweight remote sensing vehicle detection algorithm based on attention mechanism and efficient architecture. J Supercomput 80, 11241–11260 (2024). https://doi.org/10.1007/s11227-023-05872-2
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11227-023-05872-2