• DocumentCode
    533179
  • Title

    Application research on intelligent pattern recognition methods in hail identification of weather radar

  • Author

    She Yong ; Yu Lei ; Wei Yi

  • Author_Institution
    Sci. & Technol. Res. Inst., Chengdu Univ. of Inf. Technol., Chengdu, China
  • Volume
    11
  • fYear
    2010
  • fDate
    22-24 Oct. 2010
  • Abstract
    Firstly, advantages of the learning ability of intelligent pattern recognition models, which have been used in hail identification of weather radar based on echo parameters, is discussed. Then, structures and working principles of hail identification models which based on fuzzy neural network and support vector machines (SVM) are described respectively. Finally, effect validation of hail identification has been finished by using echo samples from Chengdu and Kuitun to train the hail identification models. The experimental results show that hail identification models based on intelligent pattern recognition have the better effect than which based on the statistical pattern recognition. Besides, in the case of limited training samples, the identification model based on SVM has stronger adaptability than which based on fuzzy neural network.
  • Keywords
    fuzzy neural nets; geophysical image processing; meteorological radar; pattern recognition; radar computing; support vector machines; echo parameters; fuzzy neural network; hail identification; intelligent pattern recognition; statistical pattern recognition; support vector machines; weather radar; Fuzzy neural networks; Pattern recognition; Radar; Rain; Support vector machines; Training; echo parameters; fuzzy neural network; hail identification; support vector machines(SVM); weather radar;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer Application and System Modeling (ICCASM), 2010 International Conference on
  • Conference_Location
    Taiyuan
  • Print_ISBN
    978-1-4244-7235-2
  • Electronic_ISBN
    978-1-4244-7237-6
  • Type

    conf

  • DOI
    10.1109/ICCASM.2010.5623144
  • Filename
    5623144