DocumentCode :
736812
Title :
Based on K-Means Clustering and CNN Algorithm Research in Hail Cloud Determination
Author :
Xue, Wang ; Feijia, Liao ; Wenxia, Xu ; Kun, Guo ; Guodong, Li
fYear :
2015
fDate :
13-14 June 2015
Firstpage :
232
Lastpage :
235
Abstract :
Hail is one of the main meteorological disasters in our country, also it is the hot current research. The forecasting of hail is mostly used in the data return from radar. Based on radar return image, using the statistics of the K-means clustering algorithm and cellular neural networks (CNN) algorithm, processing the cloud image, and get the difference between the inside and outside layer outline. Compare the hail cloud image with the no hail cloud image to get the variance, we can find the difference of internal and external contours Using the hail cloud images and the no hail cloud image of Shihezi in Xinjiang province verification, validation of the conclusion of this paper is effective. Show that combined with K means clustering algorithm and CNN contour extraction algorithm to get the difference of internal and external contours is a effective method of forecasting the hail.
Keywords :
Cellular neural networks; Clouds; Clustering algorithms; Image color analysis; Image edge detection; Mathematical model; Meteorology; CNN; K means clustering; Lab color space; contour extraction contour; difference of internal and external contours;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Measuring Technology and Mechatronics Automation (ICMTMA), 2015 Seventh International Conference on
Conference_Location :
Nanchang, China
Print_ISBN :
978-1-4673-7142-1
Type :
conf
DOI :
10.1109/ICMTMA.2015.64
Filename :
7263555
Link To Document :
بازگشت