Segmentation of typhoon cloud image by combining a discrete stationary wavelet transform with a continuous wavelet transform

$59.35 plus tax (Refund Policy)

Buy Article:

Abstract:

An algorithm is proposed that employs a multi-threshold technique to segment a typhoon cloud image. For both reducing the noise and enhancing the detail in the typhoon cloud image, the noise is reduced by a Wiener filter and the detail is enhanced by a nonlinear gain operator in the discrete stationary wavelet domain in the proposed algorithm. Then, the histogram equalization technique is used to enhance the global contrast of the processed image. In order to reduce the false peaks of the histogram of the denoised and enhanced typhoon cloud image (DETCI), a Bezier curve is used to smooth the histogram. An optimal segmentation threshold is then obtained from the multiple thresholds obtained from the Bezier histogram. The optimal threshold is used to segment the DETCI. The region of the maximum area in the segmented DETCI is selected as a region of interest (ROI). Thus other objects of small cloud masses are removed by the above method. We replace the area of the ROI with the corresponding area of the original typhoon cloud image to obtain the segmented ROI (SROI). Again the Bezier histogram is used to smooth the false peaks in the histogram of the SROI. In order to detect accurately the peaks and valleys in the curve of the Bezier histogram, a continuous wavelet transform is used to determine the location of peaks and valleys. After the wavelet transform, multi-segmented images at different scales are obtained. A criterion is employed to select an optimal segmentation scale. Finally, the whole typhoon cloud series is segmented accurately by the proposed method. Experimental results show that the proposed algorithm can efficiently segment the typhoon cloud series from a typhoon cloud image, and is better than the Olivo and HQ methods for analysing the structure of the typhoon wind field.

Document Type: Research Article

DOI: http://dx.doi.org/10.1080/01431160902912079

Affiliations: 1: College of Mathematics, Physics and Information Engineering, Zhejiang Normal University, Jinhua, China,State Key Laboratory of Remote Sensing Science, Jointly Sponsored by the Institute of Remote Sensing Applications of the Chinese Academy of Sciences and Beijing Normal University, Beijing, China 2: College of Mathematics, Physics and Information Engineering, Zhejiang Normal University, Jinhua, China

Publication date: April 1, 2010

More about this publication?
Related content

Share Content

Access Key

Free Content
Free content
New Content
New content
Open Access Content
Open access content
Subscribed Content
Subscribed content
Free Trial Content
Free trial content
Cookie Policy
X
Cookie Policy
ingentaconnect website makes use of cookies so as to keep track of data that you have filled in. I am Happy with this Find out more