摘要
分别介绍了卫星遥感海冰监测、分类的传统方法,以及卷积神经网络在遥感影像分类识别中的应用成果。尝试将在图像识别、语言检测等方面取得成功的卷积神经网络算法应用在海冰图像分类中,利用其能够应对非线性、网络结构简单、可并行运算等能力去解决海冰数据分类问题。
This paper analyzes the traditional methods of sea ice monitoring and classification based on satellite remote sensing, and the application results of convolutional neural network in remote sensing image classification and recognition. Moreover, the convolutional neural networks algorithm, which has been successfully used in image recognition and language detection, is applied to the classification of sea ice images, and to solve the issues of sea ice data classification based on its simple network structure, capability in coping with nonlinearity and parallel computing.
作者
崔艳荣
邹斌
韩震
石立坚
刘森
CUI Yan-rong;ZOU Bin;HAN Zhen;SHI Li-jian;LIU Sen(College of Marine Science,Shanghai Ocean University,Shanghai 201306,China;National Satellite Ocean Application Service,Beijing 10081 China;Key Laboratory of Space Ocean Remote Sensing and Application,State Oceanic Administration,Beijing 100081 China)
出处
《海洋预报》
CSCD
北大核心
2019年第5期77-85,共9页
Marine Forecasts
基金
国家重点研发计划(2018YFC1407200
2018YFC1407206)
关键词
卫星遥感
海冰分类
应用
卷积神经网络
深度学习
satellite remote sensing
sea ice classification
application
convolutional neural networks
deep learning