Cloud classification of satellite image based on convolutional neural networks

摘要

Cloud classification of satellite image is the basis of meteorological forecast. Traditional machine learning methods need to manually design and extract a large number of image features, while the utilization of satellite image features is not high. This paper constructs a convolution neural network for cloud classification, which can automatically learn features and obtain classification results. The experimental results on the FY-2C satellite image show that the features extracted by deep convolution neural network are more favorable to the classification of satellite cloud. The performance of cloud classification based on deep convolution neural network is better than that of traditional machine learning methods. The method has high precision and good robustness.