首页 > 论文 > 光电工程 > 41卷 > 11期(pp:29-35)


Fuzzy Support Vector Machines for Cumulus Cloud Detection

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文


为解决 SVM在积雨云检测中的难题, 本文构造了一种模糊支持向量机( FSVM), 首先根据训练样本的分布特性, 定义了相邻样本距离类中心的距离变化率, 然后通过计算距离变化率来剔除训练集中可能的噪声与野值样本, 从而有效克服了传统基于紧密度的 FSVM在计算最小超球半径时易受噪声与野值干扰的缺点, 使得所计算的隶属度能更好地反映不同样本的差异。实验结果表明, 对于 FY2D卫星云图, 采用从不同通道所提取的光谱特征, 本文方法的积雨云检测准确率与传统 SVM和基于紧密度的 FSVM相比, 分别平均提高 2%和 1%, 且具有更强的适应性及噪声鲁棒性。


Using satellite imagery for cumulus cloud detection has an important significance for preventing meteorological disasters. Support Vector Machine (SVM), which can seek the best compromise between the complexity of the model and the learning ability based on finite sample information, is expected to play a role in the cumulus cloud detection. However, the traditional SVM is very sensitive to the samples of noise and outlier, and doesn’t possess the skill of fuzzy treatment, which doesn’t meet the fuzzy and uneven characteristics of satellite imagery and the complex and diverse cloud patterns. In order to solve the problem of SVM, this paper introduces Fuzzy Support Vector Machine (FSVM) and defines the range-rate of the distances from the adjacent samples to the class center, based on the distribution characteristics of training samples. Then, on the basis of the range-rate, we weed out the possible noises and outliers of training set and overcome the shortcoming that the affinity FSVM is susceptible to noises and outliers at the time of calculating the radius of smallest hyper-sphere, so as to make the obtained membership better reflect the variance of different sample sets. The experimental results show that, for FY2D satellite imageries, extracting 8-d spectral features from different channels, compared with traditional SVM and affinity FSVM, the accuracies of cumulus cloud detection based on the proposed method increase respectively by about 2% and 1%. The proposed method owns stronger adaptability and noise robustness, and can make better effect on early warning disastrous weather such as thunderstorm.

广告组1 - 空间光调制器+DMD




基金项目:国家自然科学基金 (61271399, 61373068); 宁波市自然科学基金 (2011A610192, 2013A610055); 宁波市国际合作项目 (2013D10011); 宁波大学学科项目 (XKXL1306)




作者单位    点击查看

符冉迪:宁波大学信息科学与工程学院, 浙江 宁波 315211
田文哲:宁波大学信息科学与工程学院, 浙江 宁波 315211
金炜:宁波大学信息科学与工程学院, 浙江 宁波 315211
刘箴:宁波大学信息科学与工程学院, 浙江 宁波 315211
王文龙:宁波大学信息科学与工程学院, 浙江 宁波 315211


备注:符冉迪 (1971-), 男(汉族), 浙江宁波人。副教授, 主要研究方向为数字图像处理、模式识别和 GIS。

【1】LIU Yan’an, WEI Ming, GAO Wei, et al. Short-term Automatic Forecast Algorithm of Severe Convective Cloud Identification Using FY-2 IR Images [J]. Journal of Remote Sensing(S1007-4619), 2012, 16(1): 79-85.

【2】THOMAS F, REMY R. An Algorithm for the Detection and Tracking of Tropical Mesoscale Convective Systems Using Infrared Images from Geostationary Satellite [J]. IEEE Transactions on Geoscience and Remote Sensing(S0196-2892), 2013, 51(7): 4302-4315.

【3】BARNALI G, GUPINATH B, SANJAY G. Fuzzy Min-Max Neural Network for Satellite Infrared Image Clustering [C]// Emerging Applications of Information Technology, Kolkata, Nov 30-Dec 1, 2012: 239-242.

【4】XU Jinwei, WANG Ping, XIE Yiyang. Image Segmentation of Typhoon Spiral Cloud Bands Based on Support Vector Machine [C]// Proceedings of the Eighth International Conference on Machine Learning and Cybernetics, Baoding, July 12-15, 2009, 2: 1088-1093.

【5】THOMAS F, REMY R. Composite Life Cycle of Tropical Mesoscale Convective Systems from Geostationary and Low Earth Orbit Satellite Observations: Method and Sampling Considerations [J]. Quarterly Journal of the Royal Meteorological Society(S1477-870X), 2013, 139(4): 941-953.

【6】李刚, 万幼川. 商空间理念下面向对象的遥感影像分类 [J].光电工程, 2011, 38(2): 108-114.
LI Gang, WAN Youchuan. Object-oriented Classification Method Based on Quotient Space Theory [J]. Opto-Electronic Engineering, 2011, 38(2): 108-114.

【7】杨永敏, 樊继壮, 赵杰. 基于超熵和模糊集理论的带钢表面缺陷分割 [J].光学 精密工程, 2011, 19(7): 1651-1658.
YANG Yongmin, FAN Jizhuang, ZHAO Jie. Steel Strip Surface Defect Segmentation Based on Excess Entropy and Fuzzy Set Theory [J]. Optics and Process Engineering, 2011, 19(7): 1651-1658.

【8】屠礼芬, 仲思东, 彭祺. 基于模糊集理论的运动目标检测 [J].应用光学, 2013, 34(5): 820-824.
TU Lifen, ZHONG Sidong, PENG Qi. Moving Object Detection by Fuzzy Set Theory [J]. Journal of Applied Optics, 2013, 34(5): 820-824.

【9】LIN Chunfu, WANG Shengde. Fuzzy Support Vector Machines [J]. IEEE Transactions on Neural Networks(S1045-9227), 2002, 13(2): 464-471.

【10】边肇祺, 张学工. 模式识别 [M].北京: 清华大学出版社, 2010.
BIAN Zhaoqi, ZHANG Xuegong. Pattern Recognition [M]. Beijing: Tsinghua University Press, 2010.

【11】张翔, 肖小玲, 徐光祐. 基于样本之间紧密度的模糊支持向量机方法 [J].软件学报, 2006, 17(5): 951-958.
ZHANG Xiang, XIAO Xiaoling, XU Guangyou. Fuzzy Support Vector Machine Based on Affinity Among Samples [J]. Journal of Software, 2006, 17(5): 951-958.

【12】TAX D, DUIN R. Support Vector Data Description [J]. Machine Learning(S1573-0565), 2004, 54(1): 45-66.

【13】石小云. 基于神经网络方法的卫星图像云分类 [D].青岛: 中国海洋大学, 2012: 31.
SHI Xiaoyun. Cloud Classification of Satellite Imagery Based on Neural Network Methods [D]. Qingdao: Ocean University of China, 2012: 31.


FU Randi,TIAN Wenzhe,JIN Wei,LIU Zhen,WANG Wenlong. Fuzzy Support Vector Machines for Cumulus Cloud Detection[J]. Opto-Electronic Engineering, 2014, 41(11): 29-35

符冉迪,田文哲,金炜,刘箴,王文龙. 面向积雨云检测的模糊支持向量机[J]. 光电工程, 2014, 41(11): 29-35

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF