Image texture synthesis-by-analysis using moving-average models

A synthesis-by-analysis model for texture replication or simulation is presented. This model can closely replicate a given textured image or produce another image that although distinct from the original, has the same general visual characteristics and the same first and second-order gray-level statistics as the original image. The texture synthesis algorithm, proposed contains three distinct components: a moving-average (MA) filter, a filter excitation function, and a gray-level histogram. The analysis portion of the texture synthesis algorithm derives the three from a given image. The synthesis portion convolves the MA filter kernel with the excitation function, adds noise, and modifies the histogram of the result. The advantages of this texture model over others include conceptually and computationally simple and robust parameter estimation, inherent stability, parsimony in the number of parameters, and synthesis through convolution. The authors describe a procedure for deriving the correct MA kernel using a signal enhancement algorithm, demonstrate the effectiveness of the model by using it to mimic several diverse textured images, discuss its applicability to the problem of infrared background simulation, and include detailed algorithms for the implementation of the model     

ARMA Time Series Modeling: an Effective Method

The ability to generate rational models of time series plays an important role in such applications as adaptive filtering, spectral estimation, digital control, array processing, and forecasting. A method for effecting an autoregressive moving average (ARMA) model estimate is presented which possesses a number of admirable properties: 1) it has an elegant algebraic structure, 2) its modeling performance in spectral estimation applications has been empirically found to typically exceed that of such contemporary techniques as the periodogram, the Burg method, and the Box-Jenkins method on a variety of problems, 3) it is implementable by computationally efficient algorithms, and 4) it is based on pseudomaximum likelihood concepts. Taken in combination, these properties mark this method as being an effective tool in challenging applications requiring high modeling performance in a real time setting.     

Direction-of-arrival estimation using signal subspace modeling

A computationally viable algorithm for estimating the direction-of-arrival (DOA) of multiple wavefields that are incident on an array of sensors is developed. The geometry of the array is unrestricted and the incident wavefields may be generated by mixtures of incoherent and coherent emitting sources. In the approach taken, the sensor signals are modeled as a noise-contaminated linear combination of steering vectors which are functionally dependent on a set of DOA parameters. These parameters are to be chosen so that this sensor signal model is most compatible with empirically measured data (i.e., snapshot data). An iterative procedure is developed for selecting the most data compatible set of DOA parameters in both the snapshot domain and the array covariance matrix domain. Critical to the success of such iterative solution procedures is the generation of quality DOA parameters to initialize the algorithm. A sequential beamforming method for this initialization is presented. Numerical examples to illustrate the effectiveness of the proposed algorithmic approach are given     

General direction-of-arrival estimation: a signal subspace approach

A high-resolution algorithm is presented for resolving multiple incoherent and coherent plane waves that are incident on an array of sensors. The incident sources can be a mixture of narrowband and broadband sources, and, the geometry of the array is unrestricted. The algorithm makes use of a fundamental property possessed by those eigenvectors of the array spectral density matrix that are associated with eigenvalues that are larger than the sensor noise level. Specifically, it is shown that these eigenvectors can each be represented as linear combinations of the steering vectors identifying the incident plane waves. This property is then used to solve the important special cases of incoherent sources incident on a general array and coherent sources incident on an equispaced linear array. Simulation results are presented to illustrate the high-resolution performance achieved with this approach relative to that obtained with MUSIC and spatial smoothed MUSIC in which the coherent-signal-subspace focusing method is used     

An Algebraic Approach to Superresolution Array Processing

In a recent paper we made an algebraic characterization of the problem of resolving closely spaced plane waves incident on a linear array. The characterization encompassing several superresolution processing methods and encompassed the Wiener, maximum likelihood, and Pisarenko methods as well as suggesting new procedures. In this paper we amplify the algebraic approach and extend the results to consider correlated noise. An adaptive algorithm is given for a particularly effective processing method.