浏览全部资源
扫码关注微信
宁波大学 信息科学与工程学院, 浙江 宁波 315211
[ "符冉迪(1971-),男,宁波大学信息科学与工程学院副教授、硕士生导师,主要研究方向为压缩感知、数字图像处理等。E-mail: furandi@nbu.edu.cn" ]
[ "司 光(1996-),男,山东济宁人,硕士研究生,主要从事数字图像处理方面的研究。E-mail: 2529190845@qq.com" ]
收稿日期:2021-03-15,
修回日期:2021-05-14,
纸质出版日期:2022-04-25
移动端阅览
符冉迪,司光,金炜.深度网络与FSVM集成学习的卫星云图云分类[J].光学精密工程,2022,30(08):917-927.
FU Randi,SI Guang,JIN Wei.Cloud classification based on ensemble learning combining with deep neural network and FSVM[J].Optics and Precision Engineering,2022,30(08):917-927.
符冉迪,司光,金炜.深度网络与FSVM集成学习的卫星云图云分类[J].光学精密工程,2022,30(08):917-927. DOI: 10.37188/OPE.20223008.0917.
FU Randi,SI Guang,JIN Wei.Cloud classification based on ensemble learning combining with deep neural network and FSVM[J].Optics and Precision Engineering,2022,30(08):917-927. DOI: 10.37188/OPE.20223008.0917.
准确的云分类模型对气象监测有重要的意义,传统机器学习云分类模型依赖手工特征提取,容易受噪声数据影响,模型泛化能力较差。深度网络分类模型能自动学习图像深度特征,但是对于图像边缘与细节分类效果不佳。本文针对上述问题进行研究。首先提取Himawari-8卫星云图光谱特征、纹理特征用以训练模糊支持向量机(Fuzzy Support Vector Machine,FSVM)模型;同时利用不同通道云图训练深度网络,学习云图深度特征;最后,根据不同模型特性,训练元分类器对各模型输出进行融合,设计了一种基于深度网络与FSVM集成学习的云分类方法,该方法综合不同模型优势,利用不同模型间的互补性提高云分类结果的鲁棒性和可信度。相比单独使用FSVM或深度网络的分类模型,本文集成学习方法在众多评价指标中有更好的表现,平均命中率、平均误报率和平均临界成功指数分别达到0.924 5、0.079 6、0.858 1;与其它云分类模型相比,本文方法也有更好的分类效果;在具体案例测试中也发现,该方法对于不同云类混合区有更高的识别精度,而且能更加准确的识别云团边缘及细节。本文模型能够满足云分类模型稳定可靠、高精度、泛化性能强的要求。
Accurate cloud classification is of great significance for meteorological monitoring. Traditional machine learning models rely on hand-craft featurs, which is sensitive to noise data and the generalization ability is also poor. Deep neural network can automatically learn the depth features of image, but it is not good at image edge and detail classification, this paper studies on the basis of the above problems. First, the spectral features and texture features are extracted from himawari-8 satellite images to train fuzzy support vector machine (FSVM) model. At the same time, different channels of cloud images are selected to train deep neural network to learn the depth features for cloud classification. Finally, according to the characteristics of different models, the output of the two models is fused by ensemble learning to improve the classification accuracy. This article designs a cloud classification model based on ensemble learning which fuses deep neural network and FSVM. It combines the advantages of different models and makes use of the complementarity between different models to improve the robustness and reliability of the model.The experimental results show that: compared with model which uses a single model alone, the ensemble learning method proposed in this article has better performance in different evaluation indicators, The average POD, FAR and CSI were 0.9245, 0.0796 and 0.8581 respectively; this method also has better recognition effect when compared with other cloud classification models; in specific cases, it is found that this method has higher recognition accuracy in clouds mixed regions, and it can identify cloud edge and cloud details more accurately.This model can satisfy the requirements of stability, reliability, high precision and strong generalization performance of cloud classification model.
杨澄 , 袁招洪 , 顾松山 . 用多谱阈值法进行GMS-5卫星云图云型分类的研究 [J]. 南京气象学院学报 , 2002 , 25 ( 6 ): 747 - 754 . doi: 10.3969/j.issn.1674-7097.2002.06.004 http://dx.doi.org/10.3969/j.issn.1674-7097.2002.06.004
YANG C , YUAN Z H , GU S S . Cloud classification of GMS-5 satellite imagery by the use of multispectral threshold technique [J]. Journal of Nanjing Institute of Meteorology , 2002 , 25 ( 6 ): 747 - 754 . (in Chinese) . doi: 10.3969/j.issn.1674-7097.2002.06.004 http://dx.doi.org/10.3969/j.issn.1674-7097.2002.06.004
张振华 , 苗春生 , 曾智华 , 等 . 一种人工神经网络云分类方法的改进与应用 [J]. 应用气象学报 , 2012 , 23 ( 3 ): 355 - 363 . doi: 10.3969/j.issn.1001-7313.2012.03.012 http://dx.doi.org/10.3969/j.issn.1001-7313.2012.03.012
ZHANG Z H , MIAO C S , ZENG Z H , et al . Improvement and application of artificial neural networks to cloud classification [J]. Journal of Applied Meteorological Science , 2012 , 23 ( 3 ): 355 - 363 . (in Chinese) . doi: 10.3969/j.issn.1001-7313.2012.03.012 http://dx.doi.org/10.3969/j.issn.1001-7313.2012.03.012
韩丁 , 严卫 , 任建奇 , 等 . 基于支持向量机的CloudSat卫星云分类算法 [J]. 大气科学学报 , 2011 , 34 ( 5 ): 583 - 591 . doi: 10.3969/j.issn.1674-7097.2011.05.008 http://dx.doi.org/10.3969/j.issn.1674-7097.2011.05.008
HAN D , YAN W , REN J Q , et al . Cloud type classification algorithm for CloudSat satellite based on support vector machine [J]. Transactions of Atmospheric Sciences , 2011 , 34 ( 5 ): 583 - 591 . (in Chinese) . doi: 10.3969/j.issn.1674-7097.2011.05.008 http://dx.doi.org/10.3969/j.issn.1674-7097.2011.05.008
来旭 , 李国辉 , 张军 . 基于半监督FCM聚类算法的卫星云图分类 [J]. 国防科技大学学报 , 2008 , 30 ( 6 ): 73 - 77 . doi: 10.3969/j.issn.1001-2486.2008.06.016 http://dx.doi.org/10.3969/j.issn.1001-2486.2008.06.016
LAI X , LI G H , ZHANG J . Satellite cloud images classification based on semi-supervised FCM method [J]. Journal of National University of Defense Technology , 2008 , 30 ( 6 ): 73 - 77 . (in Chinese) . doi: 10.3969/j.issn.1001-2486.2008.06.016 http://dx.doi.org/10.3969/j.issn.1001-2486.2008.06.016
KIM K B , SONG D H , BAE Y . A fuzzy logic approach for cloud classification based on near-infrared image features [C]. 2013 International Conference on Fuzzy Theory and Its Applications (iFUZZY). 68,2013 , Taipei, China. IEEE , 2013 : 130 - 133 . doi: 10.1109/ifuzzy.2013.6825423 http://dx.doi.org/10.1109/ifuzzy.2013.6825423
陈珂锐 , 孟小峰 . 机器学习的可解释性 [J]. 计算机研究与发展 , 2020 , 57 ( 9 ): 1971 - 1986 . doi: 10.7544/issn1000-1239.2020.20190456 http://dx.doi.org/10.7544/issn1000-1239.2020.20190456
CHEN K R , MENG X F . Interpretation and understanding in machine learning [J]. Journal of Computer Research and Development , 2020 , 57 ( 9 ): 1971 - 1986 . (in Chinese) . doi: 10.7544/issn1000-1239.2020.20190456 http://dx.doi.org/10.7544/issn1000-1239.2020.20190456
LAI C , LIU T , MEI R W , et al . The cloud images classification based on convolutional neural network [C]. 2019 International Conference on Meteorology Observations (ICMO) . December 28-31, 2019 , Chengdu, China . IEEE , 2019 : 1 - 4 . doi: 10.1109/icmo49322.2019.9026121 http://dx.doi.org/10.1109/icmo49322.2019.9026121
毋立芳 , 贺娇瑜 , 简萌 , 等 . 局部聚类分析的FCN-CNN云图分割方法 [J]. 软件学报 , 2018 , 29 ( 4 ): 1049 - 1059 . doi: 10.13328/j.cnki.jos.005409 http://dx.doi.org/10.13328/j.cnki.jos.005409
WU L F , HE J Y , JIAN M , et al . Local clustering analysis based FCN-CNN for cloud image segmentation [J]. Journal of Software , 2018 , 29 ( 4 ): 1049 - 1059 . (in Chinese) . doi: 10.13328/j.cnki.jos.005409 http://dx.doi.org/10.13328/j.cnki.jos.005409
RUSYN B , KORNIY V , LUTSYK O , et al . Deep learning for atmospheric cloud image segmentation [C]. 2019 XIth International Scientific and Practical Conference on Electronics and Information Technologies (ELIT). 1618,2019 , Lviv, Ukraine. IEEE , 2019 : 188 - 191 . doi: 10.1109/elit.2019.8892285 http://dx.doi.org/10.1109/elit.2019.8892285
徐丽娟 . 基于纹理分析云的分类技术的研究 [D]. 南京 : 南京信息工程大学 , 2012 .
XU L J . Research on the Cloud Classification Technology Based on the Texture Analysis [D]. Nanjing : Nanjing University of Information Science & Technology , 2012 . (in Chinese)
ZHU X F , MA B , GUO G J . An adaptive-weight regularization method for multi-classifier fusion decision [C]. 2014 International Conference on Mechatronics and Control (ICMC). 35,2014 , Jinzhou, China. IEEE , 2014 : 343 - 346 . doi: 10.1109/icmc.2014.7231575 http://dx.doi.org/10.1109/icmc.2014.7231575
郑益勤 , 杨晓峰 , 李紫薇 . 深度学习模型识别静止卫星图像海上强对流云团 [J]. 遥感学报 , 2020 , 24 ( 1 ): 97 - 106 . doi: 10.11834/jrs.20208209 http://dx.doi.org/10.11834/jrs.20208209
ZHENG Y Q , YANG X F , LI Z W . Detection of severe convective cloud over sea surface from geostationary meteorological satellite images based on deep learning [J]. Journal of Remote Sensing , 2020 , 24 ( 1 ): 97 - 106 . (in Chinese) . doi: 10.11834/jrs.20208209 http://dx.doi.org/10.11834/jrs.20208209
李冰洁 . 气象卫星系统的云图自动分类识别研究 [D]. 西安 : 西安科技大学 , 2019 . doi: 10.23919/picmet.2019.8893960 http://dx.doi.org/10.23919/picmet.2019.8893960
LI B J . Automatic Classification and Recognition Research about Cloud on Meteorological Satellite System [D]. Xi'an : Xi'an University of Science and Technology , 2019 . (in Chinese) . doi: 10.23919/picmet.2019.8893960 http://dx.doi.org/10.23919/picmet.2019.8893960
CHETHAN H K , KUMAR G H , R R . Texture based approach for cloud classification using SVM [C]. 2009 International Conference on Advances in Recent Technologies in Communication and Computing . 2728,2009 , Kottayam , India . IEEE , 2009 : 688 - 690 . doi: 10.1109/artcom.2009.43 http://dx.doi.org/10.1109/artcom.2009.43
JIAO L , HUO L , HU C . Refined UNet: UNet-Based Refinement Network for Cloud and Shadow Precise Segmentation [J]. Remote Sensing , 2020 , 12 ( 12 ): 2001 . doi: 10.3390/rs12122001 http://dx.doi.org/10.3390/rs12122001
邱云志 , 汪廷华 , 余武清 . 模糊支持向量机研究综述 [J]. 赣南师范大学学报 , 2020 , 41 ( 3 ): 26 - 32 . doi: 10.13698/j.cnki.cn36-1346/c.2020.03.007 http://dx.doi.org/10.13698/j.cnki.cn36-1346/c.2020.03.007
QIU Y Z , WANG T H , YU W Q . A review of fuzzy support vector machines [J]. Journal of Gannan Normal University , 2020 , 41 ( 3 ): 26 - 32 . (in Chinese) . doi: 10.13698/j.cnki.cn36-1346/c.2020.03.007 http://dx.doi.org/10.13698/j.cnki.cn36-1346/c.2020.03.007
0
浏览量
639
下载量
1
CSCD
关联资源
相关文章
相关作者
相关机构