Font Size: a A A

Based On Satellite Images And The Automatic Identification Method Of Convetive Clouds

Posted on:2012-02-17Degree:MasterType:Thesis
Country:ChinaCandidate:X D BaoFull Text:PDF
GTID:2218330344951083Subject:Transportation planning and management
Abstract/Summary:PDF Full Text Request
Convective clouds are often accompanied by hail, tornadoes, heavy rain and other inclement weather. It makes a great impact on the flight, aviation is generally found ahead of convective clouds and then flying around, so the identification,monitoring and nowcasting of convective clouds is a key. This paper selects satellite clouds in June 2010 as samples to make a reach about identification of convective clouds. We use three ways of the texture, the threshold, threshold with texture to identify the convective cloudsTexture Recognition Methods used two algorithms of GLCM and gray-Gradient Co-occurrence Matrix to identify convective clouds. Both algorithms are applied to identify the convective clouds. But, gray - Gradient Co-occurrence Matrix is better than GLCM. The disadvantage of statistical algorithm is complexity and statistical algorithm has a large computation. Threshold method has advantages of faster computation. But, the overall recognition rate is not high. This paper proposes a new method of identifying convective clouds and it based on the two methods of texture and brightness threshold. This new method is threshold with texture to identify the convective clouds. First of all, according to set the threshold, remove most of darker parts. Then, about the parts of images preserved, use Gray-gradient co-occurrence matrix algorithm to make texture recognition. Finally, the new method makes accurate identification of the convective clouds. It reduces the computation of using traditional gray - gradient co-occurrence matrix method to identify convective clouds and increases the recognition rate.Through the sample test, statistics of the recognition rate of each of the three identification methods and summed up the advantages and disadvantages of them. Brightness hreshold with texture recognition method can effectively identify the satellite images of convective clouds. This is conducive to the monitoring and nowcasting of convective clouds and provides a method and guidance for dispatchers and other aiviation personnal to identify convective clouds.
Keywords/Search Tags:Gray level co-occurrence matrix, Gray-Gradient Co-occurrence Matrix, Convective clouds, Satellite image
PDF Full Text Request
Related items