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

Revealing Hierarchical Structure of Leaf Venations in Plant Science via Label-Efficient Segmentation: Dataset and Method (2405.10041v1)

Published 16 May 2024 in cs.CV

Abstract: Hierarchical leaf vein segmentation is a crucial but under-explored task in agricultural sciences, where analysis of the hierarchical structure of plant leaf venation can contribute to plant breeding. While current segmentation techniques rely on data-driven models, there is no publicly available dataset specifically designed for hierarchical leaf vein segmentation. To address this gap, we introduce the HierArchical Leaf Vein Segmentation (HALVS) dataset, the first public hierarchical leaf vein segmentation dataset. HALVS comprises 5,057 real-scanned high-resolution leaf images collected from three plant species: soybean, sweet cherry, and London planetree. It also includes human-annotated ground truth for three orders of leaf veins, with a total labeling effort of 83.8 person-days. Based on HALVS, we further develop a label-efficient learning paradigm that leverages partial label information, i.e. missing annotations for tertiary veins. Empirical studies are performed on HALVS, revealing new observations, challenges, and research directions on leaf vein segmentation.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (40)
  1. Leaf venation networks of bornean trees: images and hand-traced segmentations. Ecology, 100(11), 2019.
  2. phenovein—a tool for leaf vein segmentation and analysis. Plant Physiology, 169(4):2359–2370, 2015.
  3. Swin-unet: Unet-like pure transformer for medical image segmentation. In European Conference on Computer Vision, pages 205–218. Springer, 2022.
  4. Semi-supervised learning (chapelle, o. et al., eds.; 2006)[book reviews]. IEEE Transactions on Neural Networks, 20(3):542–542, 2009.
  5. Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected crfs. IEEE Transactions on Pattern Analysis and Machine Intelligence, 40(4):834–848, 2017.
  6. Encoder-decoder with atrous separable convolution for semantic image segmentation. In Proceedings of the European Conference on Computer Vision (ECCV), pages 801–818, 2018.
  7. François Chollet. Xception: Deep learning with depthwise separable convolutions. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 1251–1258, 2017.
  8. Quantitative analysis of venation patterns of arabidopsis leaves by supervised image analysis. The Plant Journal, 69(3):553–563, 2012.
  9. Towards robust partially supervised multi-structure medical image segmentation on small-scale data. Applied Soft Computing, 114:108074, 2022.
  10. Revisiting vicinal risk minimization for partially supervised multi-label classification under data scarcity. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 4212–4220, 2022.
  11. Manual of Leaf Architecture. 2009.
  12. Xi Fang and Pingkun Yan. Multi-organ segmentation over partially labeled datasets with multi-scale feature abstraction. IEEE TMI, 2020.
  13. Automatic hierarchy classification in venation networks using directional morphological filtering for hierarchical structure traits extraction. Computational Biology and Chemistry, 80:187–194, 2019.
  14. Deep learning for plant identification using vein morphological patterns. Computers and Electronics in Agriculture, 127:418–424, 2016.
  15. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 770–778, 2016.
  16. Leo J Hickey. Classification of the architecture of dicotyledonous leaves. American Journal of Botany, 60(1):17–33, 1973.
  17. Network feature-based phenotyping of leaf venation robustly reconstructs the latent space. PLOS Computational Biology, 19(7):e1010581, 2023.
  18. An automatic hierarchical leaf venation segmentation based on region growing. In 2020 IEEE 5th International Conference on Signal and Image Processing (ICSIP), pages 124–128. IEEE, 2020.
  19. Net: a new framework for the vectorization and examination of network data. Source Code for Biology and Medicine, 12:1–11, 2017.
  20. Leaf vein segmentation with self-supervision. Computers and Electronics in Agriculture, 203:107352, 2022.
  21. Swin transformer: Hierarchical vision transformer using shifted windows. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 10012–10022, 2021.
  22. Plant leaf veins coupling feature representation and measurement method based on deeplabv3+. Frontiers in Plant Science, 13:1043884, 2022.
  23. Weakly-and semi-supervised learning of a deep convolutional network for semantic image segmentation. In Proceedings of the IEEE International Conference on Computer Vision, pages 1742–1750, 2015.
  24. Pytorch: An imperative style, high-performance deep learning library. Advances in Neural Information Processing Systems, 32, 2019.
  25. Corrigendum to: New handbook for standardised measurement of plant functional traits worldwide. Australian Journal of Botany, 64(8):715–716, 2016.
  26. Leaf extraction and analysis framework graphical user interface: segmenting and analyzing the structure of leaf veins and areoles. Plant Physiology, 155(1):236–245, 2011.
  27. Covariate shift and local learning by distribution matching. Dataset Shift in Machine Learning, pages 131–160, 2008.
  28. How do leaf veins influence the worldwide leaf economic spectrum? review and synthesis. Journal of experimental botany, 64(13):4053–4080, 2013.
  29. Leaf vein segmentation of medicinal plant using hessian matrix. In 2015 International Conference on Advanced Computer Science and Information Systems (ICACSIS), pages 275–279. IEEE, 2015.
  30. Improved non-destructive 2d and 3d x-ray imaging of leaf venation. Plant Methods, 14(1):1–15, 2018.
  31. Marginal loss and exclusion loss for partially supervised multi-organ segmentation. Medical Image Analysis, 70:101979, 2021.
  32. Fixmatch: Simplifying semi-supervised learning with consistency and confidence. Advances in Neural Information Processing Systems, 33:596–608, 2020.
  33. United Nations. 17 sustainable development goals. https://sdgs.un.org/goals, 2023.
  34. United Nations. Leave no one behind. https://unsdg.un.org/2030-agenda/universal-values/leave-no-one-behind, 2023.
  35. Johann Vollmann. Soybean versus other food grain legumes: A critical appraisal of the united nations international year of pulses 2016. Die Bodenkultur: Journal of Land Management, Food and Environment, 67(1):17–24, 2016.
  36. Semi-supervised semantic segmentation using unreliable pseudo-labels. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 4248–4257, 2022.
  37. Automated and accurate segmentation of leaf venation networks via deep learning. New Phytologist, 229(1):631–648, 2021.
  38. Revisiting weak-to-strong consistency in semi-supervised semantic segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 7236–7246, 2023.
  39. Prior-aware neural network for partially-supervised multi-organ segmentation. In ICCV, pages 10672–10681, 2019.
  40. A fast and automatic method for leaf vein network extraction and vein density measurement based on object-oriented classification. Frontiers in Plant Science, 11:499, 2020.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets