Automatic extraction of ocean internal waves based on multi-scale fusion and elliptical tracking region method
Qicong Wang, Shaoling Shang, Qingkai Zhen, Yan Li, Quan Zou · 2010
Since the remote sensing images are always contaminated by multiplicative noise, solar flares and clouds, many interference curves are produced and the curves of internal waves are divided into several segments after image processing. Most of present detection methods cannot extract the internal wave automatically and directly. To handle this problem, we propose a novel method which utilizes wavelet multi-scale fusion and elliptic tracking region to eliminate the impact of different interference factors and extract the continuous and complete curves of most internal waves. Experimental results show that the proposed method can work well even for the remote sensing images with the strong interferences.