Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
38 tokens/sec
GPT-4o
59 tokens/sec
Gemini 2.5 Pro Pro
41 tokens/sec
o3 Pro
7 tokens/sec
GPT-4.1 Pro
50 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

FlightScope: An Experimental Comparative Review of Aircraft Detection Algorithms in Satellite Imagery (2404.02877v4)

Published 3 Apr 2024 in cs.CV and cs.AI

Abstract: Object detection in remotely sensed satellite pictures is fundamental in many fields such as biophysical, and environmental monitoring. While deep learning algorithms are constantly evolving, they have been mostly implemented and tested on popular ground-based taken photos. This paper critically evaluates and compares a suite of advanced object detection algorithms customized for the task of identifying aircraft within satellite imagery. Using the large HRPlanesV2 dataset, together with a rigorous validation with the GDIT dataset, this research encompasses an array of methodologies including YOLO versions 5 and 8, Faster RCNN, CenterNet, RetinaNet, RTMDet, and DETR, all trained from scratch. This exhaustive training and validation study reveal YOLOv5 as the preeminent model for the specific case of identifying airplanes from remote sensing data, showcasing high precision and adaptability across diverse imaging conditions. This research highlight the nuanced performance landscapes of these algorithms, with YOLOv5 emerging as a robust solution for aerial object detection, underlining its importance through superior mean average precision, Recall, and Intersection over Union scores. The findings described here underscore the fundamental role of algorithm selection aligned with the specific demands of satellite imagery analysis and extend a comprehensive framework to evaluate model efficacy. The benchmark toolkit and codes, available via https://github.com/toelt-llc/FlightScope_Bench, aims to further exploration and innovation in the realm of remote sensing object detection, paving the way for improved analytical methodologies in satellite imagery applications.

PDF HTML Abstract
Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize File Document Download Save Streamline Icon: https://streamlinehq.com PDF File Document Download Save Streamline Icon: https://streamlinehq.com Markdown Bookmark Chat Bubble Oval Streamline Icon: https://streamlinehq.com Chat (Pro)
Definition Search Book Streamline Icon: https://streamlinehq.com
References (94)
  1. A comparative study of rgb and multispectral sensor-based cotton canopy cover modelling using multi-temporal uas data. Remote Sensing, 11(23):2757, 2019.
  2. Anchor-free object detection in remote sensing images using a variable receptive field network. EURASIP Journal on Advances in Signal Processing, 2023(1):53, 2023.
  3. Sushobhan Majumdar. The Role of Remote Sensing and GIS in Military Strategy to Prevent Terror Attacks, pages 79–94. 2021.
  4. Remote sensing and urban analysis. In Osvaldo Gervasi, Beniamino Murgante, Antonio Laganà, David Taniar, Youngsong Mun, and Marina L. Gavrilova, editors, Computational Science and Its Applications – ICCSA 2008, pages 300–315, Berlin, Heidelberg, 2008. Springer Berlin Heidelberg.
  5. Harm Greidanus. Satellite imaging for maritime surveillance of the european seas. In Remote Sensing of the European Seas, pages 343–358. Springer, 2008.
  6. Aircraft detection in high-resolution satellite images using deep learning. International Journal of Remote Sensing, 43(10):4349–4367, 2022.
  7. Sar-dad: A novel small target detection algorithm for spaceborne sar imagery. Complexity, 2022:2262549, 2022.
  8. Factors affecting the quality of formation and resolution of images in remote sensing systems. JCPEE, 5(1):41–46, 2015.
  9. Comparative research on deep learning approaches for airplane detection from very high-resolution satellite images. Remote Sensing, 12(3), 2020.
  10. Ultralytics yolov8, 2023.
  11. SSD: Single Shot MultiBox Detector. In Bastian Leibe, Jiri Matas, Nicu Sebe, and Max Welling, editors, Computer Vision – ECCV 2016, Lecture Notes in Computer Science, pages 21–37, Cham, 2016. Springer International Publishing.
  12. Region-Based Convolutional Networks for Accurate Object Detection and Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 38(1):142–158, January 2016. Conference Name: IEEE Transactions on Pattern Analysis and Machine Intelligence.
  13. Ross Girshick. Fast R-CNN. In 2015 IEEE International Conference on Computer Vision (ICCV), pages 1440–1448, December 2015. ISSN: 2380-7504.
  14. Faster r-cnn: Towards real-time object detection with region proposal networks, 2016.
  15. Mask R-CNN. In 2017 IEEE International Conference on Computer Vision (ICCV), pages 2980–2988, October 2017. ISSN: 2380-7504.
  16. EfficientDet: Scalable and Efficient Object Detection. pages 10778–10787, 2020.
  17. Focal Loss for Dense Object Detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 42(2):318–327, February 2020. Conference Name: IEEE Transactions on Pattern Analysis and Machine Intelligence.
  18. CenterNet: Keypoint Triplets for Object Detection, April 2019. arXiv:1904.08189 [cs].
  19. Attention mechanisms in computer vision: A survey. Computational visual media, 8(3):331–368, 2022.
  20. Few-shot image classification algorithm based on attention mechanism and weight fusion. Journal of Engineering and Applied Science, 70(1):14, 2023.
  21. Object detection based on an adaptive attention mechanism. Scientific Reports, 10(1):11307, 2020.
  22. YOLOv8 by Ultralytics. Software.
  23. Rtmdet: An empirical study of designing real-time object detectors, 2022.
  24. Detrs beat yolos on real-time object detection. arXiv preprint arXiv:2304.08069, 2023.
  25. Cornernet: Detecting objects as paired keypoints. CoRR, abs/1808.01244, 2018.
  26. Focal loss for dense object detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 42(2):318–327, 2020.
  27. Satellite Earth Observations in Environmental Problem-Solving, pages 3–27. Springer Singapore, Singapore, 2017.
  28. Image and object geo-localization. International Journal of Computer Vision, pages 1–43, 2023.
  29. Object detection in agricultural contexts: A multiple resolution benchmark and comparison to human. Computers and Electronics in Agriculture, 189:106404, 2021.
  30. Stp-som: Scale-transfer learning for pansharpening via estimating spectral observation model. International Journal of Computer Vision, 131(12):3226–3251, 2023.
  31. Vehicle detection from aerial imagery using principal component analysis and deep learning. In International Conference on Innovations in Bio-Inspired Computing and Applications, pages 129–140. Springer, 2022.
  32. Dota: A large-scale dataset for object detection in aerial images. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 3974–3983, 2018.
  33. Yolov3: An incremental improvement, 2018. cite arxiv:1804.02767Comment: Tech Report.
  34. A benchmark dataset for deep learning-based airplane detection: Hrplanes. International Journal of Engineering and Geosciences, 8(3):212–223, 2023.
  35. Yolov4: Optimal speed and accuracy of object detection. arXiv preprint arXiv:2004.10934, 2020.
  36. Class. Airbus aircraft detection dataset. https://universe.roboflow.com/class-dvpyb/airbus-aircraft-detection, jan 2023. visited on 2024-01-21.
  37. Rareplanes: Synthetic data takes flight. CoRR, abs/2006.02963, 2020.
  38. GDIT. Aerial airport dataset. https://universe.roboflow.com/gdit/aerial-airport, sep 2023. visited on 2024-01-21.
  39. Robert Hammell. Planes in satellite imagery, 2023.
  40. A dataset for detecting flying airplanes on satellite images, 2021.
  41. Dataset of aircraft classification in remote sensing images. J. Glob. Change Data Discov, 2:183–190, 2020.
  42. C. Bhagya and A. Shyna. An overview of deep learning based object detection techniques. In 2019 1st International Conference on Innovations in Information and Communication Technology (ICIICT), pages 1–6, 2019.
  43. Deep learning-based object detection in low-altitude uav datasets: A survey. Image and Vision Computing, 104:104046, 2020.
  44. Attention is all you need. Advances in neural information processing systems, 30, 2017.
  45. Deep learning for generic object detection: A survey. International journal of computer vision, 128:261–318, 2020.
  46. Oriented r-cnn and beyond. International Journal of Computer Vision, pages 1–23, 2024.
  47. A study on transformer-based object detection. In 2021 International Conference on Intelligent Technologies (CONIT), pages 1–6, 2021.
  48. Single-shot refinement neural network for object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 4203–4212, 2018.
  49. Cascade r-cnn: Delving into high quality object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 6154–6162, 2018.
  50. R-fcn: Object detection via region-based fully convolutional networks. Advances in neural information processing systems, 29, 2016.
  51. R-cnn for small object detection. In Computer Vision–ACCV 2016: 13th Asian Conference on Computer Vision, Taipei, Taiwan, November 20-24, 2016, Revised Selected Papers, Part V 13, pages 214–230. Springer, 2017.
  52. End-to-end object detection with transformers, 2020.
  53. Exploring plain vision transformer backbones for object detection. In European Conference on Computer Vision, pages 280–296. Springer, 2022.
  54. MimicDet: Bridging the Gap Between One-Stage and Two-Stage Object Detection. pages 541–557. November 2020.
  55. Jeremy Jordan. An overview of object detection: One-stage methods. https://www.jeremyjordan.me/object-detection-one-stage/, 2021. Accessed on 8 June 2023.
  56. Review on one-stage object detection based on deep learning. EAI Endorsed Transactions on e-Learning, 7(23), 6 2022.
  57. A survey of deep learning-based object detection. IEEE Access, 7:128837–128868, 2019.
  58. Bibliometric analysis of one-stage and two-stage object detection. 02 2021.
  59. You Only Look Once: Unified, Real-Time Object Detection. pages 779–788, June 2016. ISSN: 1063-6919.
  60. Mimicdet: Bridging the gap between one-stage and two-stage object detection. CoRR, abs/2009.11528, 2020.
  61. Temporal roi align for video object recognition. CoRR, abs/2109.03495, 2021.
  62. Object detection using yolo: Challenges, architectural successors, datasets and applications. multimedia Tools and Applications, 82(6):9243–9275, 2023.
  63. Yolo9000: Better, faster, stronger. In 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 6517–6525, 2017.
  64. ultralytics/yolov5: v7.0 - YOLOv5 SOTA Realtime Instance Segmentation. Zenodo, November 2022.
  65. Scaled-yolov4: Scaling cross stage partial network. pages 13024–13033, 06 2021.
  66. Super-gradients, 2021.
  67. Neural architecture search benchmarks: Insights and survey. IEEE Access, PP:1–1, 01 2023.
  68. Ssd: Single shot multibox detector. In Bastian Leibe, Jiri Matas, Nicu Sebe, and Max Welling, editors, Computer Vision – ECCV 2016, pages 21–37, Cham, 2016. Springer International Publishing.
  69. Evaluating the single-shot multibox detector and yolo deep learning models for the detection of tomatoes in a greenhouse. Sensors, 21(10), 2021.
  70. Deep pre-trained models for computer vision applications: Traffic sign recognition. In 2021 18th International Multi-Conference on Systems, Signals & Devices (SSD), pages 23–28, 2021.
  71. Multi-task learning using uncertainty to weigh losses for scene geometry and semantics, 2018.
  72. Loss functions and metrics in deep learning, 2023.
  73. Enhancement of SSD by concatenating feature maps for object detection. CoRR, abs/1705.09587, 2017.
  74. Object detection in real time based on improved single shot multi-box detector algorithm. EURASIP Journal on Wireless Communications and Networking, 2020(1):204, October 2020.
  75. Rich feature hierarchies for accurate object detection and semantic segmentation. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 11 2013.
  76. Convolutional neural networks: an overview and application in radiology. Insights into Imaging, 9, 06 2018.
  77. Class-agnostic object detection. 2021 IEEE Winter Conference on Applications of Computer Vision (WACV), pages 918–927, 2020.
  78. Selective search for object recognition. International Journal of Computer Vision, 104:154–171, 09 2013.
  79. Imagenet classification with deep convolutional neural networks. 25, 2012.
  80. Support vector machines. IEEE Intelligent Systems and their Applications, 13(4):18–28, 1998.
  81. Object detection in 20 years: A survey, 2023.
  82. A brief review and challenges of object detection in optical remote sensing imagery. Multiagent and Grid Systems, 16:227–243, 10 2020.
  83. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 770–778, 2016.
  84. A survey of transformers, 2021.
  85. Object detection with deep learning: A review, 2019.
  86. Backbones-review: Feature extraction networks for deep learning and deep reinforcement learning approaches, 2022.
  87. A survey of transformers. AI Open, 3:111–132, 2022.
  88. A deep learning review of resnet architecture for lung disease identification in cxr image. Applied Sciences, 13(24), 2023.
  89. A review on 2d instance segmentation based on deep neural networks. Image and Vision Computing, 120:104401, 2022.
  90. Mmdetection: Open mmlab detection toolbox and benchmark, 2019.
  91. Scaling up your kernels to 31x31: Revisiting large kernel design in cnns, 2022.
  92. Dynamic label assignment for object detection by combining predicted ious and anchor ious. Journal of Imaging, 8(7):193, July 2022.
  93. Detectron2. https://github.com/facebookresearch/detectron2, 2019.
  94. A comparative analysis of object detection metrics with a companion open-source toolkit. Electronics, 10(3):279, 2021.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (6)
  1. Safouane El Ghazouali (5 papers)
  2. Arnaud Gucciardi (5 papers)
  3. Nicola Venturi (1 paper)
  4. Michael Rueegsegger (1 paper)
  5. Umberto Michelucci (17 papers)
  6. Francesca Venturini (17 papers)
Citations (1)
View on Semantic Scholar