Typhoon Classification Model Based on Multi-Scale Convolution Feature Fusion

Table 1. Typhoon classification standard
View table
Table 1. Typhoon classification standard
Label Levelof typhoon Maximum windspeed /(m·s^-1)
Class 1 Tropical storm ≤24.4
Class 2 Severe tropical storm 24.5-32.6
Class 3 Typhoon 32.7-41.4
Class 4 Violent typhoon 41.5-50.9
Class 5 Super typhoon ≥51.0

Table 2. Accuracies of different models on typhoon datasets
View table
Table 2. Accuracies of different models on typhoon datasets
Model Train accuracy Test accuracy
LeNet-5 0.8657 0.8559
AlexNet 0.9508 0.9432
Hybrid Model^[17] 0.9714 0.9366
SIFT +CNN^[18] 0.9272 0.9205
MS-TyCNN 1.0000 0.9988

Table 3. Distribution of datasets
View table
Table 3. Distribution of datasets
Dataset Number oftraining sample Number oftesting sample Class
MNIST 42000 10000 10
CIFAR-10 50000 10000 10

Table 4. Generalization of MS-TyCNN model
View table
Table 4. Generalization of MS-TyCNN model
Dataset LeNet-5 AlexNet Hybrid model SIFT+CNN MS-TyCNN
MNIST 0.9637 0.9806 0.9725 0.9837 0.9814
CIFAR-10 0.7320 0.8431 0.8926 0.8759 0.8763

Tools

Get Citation

Copy Citation Text

Peng Lu, Peiqi Zou, Guoliang Zou. Typhoon Classification Model Based on Multi-Scale Convolution Feature Fusion[J]. Laser & Optoelectronics Progress, 2019, 56(16): 160101

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites