A CNN-based vortex identification method
- 94 Downloads
Vortex identification and visualization are important for understanding the underlying physical mechanism of the flow field and have been intensively studied recently. Local vortex identification methods could provide results in a rapid way, but they require the choice of a suitable criterion and threshold, which leads to poor robustness. Global vortex identification methods could obtain reliable results, while they require considerable user input and are computationally intractable for large-scale data sets. To address the problems described above, we present a novel vortex identification method based on the convolutional neural network (CNN). The proposed method integrates the advantages of both the local and global vortex identification methods to achieve higher precision and recall efficiently. In specific, the proposed method firstly obtains the labels of all grid points using a global and objective vortex identification method and then samples local patches around each point in the velocity field as the inputs of CNN. After that it trains the CNN to decide whether the central points of these patches belong to vortices. By this way, our method converts the vortex identification task to a binary classification problem, which could detect vortices quickly from the flow field in an objective and robust way. Extensive experimental results demonstrate the efficacy of our proposed method, and we expect this method can replace or supplement existing traditional methods.
KeywordsVortex identification CNN Unsteady flow field
This work was supported in part by the National Key Research and Development Program of China (# 2016YFB0200701) and the National Natural Science Foundation of China (# 61806205, # 91530324 and # 91430218).
- Biswas A, Thompson D, He W, Deng Q, Chen C-M, Shen H-W, Machiraju R, Rangarajan A (2015) An uncertainty-driven approach to vortex analysis using oracle consensus and spatial proximity. In: IEEE pacific visualization symposium, IEEE Computer Society, Los Alamitos, pp 1–8. https://doi.org/10.1109/PACIFICVIS.2015.715638
- Franz K, Roscher R, Milioto A, Wenzel S, Kusche J (2018) Ocean eddy identification and tracking using neural networks. In: Computer vision and pattern recognitionGoogle Scholar
- Girshick RB, Donahue J, Darrell T, Malik J (2014) Rich feature hierarchies for accurate object detection and semantic segmentation. In: IEEE conference on computer vision and pattern recognition proceedings of the 2014 CVPR ’14. IEEE Computer Society, Los Alamitos, pp 580–587Google Scholar
- Jiang M, Machiraju R, Thompson D (2005) Detection and visualization of vortices. In: Visualization handbook, pp 295–309. https://doi.org/10.1016/B978-012387582-2/50016-2
- Kingma DP, Ba J (2014) Adam: a method for stochastic optimization. In: Computer sciencesGoogle Scholar
- Lguensat R, Sun M, Fablet R, Mason E, Tandeo P, Chen G (2017) Eddynet: a deep neural network for pixel-wise classification of oceanic eddies. In: Computer vision and pattern recognitionGoogle Scholar
- Lucas BD, Kanade T (1981) An iterative image registration technique with an application to stereo vision. In: International joint conference on artificial intelligence. Elsevier, Amsterdam, pp 674–679Google Scholar
- Sadarjoen A, Post FH, Ma B, Banks DC, Pagendarm HG (2002) Selective visualization of vortices in hydrodynamic flows. In: Proceedings of visualization. IEEE Computer Society, Los Alamitos. pp 419–422, https://doi.org/10.1109/VISUAL.1998.745333
- Simard PY, Steinkraus D, Platt JC (2003) Best practices for convolutional neural networks applied to visual document analysis. In: International conference on document analysis and recognition. IEEE Computer Society, Los Alamitos, pp 958–963, https://doi.org/10.1109/ICDAR.2003.1227801
- Ströfer CM, Wu J, Xiao H, Paterson E (2018) Data-driven, physics-based feature extraction from fluid flow fields. In: Fluid dynamicsGoogle Scholar