Initial Model Parameters Research Articles

Real-Time Landmine Detection with Ground-Penetrating Radar Using Discriminative and Adaptive Hidden Markov Models

We propose a real-time software system for landmine detection using ground-penetrating radar (GPR). The system includes an efficient and adaptive preprocessing component; a hidden Markov model- (HMM-) based detector; a corrective training component; and an incremental update of the background model. The preprocessing is based on frequency-domain processing and performs ground-level alignment and background removal. The HMM detector is an improvement of a previously proposed system (baseline). It includes additional pre- and postprocessing steps to improve the time efficiency and enable real-time application. The corrective training component is used to adjust the initial model parameters to minimize the number of misclassification sequences. This component could be used offline, or online through feedback to adapt an initial model to specific sites and environments. The background update component adjusts the parameters of the background model to adapt it to each lane during testing. The proposed software system is applied to data acquired from three outdoor test sites at different geographic locations, using a state-of-the-art array GPR prototype. The first collection was used as training, and the other two (contain data from more than 1200 m of simulated dirt and gravel roads) for testing. Our results indicate that, on average, the corrective training can improve the performance by about 10% for each site. For individual lanes, the performance gain can reach 50%.

Localized Parametric Electromagnetic Inversion for Pavement Profiling with Ground Penetrating Radar

Ground penetrating radar (GPR) has proven to be an effective non-destructive tool for pavement profiling. Layer-stripping algorithm and parametric electromagnetic (EM) inversion algorithm (referred to as EMI algorithm) are two inverse scattering approaches to estimate multilayered media properties using monostatic GPR. In this paper, a localized parametric EM inversion approach (referred to as NHD-EMI algorithm) is proposed for analyzing pavements with spatial varying characteristics. A non-homogeneous detector (NHD) is first used to divide the whole data into multiple subsegments with similar media properties. Then EMI is applied to each segment with layer-stripping technique to provide the initial model parameter estimates needed by the EMI algorithm. Numerical results show that the proposed NHD-EMI algorithm exhibits better performance over the existing EMI algorithm for nonhomogeneous pavements.

A genetic classification method for speaker recognition

Gaussian mixture model (GMM) has been widely used for modeling speakers. In speaker identification, one major problem is how to generate a set of GMMs for identification purposes based upon the training data. Due to the hill-climbing characteristic of the maximum likelihood (ML) method, any arbitrary estimate of the initial model parameters will usually lead to a sub-optimal model in practice. To resolve this problem, this paper proposes a hybrid training method based on genetic algorithm (GA). It utilizes the global searching capability of GA and combines the effectiveness of the ML method. Experimental results based on TI46 and TIMIT showed that this hybrid GA could obtain more optimized GMMs and better results than the simple GA and the traditional ML method.

Segmentation of 4D cardiac MR images using a probabilistic atlas and the EM algorithm

In this paper an automatic atlas-based segmentation algorithm for 4D cardiac MR images is proposed. The algorithm is based on the 4D extension of the expectation maximisation (EM) algorithm. The EM algorithm uses a 4D probabilistic cardiac atlas to estimate the initial model parameters and to integrate a priori information into the classification process. The probabilistic cardiac atlas has been constructed from the manual segmentations of 3D cardiac image sequences of 14 healthy volunteers. It provides space and time-varying probability maps for the left and right ventricles, the myocardium, and background structures such as the liver, stomach, lungs and skin. In addition to using the probabilistic cardiac atlas as a priori information, the segmentation algorithm incorporates spatial and temporal contextual information by using 4D Markov Random Fields. After the classification, the largest connected component of each structure is extracted using a global connectivity filter which improves the results significantly, especially for the myocardium. Validation against manual segmentations and computation of the correlation between manual and automatic segmentation on 249 3D volumes were calculated. We used the 'leave one out' test where the image set to be segmented was not used in the construction of its corresponding atlas. Results show that the procedure can successfully segment the left ventricle (LV) ( r=0.96), myocardium ( r=0.92) and right ventricle ( r=0.92). In addition, 4D images from 10 patients with hypertrophic cardiomyopathy were also manually and automatically segmented yielding a good correlation in the volumes of the LV ( r=0.93) and myocardium (0.94) when the atlas constructed with volunteers is blurred.

Application of TOPNET in the distributed model intercomparison project

This paper describes the application of a networked version of TOPMODEL, TOPNET, as part of the Distributed Model Intercomparison Project (DMIP). The model implementation is based on a topographically derived river network with spatially distributed sub-basins draining to each network reach. The river network is mapped from the US National Elevation Dataset Digital Elevation Model (DEM) using procedures that objectively estimate drainage density from geomorphic principles. Rainfall inputs are derived from NEXRAD (radar) for each sub-basin. For each sub-basin, the wetness index distribution is derived from the DEM. The initial model parameters for each sub-basin are estimated using look up tables based on soils (STATSGO) and vegetation (1-km AVHRR). These initial model parameters provide the spatially distributed pattern of parameters at the scale of each sub-basin. Calibration uses a multiplier for each parameter to adjust the parameters while retaining the relative spatial pattern obtained from the soils and vegetation data. Parameter multipliers were calibrated using the shuffled complex evolution algorithm [J. Optim. Theory Appl. 61 (1993)] with the objective to minimize the mean square error between observed and modeled hourly streamflows. We describe the model and calibrated results submitted for all basins for the time periods involved in the DMIP study. We were encouraged by the relatively good performance of the model, especially in comparison to streamflow from smaller interior watersheds not used in calibration and simulated as ungaged basins. The limited resources used to achieve these results show some of the potential for distributed models to be useful operationally.

Theoretical analyses and numerical experiments of variational assimilation for one-dimensional ocean temperature model with techniques in inverse problems

In the present work, the data assimilation problem in meteorology and physical oceanography is re-examined using the variational optimal control approaches in combination with regularization techniques in inverse problem. Here the estimations of the initial condition, boundary condition and model parameters are performed simultaneously in the framework of variational data assimilation. To overcome the difficulty of ill-posedness, especially for the model parameters distributed in space and time, an additional term is added into the cost functional as a stabilized functional. Numerical experiments show that even with noisy observations the initial conditions and model parameters are recovered to an acceptable degree of accuracy.

Frequency Domain Finite Element Inversion of Elastic Wave Velocity Using the Geologic Regular‐Block‐Model Method

AbstractIn order to obtain the elastic velocity of subsurface media directly, we propose a frequency domain elastic‐wave finite element least squares inversion method according to the criterion of minimizing the misfit function (the l2‐norm of the data residuals between observed and froward modeled data) on the basis of the frequency domain elastic wave finite element modeling equation. This inversion method includes the matrix compressional storage and LU decomposition techniques to save and solve the large sparse complex‐valued coefficient matrix in frequency domain modeling equation and the Levenberg‐Marquard technique with an adjustable damping term to solve the inverse equation group. To exploit subsurface geologic distribution features to decrease the number of inverse unknowns, we suggest a regular‐block‐model method for the inversion. Being verified by numerical models, this inversion method can obtain satisfied results using the shot profile contaminated with great random noise even up to 50% or the initial model parameters that have a great difference from the model. It proves that the inversion is noise‐resistable and robust.

Direct and adjoint sensitivity analysis of chemical kinetic systems with KPP: Part I—theory and software tools

The analysis of comprehensive chemical reactions mechanisms, parameter estimation techniques, and variational chemical data assimilation applications require the development of efficient sensitivity methods for chemical kinetics systems. The new release (KPP-1.2) of the kinetic preprocessor (KPP) contains software tools that facilitate direct and adjoint sensitivity analysis. The direct-decoupled method, built using BDF formulas, has been the method of choice for direct sensitivity studies. In this work, we extend the direct-decoupled approach to Rosenbrock stiff integration methods. The need for Jacobian derivatives prevented Rosenbrock methods to be used extensively in direct sensitivity calculations; however, the new automatic and symbolic differentiation technologies make the computation of these derivatives feasible. The direct-decoupled method is known to be efficient for computing the sensitivities of a large number of output parameters with respect to a small number of input parameters. The adjoint modeling is presented as an efficient tool to evaluate the sensitivity of a scalar response function with respect to the initial conditions and model parameters. In addition, sensitivity with respect to time-dependent model parameters may be obtained through a single backward integration of the adjoint model. KPP software may be used to completely generate the continuous and discrete adjoint models taking full advantage of the sparsity of the chemical mechanism. Flexible direct-decoupled and adjoint sensitivity code implementations are achieved with minimal user intervention. In a companion paper, we present an extensive set of numerical experiments that validate the KPP software tools for several direct/adjoint sensitivity applications, and demonstrate the efficiency of KPP-generated sensitivity code implementations.

Restoration of structure of thin multilayer films from reflectivity data by stepwise model refinement

The computational experiment on the restoration of the model parameters of the cross-section profile of the scattering density of thin multilayer films is described. It is shown that the use of the stepwise model of the structure, which is described by the thickness, density, absorption, and roughness parameters makes the problem of profile restoration multimodal and ill-conditioned. The analysis of the numerical experiments performed allowed us to propose a procedure of consecutive refinement of the initial-model parameters using the results of the local minimization of the discrepancy functional applied for the characteristic segments of the reflectivity curve. The methods proposed were used to restore the structure of a multilayer film deposited on a silicon substrate by the Langmuir-Blodgett method.

Experimental Modeling of Monolithic Resistors for Silicon ICS with a Robust Optimizer-Driving Scheme

Today, an exhaustive library of models describing the electrical behavior of integrated passive components in the radio-frequency range is essential for the simulation and optimization of complex circuits. In this work, a preliminary study has been done on Tantalum Nitride (TaN) resistors integrated on silicon, and this leads to a single p-type lumped-element circuit. An efficient extraction technique will be presented to provide a computer-driven optimizer with relevant initial model parameter values (the "guess-timate"). The results show the unicity in most cases of the lumped element determination, which leads to a precise simulation of self-resonant frequencies.

Extended quintessence and its late-time domination

Various astronomical observations point towards the evidence for dark energy. One of the most mysterious problem is the coincidence problem: why dark energy becomes dominant only recently. We present a scenario based on extended quintessence models to explain the late-time domination of dark energy without severe fine-tuning of initial conditions and model parameters.

Self-organized critical behavior in a deforming metal

It has been shown that the transport of mobile dislocations through blocking dislocation walls in a deforming metal can be treated by a simple percolation theory. Two different mechanisms for strain propagation are proposed in the strain percolation model. In the first case, the strain propagates between adjacent dislocation cells by activation of sources within the walls. In the second case, as an additional mechanism, unstable locks can be unzipped by a nearby dislocation pileup which can lead to a large localized strain. Previous simulations have shown that both cases belong to the same universality class as standard percolation. Further extensive simulations of the model have been performed to understand how the geometrical aspects of a strained percolating cluster are related to the strain itself. In our case, the strain is an additional variable not present in standard percolation theory. We find that the total strain and the mean strain per strained cell show power-law behavior in the critical regime, and obtain a scaling function which explains its critical behavior. Other percolation and critical aspects of the model are also discussed in terms of the initial strain, correlation length (which is a characteristic length scale), and model parameters.

A habitat-based population model for the Great Lakes population of the piping plover ( Charadrius melodus)

We created a stochastic habitat-based population model to compare the relative effectiveness of potential conservation strategies to increase the endangered Great Lakes population of piping plovers. Initial model parameters were based on an extensive 14-year dataset obtained by annual studies of breeding pairs in Michigan and 6 years of observations of color-marked individuals. Cumulative persistence probability curves and population trends were compared for scenarios with all possible combinations of the following: (1) empirical, medium and high site specific probabilities of reproductive success and (2) empirical, medium and high numbers of available breeding territories. Sites were also categorized according to land ownership to determine if long-term preservation of publicly owned lands is sufficient for population recovery under any of the scenarios. Only scenarios in which both reproduction and habitat availability were high and all ownership classes were included resulted in a model population that persisted for 100 years and was likely to reach the US Fish and Wildlife Service current population recovery goal (100 breeding pairs) for Michigan. The model was most sensitive to survival rates of adults and more sensitive to changes in reproductive parameters than to number of breeding sites or territories. Model results suggest that intensive measures to increase productivity alone are not sufficient for population recovery; increases in nesting densities or total number of available territories are also likely necessary. Given uncertainties about nesting densities possible at specific breeding sites, preservation of presently unoccupied habitat throughout the Great Lakes region and management of this habitat to encourage breeding by plovers are recommended. Use of techniques to augment the breeding population in conjunction with protection of habitat and reproduction may be warranted.

Validation of Partial Tissue Segmentation of Single-Channel Magnetic Resonance Images of the Brain

We describe and evaluate a practical, automated algorithm based on local statistical mixture modeling for segmenting single-channel, T1-weighted volumetric magnetic resonance images of the brain into gray matter, white matter, and cerebrospinal fluid. We employed a stereological sampling method to assess, prospectively, the performance of the method with respect to human experts on 10 normal T1-weighted brain scans acquired with a three-dimensional gradient echo pulse sequence. The overall kappa statistic for the concordance of the algorithm with the human experts was 0.806, while that among raters, excluding the algorithm, was 0.802. The algorithm had better agreement with the modal expert decision (κ = 0.878). The algorithm could not be distinguished from the experts by this measure. We also validated the algorithm on a simulated MR scan of a digital brain phantom with known tissue composition. Global gray matter and white matter errors were 1% and <1%, respectively, and correlation coefficients with the underlying tissue model were 0.95 for gray matter, 0.98 for white matter, and 0.95 for cerebrospinal fluid. In both approaches to validation, we evaluated both local and global performance of the algorithm. Human experts generated slightly higher global gray matter proportion estimates on the test brain scans relative to the algorithm (3.7%) and on the simulated MR scan relative to the true tissue model (4.4%). The algorithm underestimated gray in some subcortical nuclei which contain admixed gray and white matter. We demonstrate the reliability of the method on individual 1 NEX data sets of the test subjects, and its insensitivity to the precise values of initial model parameters. The output of this algorithm is suitable for quantifying cerebral cortical tissue, using a commonly performed commercial pulse sequence.

Data assimilation by an intermediate coupled ocean‐atmosphere model: Application to the 1997–1998 El Niño

Sea surface temperature, sea level, and pseudo wind stress anomaly data from late 1996 to early 1998 are assimilated into an intermediate coupled model of the Tropical Pacific. Model data consistency is examined. Impact of the assimilation on forecast is evaluated. The ocean component of the coupled model consists of a shallow water model with two baroclinic modes, an Ekman shear layer, and a mixed layer temperature equation. The atmospheric model is a statistical one (based on dominant covariance of historical surface temperature and pseudo wind stress anomaly data). The adjoint method is used to fit the coupled model to 6 months of data by optimally adjusting the initial state and model parameters. A forecast is performed using the end state of an assimilation experiment as initial conditions and using parameters estimated during the assimilation period. Thus the model state during the assimilation and that during the forecast belong to the same model trajectory in different periods. Such an initialization procedure is useful in avoiding initial shock during forecast due to inconsistency of an initial state with the coupled model physics. As a result of optimal adjustments of initial state and parameters, the model is able to reproduce observed interannual variability of sea surface temperature and sea level reasonably well. The averaged residual model data misfits over various 6 month periods are 0.5°C and 5 cm, respectively. The model has a limited skill in reproducing much of the off‐equatorial wind anomalies. The residual model data misfit in pseudo wind stress anomaly is larger than 10 m2 s−2. Forecasts initialized from the assimilation product are overall more realistic than those simply initialized from wind‐forced ocean states. Consistent improvement due to optimal initialization is found for sea surface temperature and sea level anomalies in the central‐eastern Pacific and zonal pseudo wind stress anomaly in the central Pacific, both in terms of root‐mean‐squared deviation from and correlation with the data. The adjustments of parameters in addition to initial state in a coupled context is found to be important to improving the model data consistency during the assimilation and the forecast. In particular, the estimated drag and damping coefficients properly regulate the relative strength of forcing and damping of the ocean state so as to fit the three types of observations during the assimilation (initialization) period, which facilitates the development of a large‐amplitude warming event during the forecast. The study demonstrates the utility of oceanic and atmospheric data to estimate initial state and model parameters in a coupled context, which is useful to the evaluation, improvement, and initialization of El Niño‐Southern Oscillation forecast models.

Three-dimensional inversion of gravity data from the Main Ethiopian Rift

The crustal thickness distribution beneath the Main Ethiopian Rift is determined using a 3-D inverse approach to gravity data interpretation. The depths to the crust-upper mantle interface form the inversion parameters. These initial model parameters are constrained with the results of the seismic refraction experiment of the region. The degree of ambiguity of the final model parameters is then quantified using a Q-mode factor analysis approach. The results of the modelling indicate a strong crustal attenuation zone (≤31 km) closely associated with the rifting of the graben and an abrupt fall of the Moho interface on either side of the rift (51 km) related to the formation of the western and southeastern plateaus. However, no indication of crustal separation is observed. The semi-quantitative ambiguity analysis reveals the association of the model parameters from the southeastern plateau with the strong ambiguity region. These model parameters represent the depths to the Moho interface where the seismic control is relatively less.

Fractal convergence properties of geophysical inversion

Geophysical data, such as measurements of the Earth's gravitational or magnetic field, are routinely collected and studied to get information on the structure of the subsurface geology. The standard method of data analysis involves the least-squares inversion of the data with respect to various user-chosen model parameters (such as the geometry or density of the lithology). One serious problem with most inversion schemes is that they are liable to converge to local minima – that is they reach a set of model parameters whose geophysical response is a better fit to the observed data than the starting model, but they do not reach the set of parameters that would provide the best fit possible. The set of initial model parameters that converge to a particular minima of the misfit surface is studied here for some magnetic models, and is found to be a fractal when there are two minima available for the model to converge to, and at least two model parameters are inverted. The fractal dimension of the set is shown to be inversely proportional to the damping of the inversion process.

Three-Dimensional Soil Resistivity Inversion Using Patching Method.

To apply Electrical Resistivity Tomography (ERT) in monitoring temporal and spatial variation of soil water, by using finite-element method as forward modeling, we proposed a patching algorithm of inversely calculating 3D soil resistivity in this paper. From finite-element forward modeling, sensitivity calculation to formation of initial model parameters, we systematically described the inversion calculation process of using patching algorithm. At the end, we gave out results about two numerical experiments and one field measured data inversion to illustrate the effectiveness of the algorithm. By patching algorithm, the initial soil resistivity model is step by step modified to make the calculated potential approximate that measured. Each measurement event (group of pole-pole array measurements which have same current electrode and input current) composes a step of modification to the model. This algorithm is appropriate when the data number for one measurement event is quite less than the parameter number of the model and the measurement event number varies with the equipment capacity or prospecting objectives, and even when the resistivity changes with time. Numerical experiments and field measured data inversion indicated the effectiveness of the inverse algorithm, but also revealed the difficulty of completely inversely calculating out the 3D subsurface soil resistivity only by using the data measured at the soil surface.

Frequency–wavenumber modelling and migration of 2D GPR data in moderately heterogeneous dispersive media

An algorithm for modelling and migrating ground penetrating radar (GPR) data in moderately heterogeneous dispersive media is presented. The method is based on wavefield extrapolation in the frequency–wavenumber (f–k) domain, from the solution of the 2D Maxwell's equations. The wavefield is extrapolated by a phase‐shift technique using a constant relative permittivity K and a quality factor Q. It is then modified by a correction term to handle the lateral K and Q variations. The spatial distribution of the K and Q‐factor values, representing the given model parameters, is introduced into the algorithm by a regular grid parametrization. The radar wave dispersion and attenuation, induced by relaxation processes, are taken into account by a linear frequency‐dependent Q model, and expressed by a complex wavenumber in the propagation equation.A synthetic case and a field data set illustrate the potential of the method for frequencies of 300, 500 and 900 MHz. In the first case, a typical civil engineering problem is considered. The frequency dependence of the wave velocity and attenuation is well illustrated. The synthetic data are afterwards migrated using the initial model parameters. The results show the importance of using spatially varying model parameters in the migration processes. The second case concerns an application of the method to a real data set. In order to adjust the model parameters, a forward modelling sequence is performed until the best match between the measured and the synthetic data is achieved. A depth migration is then applied to the data, and the result is compared with the initial model parameters.In conclusion, we assess the contributions of the method to industrial applications, by discussing the performance of the algorithm compared with its limitations.

Geoelectrical soundings and their relationship to hydraulic parameters in semiarid regions of Jalore, northwestern India

Geoelectrical soundings using the Schlumberger configuration were carried out in the close vicinity of twelve pumping test sites in the Jalore district of Rajasthan, India, to relate geoelectric and hydraulic parameters. Geoelectrical sounding data were interpreted by partial curve matching technique to obtain the initial model parameters. This initial starter model was then used to obtain the final layer parameters through inversion technique. From the inversion technique, the best-fitting, lower bound (minimum) and upper bound (maximum) models were derived under 1% error level. The best-fitted models were correlated with the litholog. The resistivities for most of the aquifer zones are found to be very low which may be due to deteriorating quality of groundwater besides their packing and grain size. The resistivities of collected water samples were determined to compute the formation factor and porosity. Computed formation factor and porosity were correlated with hydraulic conductivity. The formation factor is directly related to hydraulic conductivity whereas porosity is inversely related. The normalised aquifer resistivity was correlated with hydraulic conductivity and normalised transverse resistance with aquifer transmissivity. The hydraulic conductivity is found to be closely related with normalised resistivity and the transmissivity with normalised transverse resistance.