### Table 2. PSNR (dB) results for the wavelet-based method from [10] and for TV denoising.

"... In PAGE 4: ... For the Barbara image, the presence of strong texture makes the TV criterion less competitive. Table2 shows the results of TV and the method from [10]; in these results, TV is outperformed, at all noise levels and for all im- ages, by this state of the art method, which uses explicit models of inter-coefficient dependencies and sophisticated wavelet represen- tations (the steerable pyramid). Given the difference in complexity of the TV penalty and the model used in [10], this is not surprising.... ..."

### Table 2. PSNR (dB) results for the wavelet-based method from [10] and for TV denoising.

"... In PAGE 4: ... For the Barbara image, the presence of strong texture makes the TV criterion less competitive. Table2 shows the results of TV and the method from [10]; in these results, TV is outperformed, at all noise levels and for all im- ages, by this state of the art method, which uses explicit models of inter-coefficient dependencies and sophisticated wavelet represen- tations (the steerable pyramid). Given the difference in complexity of the TV penalty and the model used in [10], this is not surprising.... ..."

### Table 2. Comparison of the Image Quality of Reconstructed Images after Thresholding 90% of the Transform Coefficients for Several SAR Images

"... In PAGE 19: ... In this case, the image quality (as judged by several subjects with experience in viewing SAR imagery) is closely related to the mean-square error. Table2 summarizes the results of applying this thresholding and remapping process to several SAR images. In general, the image quality of the reconstructed images based on the thresholded wavelet and WP-Gabor transforms was significantly higher than that of the Gabor transform when 90% of the coefficients were set to zero.... ..."

### Table 5-5.1 Comparison of linear averaging to novel wavelet denoising for 512 data frames

"... In PAGE 15: ...ignal being estimated. The right side of Figure 5 shows the variance of the linear estimator (top) vs. the variance of the novel wavelet estimator (bottom), as a function of number of frames averaged. The numerical results corresponding to Figures 5 are in Table5 -1. The linear averaging and wavelet denoised variances were calculated and compared, as were the SNRs for the two approaches.... In PAGE 17: ... We also note from this figure that the CSTD algorithm results in reconstructed wave shapes with significantly better fidelity to the noise-free AMRL template than those obtained by simply averaging 256 frames. SNR improvements of the CSTD result versus linear averaging are listed in Table5 -2 for averages of 256 and 512 frames. Again, we can easily verify that linear averaging improves the SNR by the predictable amount given by Equation 3.... In PAGE 17: ...3. Table5 -2 Comparison of linear averaging to CSTD denoising for 256 and 512 AMLR 512 data frames Frames SNR noisy signal Linear SNR (dB) Denoise SNR (dB) SNR dB Improvement 256 -15 9.2 14.... In PAGE 18: ... We can see that the variance of this wavelet denoising algorithm is reduced substantially relative to linear averaging. Table5 -3 summarizes the results for these waveforms. Table 5-3 Denoised signal root-mean-square error (rms; i.... In PAGE 18: ... Table 5-3 summarizes the results for these waveforms. Table5 -3 Denoised signal root-mean-square error (rms; i.e.... ..."

### Table 1 presents results of a simulated image de-noising using selected wavelet functions, which give quite different wavelet coefficients because of their different definitions (analytic or numerical). Clearly, the wavelet transform provides then many ways for the wavelet decomposition of a given signal. The best result of given simulated image de-noising has been obtained by the Symmlet wavelet function of the 4th order. The second best result provides the Symmlet wavelet function of the 8th or-

"... In PAGE 3: ...873 7 Wavelet Symmlet8 24.713 Table1 : The peak-signal-to-noise-ratio (PSNR) of the simu- lated image with noise (PSNR1) and the same simulated im- age after de-noising (PSNR2) from the previous figure de- noised by the selected wavelet functions der. Criteria of an image de-noising quality include evaluation of peak-signal-to-noise-ratio (PSNR) value in dB.... ..."

### Table 1. Distribution model based on InSAR information

2000

"... In PAGE 6: ...nside the fault. The position of the interpolating point relative to the fault can be determined in the same way. The combination of open- and closed-curve models can deal with comparatively complex fault systems. After the classification of GPS stations and the intended interpolating points into different groups, a distribution model for interpolation for each group based on the GPS-corrected InSAR results is proposed, as illustrated in Table1 . In some sites of interest (Sites 1 to 4) there are both GPS and InSAR results for the deformation (the CGPS results may have to span one or more InSAR repeat cycles depending on the availability of suitable SAR image pairs).... ..."

Cited by 1

### Table 2 Denoising results (dB) using wavelets sym12 and coif4 with the Barbara image

"... In PAGE 15: ... We use the Barbara image to study the performance of the denoising algorithm under different noise levels. Table2 shows the PSNR (dB) results. We can see that as a result of using two wavelets and the last step of averaging, the PSNR of s12 is about 0.... In PAGE 15: ...ection 2.3.2 for discussion of this special case. Actually, results for images b1, b2 and b12 were obtained by running the proposed algorithm with the WHT and IWHT being by-passed. We can see from Table2 that the PSNR for b12 is also about 0.... In PAGE 19: ... We can see that the difference in PSNR is smaller as the noise level grows higher. We can also see that the results for each individual wavelet pair is very similar to that presented in Table2 . The general observation (stated in previous subsection) about the performance of proposed algorithm is still valid for a pair of wavelets originating from one wavelet.... ..."

### Table 1. Comparison of SAR and buoy wave parameters

"... In PAGE 7: ... Ten separate ERS- 1 SAR data sets were utilized, five for buoy 46001 and five for the ASREX Seatex buoy. Table1 lists the dates and time of each SAR pass. The environmental conditions for each date are detailed in Appendix A.... In PAGE 7: ... For the dominant peak wave component, the SAR spectrum compares reasonably well with the buoy results, however the SAR spectral shape is slightly distorted in the azimuth direction by the radar transfer function as compared to the model results. Referring back to Table1 , it can be seen that of the 10 cases, 8 comparisons between the SAR measurements and the buoy measurements are favorable, while 2 cases, October 15 and November 16, show considerable differences in wave length and direction. We.... In PAGE 8: ... For ERS- 1 the R/V ratio is about 115 s. As shown in Table1 , values of C near 2 or greater are obtained for the two nonlinear cases on October 15 and November 16, due to the comparatively short wave length and large Hs in both cases. The remaining 8 cases have values generally much less than 1.... In PAGE 16: ...A brief summary is provided of the weather and wave environment for each of the ten dates used in the buoy/SAR validation ( Table1 ) based on NMC surface charts, site descriptions (Galbraith et al., 1994), and buoy and wave model results.... In PAGE 23: ... Comparison of measurements of wave period from buoy 46001 and wave length from ERS- 1 SAR using the wave dispersion relation. See Table1 for details. 5.... In PAGE 26: ... ADIR = Difference in wave direction between buoy and SAR. Table1 Notes1). ERS- 1 flight (azimuth) direction is 196 T for all images.... ..."

### Table 3: Denoising of Lena image via wavelet shrinkage.

1999

"... In PAGE 22: ... As in the 1-D case, the depth of the cascade was chosen to be J = 4 for GHM with approximation and J = 5 for GHM with repeated row and D4. The experimental results are shown in Table3 and in Figure 12. Multiwavelet schemes were superior to D4 both numerically and subjectively.... ..."

Cited by 42

### Table 1: The peak-signal-to-noise-ratio (PSNR) of the simu- lated image with noise (PSNR1) and the same simulated im- age after de-noising (PSNR2) from the previous figure de- noised by the selected wavelet functions

"... In PAGE 2: ... 3 . Table1 presents results of a simulated image de-noising using selected wavelet functions, which give quite different wavelet coefficients because of their different definitions (analytic or numerical). Clearly, the wavelet transform provides then many ways for the wavelet decomposition of a given signal.... ..."