Extraction of Ocean wave information from microwave nautical radar images: fourier method vs wavelet method

An, Jiaqi (2013) Extraction of Ocean wave information from microwave nautical radar images: fourier method vs wavelet method. Masters thesis, Memorial University of Newfoundland.

[English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (9MB)

Abstract

The development of algorithms for extracting ocean wave information from X-band nautical radar images is addressed. The algorithms are designed based on the three-dimensional (3-D) fast Fourier transform (FFT) and two-dimensional (2-D) continuous wavelet transform (CWT). First, in order to design the 3-D FFT-based wave algorithm, methods for extracting sea surface current information using least squares (LS), iterative LS, and normalized scalar product (NSP) methods are examined. With the results of mode classification of wave components from the iterative LS current estimation process, an improved wave measurement algorithm is proposed in which a modified band-pass filter is constructed to retain the contributions of the fundamental wave only and to remove the harmonics and non-wave components for wave information extraction. Noise components and aliasing effects are also eliminated based on existing theories and techniques associated with 3-D FFT-based wave field analysis. An alternative scheme for wave measurement using the 2-D CWT is also presented. The relationship of the scaling factor of the CWT and the minimum distinguishable wavenumber is developed and its expression is presented with a calibration factor. Empirical optimal values of the scaling parameter for variable wave conditions are provided based on extensive simulated radar tests. Based on the existing techniques employed in the classic 3-D FFT analysis, a look-up-table method and an iterative method are proposed to determine the calibration factor. For the iterative method, a self-adaptive wavelet analysis algorithm for wave measurement is developed. The designed wave algorithms are first evaluated with the numerical simulation under various sea conditions, assuming shadowing modulation (SH) and tilt modulation (TM). Then for further validation, the algorithms are applied to dual-polarized radar field data collected on the east coast of Canada. Results derived from FFT- and CWT- based algorithms are analyzed and compared with in situ TRIAXYS directional wave buoy da ta collected in the same region. The comparisons reveal that the proposed algorithms are practical and effective for retrieving ocean wave information from the radar data.

Actions (login required)

View Item