As the LMS algorithm is a kind of noise gradient method, the convergence gain should be lessened gradually to ensure that the quadratic mean deviation of the steady state takes the quadratic mean deviation of the system as its limit.

For approximation,the average deviation of the regional mean as a threshold is proposed,if the regional mean deviation about the approximation coefficients for each sub band is higher than the threshold,a logical filter fused rule is employed,else a weighted average fused rule is applied and the weighted coefficients are given by the approximation coefficients.

The light scattering by a rough surface with random Gaussian fluctuations of roughness is studied in the case of coarse roughness, whose parameters-mean deviation and correlation length-are much greater than the radiation wavelength.

The mean deviation between the total atmospheric ozone content derived from TOVS data and that got from ground-based Dobson spectrophotometer observation is 3.67%.

An appreciable reduction of the mean deviation was achieved in all spectra considered.

To determine the false-negative response rate in patients with primary open-angle glaucoma (POAG) and its relationship with the mean deviation, we evaluated 286 visual fields of patients with POAG.

In addition, 38 CL wearers (79%) showed higher positive mean deviation values than 10 eyes (20.8%) with spectacles.

LMS algorithm is used widely in various adaptive filters designing. A lot of work has been done on it.The adequate convergence condition of this algorithm has been got. And the upper and lower bounds of the misadjustment in the algorithm for a constant convergence gain have been got. The quantitative study on the convergence gain is yet insufficient. Because of the statistical characteristic of input data of the weights in a LMS algorithm, more attention ...

LMS algorithm is used widely in various adaptive filters designing. A lot of work has been done on it.The adequate convergence condition of this algorithm has been got. And the upper and lower bounds of the misadjustment in the algorithm for a constant convergence gain have been got. The quantitative study on the convergence gain is yet insufficient. Because of the statistical characteristic of input data of the weights in a LMS algorithm, more attention should be paid to the recent historical calculation data. As the LMS algorithm is a kind of noise gradient method, the convergence gain should be lessened gradually to ensure that the quadratic mean deviation of the steady state takes the quadratic mean deviation of the system as its limit. According to this condition, a new method to estimate the optimum convergence gain is proposed. In the new method the increment of the weights vector in previous steps was rotated and placed to a plane , then the extended method for a one variable convex function was used. In this way, the misadjustment of the LMS algorithm can converge to zero quickly.

Fractal image coding is a very promising compression technique, in which an image is encoded by a contractive transformation whose fixed point is close to the original image, and then is decoded by using the iteration procedure stemmed from Banach fixed-point theorem. However, it has not been widely used because of its long encoding time and high computational complexity. A fast fractal encoding algorithm is thus proposed in this paper. The proposed algorithm uses an inequality linking the root-mean-square...

Fractal image coding is a very promising compression technique, in which an image is encoded by a contractive transformation whose fixed point is close to the original image, and then is decoded by using the iteration procedure stemmed from Banach fixed-point theorem. However, it has not been widely used because of its long encoding time and high computational complexity. A fast fractal encoding algorithm is thus proposed in this paper. The proposed algorithm uses an inequality linking the root-mean-square (RMS) and mean intensity deviation to convert the range-domain block matching problem to the nearest neighbors search problem in the sense of mean deviations. In detail, after the codebook blocks are sorted according to their mean deviations of intensities, the encoder uses the bisection search method to find out the best matched codebook block regarding to mean deviations of a given range block. Because the closeness of mean intensity deviations of two blocks cannot ensure their good approximations in the RMS sense, the encoder utilizes the inequality to again search for the best-matched block (in the RMS sense) in the nearest k-neighbor of the best-matched block (in the sense of mean deviations) to a given range block in order to further improve the image quality. The experimental results demonstrate that the encoding procedure is much faster than that of the baseline fractal algorithm, while it gives an insignificant degradation in the subjective quality.

The paper presents a region-fused image fusion algorithm,according to the different imagery characteristic between the SAR and multi-spectral images based on studying the lifting scheme wavelet transform.The details selected are based on maximum standard deviation of the detail coefficients for each sub band.For approximation,the average deviation of the regional mean as a threshold is proposed,if the regional mean deviation about the approximation coefficients for each sub band is higher than the threshold,a...

The paper presents a region-fused image fusion algorithm,according to the different imagery characteristic between the SAR and multi-spectral images based on studying the lifting scheme wavelet transform.The details selected are based on maximum standard deviation of the detail coefficients for each sub band.For approximation,the average deviation of the regional mean as a threshold is proposed,if the regional mean deviation about the approximation coefficients for each sub band is higher than the threshold,a logical filter fused rule is employed,else a weighted average fused rule is applied and the weighted coefficients are given by the approximation coefficients.The experiment result shows the algorithm effectively fused the complementary information of SAR and multi-spectral images and it is easy to be realized in DSP compared with other wavelet transform.