Showing papers by "Charles G. Boncelet published in 1989"
TL;DR: The Ll- Filters are introduced to generate the order statistic filters (L-filters) and the nonrecursive linear, or finite-duration impulse-response (FIR), filters.
Abstract: The Ll-filters are introduced to generate the order statistic filters (L-filters) and the nonrecursive linear, or finite-duration impulse-response (FIR), filters. Such estimators are particularly effective filtering signals that do not necessarily follow Gaussian distributions. They can be designed to restore one-dimensional or multidimensional signals corrupted by noise of impulsive type. Such filters are appealing since they are suitable for being made robust against the presence of spurious outliers in the data. >
101 citations
14 Aug 1989
TL;DR: In this article, the removal of noise by a class of order statistic filters, called multistage order statistics filters or multi-window filters, is described and its equivalence under special circumstances to a filter defined in the literature is explored.
Abstract: The removal of noise by a class of order statistic filters, called multistage order statistic filters or multiwindow filters, is described. These filters take 'ranks of ranks'. A new filter type is defined and its equivalence under special circumstances to a filter defined in the literature is shown. This filter's deterministic and statistical properties are explored. This filter is believed to be the best available for smoothing high signal-to-noise ratio images. >
4 citations
01 Jan 1989
TL;DR: In this article, the porosity value of composite materials is quantified using a NDE technique based on digitized waveforms obtained from pulse-echo scans and a software framework was created with compatability between the signal and image processing modules.
Abstract: The method presented to detect and quantify the porosity value content of composite materials is part of a development effort to establish an integrated ultrasonic evaluation system 1. The overall research effort aims to automate the NDE process and predict mechanical and structural properties of composite materials using NDE techniques. The ultrasonic NDE technique used herein is based on digitized waveforms obtained from pulse-echo scans. For the fast integration of new algorithms into the system, a software framework was created with compatability between the signal and image processing modules. The porosity volume content of the test specimens is related to statistical properties of the ultrasonic data rather than to the ultrasonic attenuation with respect to frequency2.
3 citations