scispace - formally typeset
Search or ask a question

Showing papers on "Median filter published in 1976"


Journal ArticleDOI
01 Dec 1976
TL;DR: Two procedures are developed to adapt continuously the finite impulse response of a two-dimensional, noncausal, linear digital filter to remove random noise from gray tone images without significantly sacrificing the subjective resolution.
Abstract: The problem of removing random noise from gray tone images without significantly sacrificing the subjective resolution is considered. Based on a subjective visibility function, which gives the relationship between the visibility of a unit noise and a measure of local spatial detail (spatial masking), two procedures are developed to adapt continuously the finite impulse response of a two-dimensional, noncausal, linear digital filter. At sharp transitions in the image intensity, the filter operator is strongly peaked to preserve the resolution, whereas in flat areas it is flat to effectively average out the random noise. The first procedure (S-filter) is computationally more efficient, but does not perform as well as the second method (SD-filter) which requires solution of a new optimization problem at every picture element. Results of several simulations are presented to demonstrate the feasibility of our approach. Extensions are pointed out to incorporate different adaptation procedures and psychovisual criteria other than the type of spatial masking used here.

201 citations


Patent
04 May 1976
TL;DR: In this paper, the authors identify and analyse the parameters of an input signal that contains speech in the presence of simultaneously occuring near-stationary noise, pauses between speech intervals as well as the termination of such noise can be recognized.
Abstract: By identifying and analyzing the properties of the parameters of an input signal that contains speech in the presence of simultaneously occuring near-stationary noise, pauses between speech intervals as well as the termination of such noise can be recognized. When a pause interval containing noise is recognized, the parameters identified during such interval are used to set the parameters of an adaptive filter through which the input signal is passed during subsequent intervals of speech and until the noise terminates. During the time the input signal passes through the filter, the near-stationary noise is filtered out. In response to recognition of the termination of noise, the input signal is caused to by-pass the filter which is then prepared to accept the parameters of noise occuring in a subsequent pause.

97 citations


Journal ArticleDOI
TL;DR: In this paper, a new noise expression for the class of fixed-point digital filters described by the state equations is formulated, and two methods of its computation are discussed, and the effects of possible structure transformation and state-amplitude scalings are then incorporated in this expression, and results have been analyzed.
Abstract: A new noise expression is formulated for the class of fixed-point digital filters described by the state equations, and two methods of its computation are discussed. The effects of possible structure transformation and state-amplitude scalings are then incorporated in this expression, and the results have been analyzed. In particular, it is shown that the output noise and state amplitudes are inversely proportional, and that an elementary transformation is well suited for a step-by-step generation of a low-noise filter.

94 citations


Journal ArticleDOI
TL;DR: In this paper, a method for image deblurring in spatially incoherent illumination is described, where an additional step of illuminance subtraction is added to simulate negative intensities, and the use of computer-generated holograms of a low number of cells.
Abstract: Under coherent illumination, the interference between signal and noise produces an amplification of the noise. That is why image processing by optical methods may be done better with incoherent light than with coherent light. A method is described where image deblurring is performed in spatially incoherent illumination. The two main features are an additional step of illuminance subtraction to simulate negative intensities, and the use of computer-generated holograms of a low number of cells.

56 citations


Journal ArticleDOI
TL;DR: An estimate of the variance of the registration error that can be expected via two approaches is derived, and it is indicated that for most cases registration variances will be significantly less than the diameter of one picture element.
Abstract: When one image (the sigal) is to be registered with a second image (the signal plus noise) of the same scene, one would like to know the accuracy possible for this registration. This paper derives an estimate of the variance of the registration error that can be expected via two approaches. The solution in each instance is found to be a function of the effective bandwidth of the signal and the noise, and the signal-to-noise ratio. Application of these results to LANDSAT-1 data indicates that for most cases registration variances will be significantly less than the diameter of one picture element.

44 citations


Journal ArticleDOI
TL;DR: A number of signal processing operations can be carried out on spectra using a digital filter based on a simple trapezoid function, applied to the Fourier transform of the spectral signal.
Abstract: A number of signal processing operations can be carried out on spectra using a digital filter based on a simple trapezoid function. The filter is applied to the Fourier transform of the spectral signal. Several signal processing examples are presented to illustrate the capabilities of this filter. These examples include filtering and diagnosis of high frequency noise on a signal, removal of fixed frequency noise, minimization of quantization noise, and differentiation and approximate deconvolution for the purpose of resolution enhancement.

39 citations


Patent
20 Dec 1976
TL;DR: In this paper, a combiner combines the estimated noise and original input signal to develop an output speech signal free of noise, which is then applied to a spectrum shaper to restore a natural shape.
Abstract: Noise such as hum (whose components are self-correlated over a long time-interval) is estimated (separated) from Signal such as speech (whose components are self-correlated only over a short time-interval) by using a time-domain transversal filter, whose tap-weight coefficients may be fixed (Wiener filter) or variable (Adaptive filter). A combiner combines the estimated noise and original input signal to develop an output speech signal free of noise. The output speech may by further applied to a spectrum shaper to restore a natural shape.

35 citations


Proceedings ArticleDOI
09 Jul 1976
TL;DR: The results indicate that subjectively preferable restorations are achieved by Wiener filtering when blur, rather than image noise, is the dominant degrading factor, while noise cheating produces preferable results when noise is predominant.
Abstract: An image enhancement technique called "Noise Cheating" allows the detection of low contrast objects in a noisy background. This technique has been applied to imagery degraded by blur and by noise. The technique is reviewed in this paper, and the results of its application are compared to the results using Wiener filtering. Comparisons are made on both computer simulated and laboratory-generated imagery. The results indicate that subjectively preferable restorations are achieved by Wiener filtering when blur, rather than image noise, is the dominant degrading factor, while noise cheating produces preferable results when noise is predominant.© (1976) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

5 citations



01 Sep 1976
TL;DR: A discrete Wiener filter is developed which explicitly allows for the signal- dependence of the noise and adaptively alters its characteristics based on nonstationary first order statistics of an image.
Abstract: : One goal of this study has been to use the theoretical and experimental results on film characteristics obtained by photographic scientists in order to define film-grain noise within the context of estimation theory. A detailed model for the overall photographic imaging system is presented. There are linear blurring effects at the initial and the final segments of this model to account for various optical and chemical degradations. The middle segment of the model represents signal dependence effects of film-grain noise and includes a nonlinear noise term. The accuracy of this model is tested by simulating images according to it and comparing the results to images of similar targets that were actually recorded on film. These comparisons point out that the model is a good representation of the photographic imaging system. The restoration of images degraded by film-grain noise is considered in two different contexts - estimation theory and detection theory. Under the topic of estimation, a discrete Wiener filter is developed which explicitly allows for the signal- dependence of the noise. The filter adaptively alters its characteristics based on nonstationary first order statistics of an image. This filter is shown to have an advantage over the conventional Wiener filter.

4 citations


Journal ArticleDOI
TL;DR: In this paper, it is shown that time-averaging filters distort and corrupt the signal to obtain optimum signal to noise ratio, while ensemble averaging filters are shown as the better, since no signal distortion occurs.
Abstract: Linear filtering methods of noise reduction in SEM video signals are discussed. It is shown that time-averaging filters distort and corrupt the signal to obtain optimum signal to noise ratio. Ensemble averaging filters are demonstrated as the better, since no signal distortion occurs. Two types of ensemble averaging filters suitable for the SEM are discussed and experimental results in the form of line-scans are presented.

Proceedings ArticleDOI
01 Apr 1976
TL;DR: A signal processing system capable of detecting a signal in nonstationary noise is introduced which makes use of the mean and variance time functions of the non stationary noise background in order to design estimation filters which will cope with the nonstationarity.
Abstract: The problem of detecting target signals in an ocean environment using active sonar is complicated by the non-stationary background which usually consists of both ambient ocean noise and reverberation. In the present paper, a signal processing system capable of detecting a signal in nonstationary noise is introduced. This system makes use of the mean and variance time functions of the nonstationary noise background in order to design estimation filters which will cope with the nonstationarity. Appropriate statistics of the noise and signal (tone burst) plus noise have been obtained and are used to determine the probabilities of false alarm and detection and the receiver operating characteristics.

Journal ArticleDOI
01 Jun 1976
TL;DR: It is demonstrated that the response of the E-filter is almost independent of the frequency of the incoming signal under certain conditions but depends largely on the signal amplitude, which can be used to filter out low-amplitude noise which has been superimposed additively on the signals.
Abstract: A generalization of the E-filter originated by Moore and Parker [10] from a one-dimensional filter to a nonlinear two-dimensional filter is presented. It is demonstrated that the response of the E-filter is almost independent of the frequency of the incoming signal under certain conditions but depends largely on the signal amplitude. Consequently, it can be used to filter out low-amplitude noise which has been superimposed additively on the signal. The performance of a one-dimensional E-filter as applied sequentially to the rows and then the columns of a digitized image is compared with the direct application of a two-dimensional E-filter. A simple two-dimensional E-filter digital realization and several filtered biomedical images are presented to demonstrate the application of this nonlinear filtering technique.

Proceedings ArticleDOI
19 Jul 1976
TL;DR: In this article, the authors demonstrate that homomorphic filtering, parametric Wiener filtering, and constrained least squares filtering can produce nearly diffraction-limited imagery if the atmosphere is "frozen" for 1/100 - 1/10 sec.
Abstract: Conventional post detection image processing techniques have been applied to three simulations of photon noise limited images of the moon which might be obtained through atmospheric turbulence. This study demonstrates that homomorphic filtering, parametric Wiener filtering, and constrained least squares filtering can produce nearly diffraction-limited imagery if the atmosphere is "frozen" for 1/100 - 1/10 sec.

Journal ArticleDOI
Paul Diament1
TL;DR: The process is shown to improve resolution, correct certain aberrations, or reduce effects of noise, when adjustable parameters are properly selected, and tradeoffs between resolution and noise are discussed.
Abstract: Enhancement of blurred images by nutation involves accumulation of information from displaced replicas of the degraded image. The technique is analyzed theoretically and is found to be equivalent to convolution with the weight distribution assigned to the replicas. The process is shown to improve resolution, correct certain aberrations, or reduce effects of noise, when adjustable parameters are properly selected. Plots of spectrum enhancement and of SNR calculations are presented, and tradeoffs between resolution and noise are discussed.

Proceedings ArticleDOI
01 Apr 1976
TL;DR: It is believed that the filtering problem for a digital local exchange is best studied by division into two parts, and several designs of digital filter are described which may be suitable for large scale integration.
Abstract: We believe that the filtering problem for a digital local exchange is best studied by division into two parts. The quantisation noise filter design entails a trade-off between this filter and the type of coder. The complexity of the bandlimiting filter to achieve the CCITT specification is divided between the analogue prefilter and the second part of the digital filter. Several designs of digital filter are described which may be suitable for large scale integration.