Harvey F. Silverman

A Class of Algorithms for Fast Digital Image Registration

The automatic determination of local similarity between two structured data sets is fundamental to the disciplines of pattern recognition and image processing. A class of algorithms, which may be used to determine similarity in a far more efficient manner than methods currently in use, is introduced in this paper. There may be a saving of computation time of two orders of magnitude or more by adopting this new approach. The problem of translational image registration, used for an example throughout, is discussed and the problems with the most widely used method-correlation explained. Simple implementations of the new algorithms are introduced to motivate the basic idea of their structure. Real data from ITOS-1 satellites are presented to give meaningful empirical justification for theoretical predictions.

Processor reconfiguration through instruction-set metamorphosis

The processor reconfiguration through instruction-set metamorphosis (PRISM) general-purpose architecture, which speeds up computationally intensive tasks by augmenting the core processors functionality with new operations, is described. The PRISM approach adapts the configuration and fundamental operations of a core processing system to the computationally intensive portions of a targeted application. PRISM-1, an initial prototype system, is described, and experimental results that demonstrate the benefits of the PRISM concept are presented.<<ETX>>

An introduction to programming the Winograd Fourier transform algorithm (WFTA)

Recently, Dr. Shmuel Winograd discovered a new approach to the computation of the discrete Fourier transform (DFT). Relative to fast Fourier transform (FFT), the Winograd Fourier transform algorithm (WFTA) significantly reduces the number of multiplication operations; it does not increase the number of addition operations in many cases. This paper introduces the new algorithm and discusses the operations comparison problem. A guide for programming is included, as are some preliminary running times.

PRISM-II compiler and architecture

This paper discusses the architecture and compiler for a general-purpose metamorphic computing platform called PRISM-II. PRISM-II improves the performance of many computationally-intensive tasks by augmenting the functionality of the core processor with new instructions that match the characteristics of targeted applications. In essence, PRISM (processor reconfiguration through instruction set metamorphosis) is a general purpose hardware platform that behaves like an application-specific platform. Two methods for hardware synthesis, one using VHDL Designer and the other using X-BLOX, are presented and synthesis results are compared.<<ETX>>

A two-stage algorithm for determining talker location from linear microphone array data☆

Abstract A microphone array system for speech data input must include a robust algorithm for determining the location of the desired talker. Here, a two-stage talker location algorithm based on filtered cross-correlation is introduced. At each stage, maximization of a sum-of-independent-cross-correlations functional is used to establish talker position. Suitable accuracy is obtained at low cost by using multirate interpolation. Experimental evidence has shown that a two-stage procedure improves performance; in the first stage, closely spaced microphone pairs are used to determine the x location of the talker (x0), and more broadly spaced pairs are used in the second stage to find y0 for a restricted range of x. Substantive results, based on real data, are presented to indicate performance. An efficient, global, non-linear optimization technique, stochastic region contraction (SRC), is briefly introduced and is shown to make this algorithm feasible in real time.

Microphone array optimization by stochastic region contraction

The authors deal with optimal microphone placement and gain for a linear one-dimensional array often in a confined environment. A power spectral dispersion function (PSD) is used as a core element for a min-max objective function (PSDX). Derivation of the optimal spacings and gains of the microphones is a hard computational problem since the min-max objective function exhibits multiple local minima (hundreds or thousands). The authors address the computational problem of finding the global optimal solution of the PSDX function. A new method, stochastic region contraction (SRC), is proposed. It achieves a computational speedup of 30-50 when compared to the commonly used simulated-annealing method. SRC is ideally suited for coarse-gain parallel processing. >

Placement of Records on a Secondary Storage Device to Minimize Access Time

The problem considered is how to place records on a secondary storage device to minimize average retrieval time, based on a knowledge of the probability for accessing the records. Theorems are presented for two limiting cases. A numerical example for an intermediate case is also given.

Some analysis of microphone arrays for speech data acquisition

Speech data for recognition, talker verification, or recording are typically acquired using a head-mounted or hand-held microphone. These devices can be very inconvenient or provide a poor signal-to-noise ratio. A microphone array has the potential for surmounting both of these problems. Here, results on optimal spacing and gain for practical linear arrays are derived under the assumption that the desired signal and intrusive speech may be accurately modeled by plane waves. Several sets of curves are presented from the solution of constrained and unconstrained multidimensional nonlinear equations derived from the theory. The curves suggest optimal spacing and gains for linear microphone arrays for speech acquisition.

A parametrically controlled spectral analysis system for speech

The parametrically controlled analyzer (PCA) is a large PL/I program which has been designed to perform spectral analysis of speech signals. PCA features parametric selection of several analysis methods, including discrete Fourier transformation and linear predictive coding. Also, selection may be made among various smoothing, normalization, and interpolation methods. PCA develops high-quality spectrographic representations of speech for standard line printers and CRT displays. The PCA is described and numerous examples of various parameter settings are presented and discussed.

A comparison of several speech-spectra classification methods

An important consideration in speech processing involves classification of speech spectra. Several methods for performing this classification are discussed. A number of these were selected for comparative evaluation. Two measures of performance-accuracy and stability-were derived through the use of an automatic performance evaluation system. Over 3000 hand-labeled spectra were used. Of those evaluated, a linearly mean-corrected minimum distance measure, on a 40-point spectral representation with a square (or cube) norm was consistently superior to the other methods.