Li Chung Wu

Ocean remotely sensed image analysis using two-dimensional continuous wavelet transforms

Ocean remote sensing is a useful way to obtain ocean wave information. Due to possible inhomogeneities from remotely sensed images, the current work proposes issues concerning ocean wave image analysis using the two-dimensional continuous wavelet transforms (2-D CWTs) to calculate local wave image spectra from inhomogeneous images. To optimize the algorithm of the 2-D CWT for wave image analysis, this work explores ideal parameter values for the wavelet function. The current study also analyses the limits of spatial image resolution and wave image size. After implementing the 2-D CWT on satellite and X-band radar images, this study presents local image spectra and ocean wave information from all the ocean images. These local image spectra reveal the phenomenon of wave refraction and wave nonlinearity nearshore. Compared to real wave spectra, the wavelet spectra present accurate results to describe local wave features in the spatial frequency domain.

Quantification of non-homogeneity from ocean remote-sensing images using two-dimensional continuous wavelet transforms

In this study, a two-dimensional continuous wavelet transform is applied to quantify the non-homogeneity from remote-sensing wave images. Our study shows that the non-homogeneity index, which is proposed here, is capable of identifying the degree of non-homogeneity from the wave-field images. However, the influence from the edges of the wave-field image should be considered in analysis. The calculated results from the simulated wave fields and from the natural wave-field images reveal that the non-homogeneity index is influenced by the bathymetry inside the area of the ocean-wave image. It conforms to the character of the natural wave non-homogeneity. After discussing the wave characteristics from the wave-field image with a high non-homogeneity index, an inhomogeneous algorithm is recommended for obtaining accurate and reasonable wave results from the wave image.

Study of wave group velocity estimation from inhomogeneous sea-surface image sequences by spatiotemporal continuous wavelet transform

In this paper, a new image processing technique is introduced for the analysis of consecutive ocean wave images by spatiotemporal continuous wavelet transform (STCWT) using the Morlet wavelet as the mother wavelet. This technique, which has been adapted to give a full time-frequency and spatiofrequency representation of ocean surface waves from remotely captured wave image sequences, provides more detailed information from wave-field measurements than the traditional Fourier transform (FT) method. A series of numerical simulations of wave image sequences was analyzed to justify the algorithm and to present quantitative theoretical results on the propagation of ocean waves for normal incidence, diffraction, reflection, and shoaling in coastal areas. The comparisons of these estimates to simulated conditions for several wave parameters show that the wavelet theory is applicable to the identification of wave spectra from inhomogeneous sea-surface image sequences.

Searching for freak waves from in-situ buoy measurements

There are several criteria to determine a freak wave in the ocean. These criteria are independent but the results are often not the same. This study tests on four presented criteria including the ratios of H/H1/3 and H/Hs, the kurtosis k, and the groupiness factor GF, and add one more condition (Hs>1m) on the searching of freak wave in the database of data buoy during typhoon from 1998 to 2009. The water elevation is inversed from buoy measured acceleration signal by wavelet transform. It has still not enough knowledge to know how a freak wave looks like in the real world. The result shows more confident cases of freak waves by the joint assessment of 5 conditions. In this study, we find 18 freak wave cases within the 9122 time series during Typhoons.

Continuous Wavelet Transform Analysis of Acceleration Signals Measured from a Wave Buoy

Accelerometers, which can be installed inside a floating platform on the sea, are among the most commonly used sensors for operational ocean wave measurements. To examine the non-stationary features of ocean waves, this study was conducted to derive a wavelet spectrum of ocean waves and to synthesize sea surface elevations from vertical acceleration signals of a wave buoy through the continuous wavelet transform theory. The short-time wave features can be revealed by simultaneously examining the wavelet spectrum and the synthetic sea surface elevations. The in situ wave signals were applied to verify the practicality of the wavelet-based algorithm. We confirm that the spectral leakage and the noise at very-low-frequency bins influenced the accuracies of the estimated wavelet spectrum and the synthetic sea surface elevations. The appropriate thresholds of these two factors were explored. To study the short-time wave features from the wave records, the acceleration signals recorded from an accelerometer inside a discus wave buoy are analysed. The results from the wavelet spectrum show the evidence of short-time nonlinear wave events. Our study also reveals that more surface profiles with higher vertical asymmetry can be found from short-time nonlinear wave with stronger harmonic spectral peak. Finally, we conclude that the algorithms of continuous wavelet transform are practical for revealing the short-time wave features of the buoy acceleration signals.

Bathymetry Determination From Marine Radar Image Sequences Using the Hilbert Transform

This letter presents an image processing technique based on the theory of the Hilbert transform to determine the coastal bathymetry from marine radar image sequences. Use of the Hilbert transform enables the difficulties and complications of inhomogeneous image analysis to be avoided. In addition, a number of steps and complex computations can be avoided using the numerical algorithm of the Hilbert transform. Because the Hilbert transform is suitable for only monocomponent signals, we first applied the image procedure to decompose the irregular wave patterns into different single wave period images. Analysis of the simulated wave field and radar image sequences demonstrated that there was a high correlation between the estimated depths and reference depths. The causes for errors in the bathymetry estimations are also discussed.

Detection of Thermal Effluent Discharge Using Sea Surface Nautical Radar Images

This letter presents the patterns of thermal effluent discharge determined by sea surface images that are acquired by nautical X-band radar. By averaging the sequential image series, we detect an obvious echo band pattern at the drain outlet of a power plant. The detection of microscale (less than 1 km) thermal effluent features using the low-grazing-angle nautical radar approach has received little attention up to now. The field sea surface temperature data confirmed that the echo band matches the spatial distribution of the thermal effluent discharge. By observing hourly image cases, we also confirm that this echo band moves approximately two times per day and that this oscillation is dominated by the tidal current.