Multiple transformation function estimation for image enhancement

Most deep learning-based image enhancement algorithms have been developed based on the image-to-image translation approach, in which enhancement processes are difficult to interpret. In this paper, we propose a novel interpretable image enhancement algorithm that estimates multiple transformation functions to describe complex color mapping. First, we develop a histogram-based multiple transformation function estimation network (HMTF-Net) to estimate multiple transformation functions by exploiting both the spatial and statistical information of the input images. Second, we estimate pixel-wise weight maps, which indicate the contribution of each transformation function at each pixel, based on the local structures of the input image and the transformed images obtained by each transformation function. Finally, we obtain the enhanced image as the weighted sum of the transformed images using the estimated weight maps. Extensive experiments confirm the effectiveness of the proposed approach and demonstrate that the proposed algorithm outperforms state-of-the-art image enhancement algorithms for different image enhancement tasks.     

Bayes estimation of the extreme-value reliability function

The authors obtain Bayes estimates of the reliability function of the extreme value distribution by using two Bayes approximation procedures: Lindley (1980), and Tierney and Kadane (1986). These estimates were compared to maximum-likelihood estimates (MLE) based on a Monte Carlo simulation study. Jeffreys invariant prior was used in the comparison for both Bayes procedures. The MLE are superior to either of the Bayes estimates, except for small values of t. The simpler Lindley Bayes procedure gives estimates with smaller root-mean-square error than estimates obtained by the Tierney and Kadane procedure except for large values of t. From a practical standpoint, the ML method is easiest to use and more accurate for the extreme value distribution than the two Bayes approximation procedures. Both Bayes procedures seem to perform equally. However, the Lindley method is easier to use with little loss of accuracy     

Rapid estimation of the range-Doppler scattering function

Under wide sense stationary uncorrelated scattering (WSSUS) conditions, the signal spreading due to a random channel may be described by the scattering function (SF). In an active acoustic system, the received signal is modeled as the superposition of delayed and Doppler spread replicas of the transmitted waveform. The SF completely describes the second-order statistics of a WSSUS channel and can be considered a density function that characterizes the average spread in delay and Doppler experienced by an input signal as it passes through the channel. The SF and its measurement will be reviewed. An estimator is proposed based on a two-dimensional (2-D) autoregressive (AR) model for the scattering function. In order to implement this estimator, we derive the conditional minimum variance unbiased estimator of the time-varying frequency response of a linear channel. Unlike conventional Fourier methods, the AR approach does not suffer from the usual convolutional smoothing due to the signal ambiguity function. Simulation results are given.     

Matching criterion function for block motion estimation: blockfeature matching function

The authors present a new block feature matching (BFM) criterion function that can be used for block estimation of H.261, H.263, MPEG1, MPEG2, and HDTV codecs. Simulation indicates that the proposed BFM function is very simple and effective, thus the corresponding complexity of block matching computation and data transfer time are greatly reduced. The BFM function can be combined with any fast search algorithm of motion estimation     

Fatigue estimation with a multivariable myoelectric mapping function

A novel approach to muscle fatigue assessment is proposed. A function is used to map multiple myoelectric parameters representing segments of myoelectric data to a fatigue estimate for that segment. An artificial neural network is used to tune the mapping function and time-domain features are used as inputs. Two fatigue tests were conducted on five participants in each of static, cyclic and random conditions. The function was tuned with one data set and tested on the other. Performance was evaluated based on a signal to noise metric which compared variability due to fatigue factors with variability due to nonfatiguing factors. Signal to noise ratios for the mapping function ranged from 7.89 under random conditions to 9.69 under static conditions compared to 3.34-6.74 for mean frequency and 2.12-2.63 for instantaneous mean frequency indicating that the mapping function tracks the myoelectric manifestations of fatigue better than either mean frequency or instantaneous mean frequency under all three contraction conditions.     

Probability density function estimation using the MinMax measure

The problem of initial probability assignment which is consistent with the available information about a probabilistic system is called a direct problem. E.T. Jaynes' (1957) maximum entropy principle (MaxEnt) provides a method for solving direct problems when the available information is in the form of moment constraints. On the other hand, given a probability distribution, the problem of finding a set of constraints which makes the given distribution a maximum entropy distribution is called an inverse problem. A method based on the MinMax measure to solve the above inverse problem is presented. The MinMax measure of information, defined by Kapur, Baciu and Kesavan (1995), is a quantitative measure of the information contained in a given set of moment constraints. It is based on both maximum and minimum entropy. Computational issues in the determination of the MinMax measure arising from the complexity in arriving at minimum entropy probability distributions (MinEPD) are discussed. The method to solve inverse problems using the MinMax measure is illustrated by solving the problem of estimating a probability density function of a random variable based on sample data     

Bayesian estimation of reliability function and hazard rate

The Bayes estimates of reliability function and hazard rate function of the finite range failure model have been developed based on life tests that are terminated at a preassigned time point or after a certain number of failures have occurred, taking the order of observations into consideration. For the prior distribution of the parameter involved, the uniform, exponential and inverted gamma densities have been considered. As an example, failure data for a V600 indicator tube used in aircraft radar sets, which fit well the finite range failure model, have been considered as the current distribution for obtaining the Bayes estimates of the reliability function.     

Detection and estimation using an adaptive rational function filter

Proposes a new nonlinear adaptive filter structure based on rational functions. There are several advantages to the use of this filter. First, it is a universal approximator and a good extrapolator. Second, it ran be trained by a linear adaptive algorithm, which makes it suitable for real-time adaptive signal processing. Third, it has a best approximation for a specified function. To demonstrate its utility as a tool for solving adaptive signal processing problems, the authors apply the adaptive rational function filter to the problem of estimation and detection. The estimation problem pertains to the direction of arrival (DOA) estimation problem in array signal processing. For the detection problem, the authors consider the detection of a weak radar target (a small piece of ice) in an ocean environment     

Nonparametric regression estimation by normalized radial basis function networks

This paper establishes weak and strong universal consistency of regression estimates based on normalized radial basis function networks when the network parameters are chosen by empirical risk minimization.     

Dictionary-based stochastic expectation-maximization for SAR amplitude probability density function estimation

In remotely sensed data analysis, a crucial problem is represented by the need to develop accurate models for the statistics of the pixel intensities. This paper deals with the problem of probability density function (pdf) estimation in the context of synthetic aperture radar (SAR) amplitude data analysis. Several theoretical and heuristic models for the pdfs of SAR data have been proposed in the literature, which have been proved to be effective for different land-cover typologies, thus making the choice of a single optimal parametric pdf a hard task, especially when dealing with heterogeneous SAR data. In this paper, an innovative estimation algorithm is described, which faces such a problem by adopting a finite mixture model for the amplitude pdf, with mixture components belonging to a given dictionary of SAR-specific pdfs. The proposed method automatically integrates the procedures of selection of the optimal model for each component, of parameter estimation, and of optimization of the number of components by combining the stochastic expectation-maximization iterative methodology with the recently developed "method-of-log-cumulants" for parametric pdf estimation in the case of nonnegative random variables. Experimental results on several real SAR images are reported, showing that the proposed method accurately models the statistics of SAR amplitude data.     

Efficient estimation of a normalised correlation function from discrete-time samples

A novel method is presented for estimating a normalised correlation function from discrete-time observations. The performance of the method approaches that of a maximum likelihood estimator. Because multiplication and division are replaced by logarithmic and hyperbolic tangent conversions, the method is well suited to hardware implementation.<>     

Parameter estimation for micro-Doppler signals based on cubic phase function

An estimator for evaluating the parameters from the radar returned multicomponent micro-Doppler (m-D) signals is presented in this paper. While time frequency distribution (TFD) is commonly used to analyze the time-varying m-D frequency features in TF domain, the proposed algorithm is based on cubic phase function (CPF) that can transform the signal to time frequency rate domain. In order to estimate the parameters of multicomponent m-D signal, the extended Hough transform (HT) of CPF is employed to estimate linear frequency modulation (LFM) or sinusoidal frequency modulation (SFM) components. For the m-D signal composed of both LFM and SFM components, the estimates involve two steps of HT-CPF. Firstly, LFM components are estimated by HT-CPF and removed from the time frequency rate plane, and then, HT of the modified time frequency rate distribution is applied to estimate SFM ones. Compared with HT-TFD, this algorithm needs less dimension of HT space and is thus computationally efficient. In addition, simulations show that the algorithm has almost the same performance signal-to-noise threshold as HT of Wigner–Ville distribution method.     

基于模糊函数的AIS信号码元序列估计算法研究

针对获取船舶状态信息的需求,研究了AIS信号码元序列估计技术,提出了一种新的AIS信号码元序列估计算法。利用AIS信号的先验信息构造了AIS样本信号,采用雷达信号处理中模糊函数的思想,获得AIS信号的起始时刻、频偏、幅度等参数进行有效估计,进而估计出AIS信号的码元序列。该方法不需要对AIS信号进行解调。算法的有效性通过对实际接收到的AIS信号数据进行处理得到了验证。     

Partition function estimation of Gibbs random field images usingMonte Carlo simulations

A Monte Carlo simulation technique for estimating the partition function of a general Gibbs random field image is proposed. By expressing the partition function as an expectation, an importance sampling approach for estimating it using Monte Carlo simulations is developed. As expected, the resulting estimators are unbiased and consistent. Computations can be performed iteratively by using simple Monte Carlo algorithms with remarkable success, as demonstrated by simulations. The work concentrates on binary, second-order Gibbs random fields defined on a rectangular lattice. However, the proposed methods can be easily extended to more general Gibbs random fields. Their potential contribution to optimal parameter estimation and hypothesis testing problems for general Gibbs random field images using a likelihood approach is anticipated     

Bayesian and non-Bayesian estimation for the pareto parameters using quasi-likelihood function

The quasi-likelihood function is used to derive Bayesian and non-Bayesian estimates for the unknown parameters of the Pareto distribution. In the Bayesian technique the likelihood function is replaced with the quasi-likelihood function when the mean-variance relationship is defined. The efficiency of the quasi-likelihood estimation relative to the maximum likelihood is calculated. Numerical illustrations for the Bayesian and quasi-Bayesian procedures are given.     

Edge operator error estimation incorporating measurements of CCD TVcamera transfer function

A method for measuring the two-dimensional modulation transfer function of an image acquisition system with a CCD TV camera input device is presented. The measurements are then used in the estimation of the image edge operator performance. A high resolution edge profile is assembled after the analysis of many profiles from the image of a step. Modulation transfer function measurements beyond the system's Nyquist limit have been obtained. A minimum of apparatus is required. The performances of several systems have been compared. The band limiting of spatial frequencies are considered with respect to signal aliasing, matching components, feature representation and image operator accuracy. The results indicate the efficiency of the system aliasing and reconstruction filters, differences in the vertical and horizontal responses of CCD arrays, and that the rectangular CCD element shape has not deflected the shape of the two dimensional modulation transfer function. Various edge operators have been tested and compared using high resolution edge spread function data computed during the transfer function measurements. These data were from realistically smoothed edges and represent the most difficult edges that the practical system would need to detect. Significant differences were found between results using the traditional pixel average and the new step edge models, indicating that in some cases a simpler, more computationally efficient operator may be sufficient     

A Bayesian estimation of reliability model using the linex loss function

In this paper the problem of Bayes estimation of the reliability and the shape parameter p of a finite range failure time model is considered (assuming scale parameter θ is known). Following Zellner [A. Zellner, J. Am. Statist. Assoc. 81, 446–451 (1986)] the asymmetric loss function is used to obtain the Bayes estimators. Efficiencies of the proposed Bayes estimators are obtained with respect to the ordinary Bayes estimators and it was found that the proposed Bayes estimators are better than the ordinary Bayes estimators for quite a wide range of parameters.     

Simultaneous estimation of parameters for different orders ofdiscrete orthonormal function models

The augmented UD identification (AUDI) method is used to simultaneously estimate parameters of all 1 to Nth-order discrete orthonormal function models in one computational step. This method is tested on different types of orthonormal functions such as the Laguerre, Kautz, FIR, and Markov-Laguerre models     

Correlation function estimation by a polarity method using stochastic reference signals

Instead of estimating correlation functions by conventional means, a specific polarity scheme may be used on bounded processes. The method is based on a rather simple relationship between the correlation functions before and after infinite clipping, provided that stochastic reference signals of uniform distribution are added to the process. This correlation technique has been known for some time. Because of the apparent computational advantages, its application to the estimation of correlation functions from discrete or sampled data is being examined. A general derivation of the appropriate moment relationship is given and a complete mean-square error analysis of estimates is provided under the assumption of white-noise-type reference signals. It is shown that correlation function estimates obtained by this polarity method possess a mean-square error that differs from the error of conventional estimates only by a term proportional to1/N, whereNis the sample size. This term may be made arbitrarily small. Thus, only small degradations in the accuracy of estimates have to be expected when using the polarity approach.     

Systematic approach to MUSIC estimation with minimisation of its cost function

Based on the minimisation of a cost function, a systematic approach to MUSIC estimation is proposed for the arrival angles of the signals impinging on a sensor array. The proposed method, which employs Brent's method for the task together with bisection bracketing, can significantly reduce the computational burden compared to the conventional algorithm for computing the whole spatial spectrum