Domain knowledge integration into deep learning for typhoon intensity classification.
Domain knowledge integration into deep learning for typhoon intensity classification.
复制标题
DOI：
10.1038/s41598-021-92286-w
复制
发表时间：
2021-06-21
影响因子：
4.6
通讯作者：
Miyata R
中科院分区：
文献类型：
--
作者：
Higa M;Tanahara S;Adachi Y;Ishiki N;Nakama S;Yamada H;Ito K;Kitamoto A;Miyata R
在本报告中，我们提出了一种深度学习技术，通过结合气象领域知识，从单个卫星图像中高精度估计台风的强度等级。通过使用Visual Geometric Group的VGG-16模型，对图像进行鱼眼畸变预处理，增强台风的眼、眼壁和云分布，我们获得了比之前研究更高的分类精度，即使是顺序分割验证。通过将VGG特征图的t-分布随机邻居嵌入（t-SNE）图与原始卫星图像进行比较，我们也验证了鱼眼预处理有利于聚类的形成，表明我们的模型可以成功提取与台风强度等级相关的图像特征。此外，采用梯度加权类激活映射（gradient-weighted class activation mapping, Grad-CAM）来突出眼睛和眼睛周围的云分布，这是强度分类的重要区域；结果表明，我们的模型定性地获得了类似于领域专家的观点。一系列分析表明，仅使用深度学习的数据驱动方法存在局限性，领域知识的整合可能带来新的突破。
In this report, we propose a deep learning technique for high-accuracy estimation of the intensity class of a typhoon from a single satellite image, by incorporating meteorological domain knowledge. By using the Visual Geometric Group’s model, VGG-16, with images preprocessed with fisheye distortion, which enhances a typhoon’s eye, eyewall, and cloud distribution, we achieved much higher classification accuracy than that of a previous study, even with sequential-split validation. Through comparison of t-distributed stochastic neighbor embedding (t-SNE) plots for the feature maps of VGG with the original satellite images, we also verified that the fisheye preprocessing facilitated cluster formation, suggesting that our model could successfully extract image features related to the typhoon intensity class. Moreover, gradient-weighted class activation mapping (Grad-CAM) was applied to highlight the eye and the cloud distributions surrounding the eye, which are important regions for intensity classification; the results suggest that our model qualitatively gained a viewpoint similar to that of domain experts. A series of analyses revealed that the data-driven approach using only deep learning has limitations, and the integration of domain knowledge could bring new breakthroughs.
登录
查看更多内容
影响因子：
1.9
作者：
Nakazawa, Tetsuo;Hoshino, Shunsuke
通讯作者：
Hoshino, Shunsuke
DOI：
10.1007/3-540-46805-6_19
发表时间：
1999-01-01
期刊：
SHAPE, CONTOUR AND GROUPING IN COMPUTER VISION
影响因子：
--
作者：
LeCun, Y;Haffner, P;Bengio, Y
通讯作者：
Bengio, Y
影响因子：
4.9
作者：
Kossin, James P.;Olander, Timothy L.;Knapp, Kenneth R.
通讯作者：
Knapp, Kenneth R.
影响因子：
2.9
作者：
Olander, Timothy L.;Velden, Christopher S.
通讯作者：
Velden, Christopher S.
影响因子：
4.9
作者：
Hagen, Andrew B.;Strahan-Sakoskie, Donna;Luckett, Christopher
通讯作者：
Luckett, Christopher