Title :
Forecast hail by analysis radar image
Author :
Xu, Wen-xia ; Li, Guo-dong ; Liao, Fei-jia
Author_Institution :
Xinjiang Weather Modification Office, Urumqi, China
Abstract :
In this paper, we use a new way to find some information from radar image. We process the image by Hole-Full Cell Nerve Network (CNN) and then add the all data in the image. We can find there seems some relation between the number and hail shooting. We process more pictures. And gave more parameters to do so. Dug some data as information. At last we find some regular to distinguish the cloud with hail or not. It is a area be find. It will be hail shooting when the data come to the area. We find those are useful way for forecasting of hail.
Keywords :
atmospheric precipitation; atmospheric techniques; geophysical image processing; ice; image segmentation; remote sensing by radar; weather forecasting; Hole-Full Cell Nerve Network; data analysis; hail identification; hail shooting forecasting; radar image segmentation; Clouds; Doppler radar; Meteorological radar; Meteorology; Radar imaging; Spaceborne radar; Hail identification; Hole-filling CNN; Radar image;
Conference_Titel :
Machine Learning and Cybernetics (ICMLC), 2010 International Conference on
Conference_Location :
Qingdao
Print_ISBN :
978-1-4244-6526-2
DOI :
10.1109/ICMLC.2010.5580568
