Showing papers on "Median filter published in 1977"

Image restoration by Wiener filtering in the presence of signal-dependent noise.

Abstract: An optimum filter to restore the degraded image due to blurring and the signal-dependent noise is obtained on the basis of the theory of Wiener filtering. Computer simulations of image restoration using signal-dependent noise models are carried out. It becomes clear that the optimum filter, which makes use of a priori information on the signal-dependent nature of the noise and the spectral density of the signal and the noise showing significant spatial correlation, is potentially advantageous.

A Semicausal Model for Recursive Filtering of Two-Dimensional Images

Abstract: A two-dimensional discrete stochastic model for representing images is developed. This representation has lower mean square error, compared to a standard autoregressive Markov representation. Application of the model to linear filtering of images degraded by white noise leads to scalar recursive filtering equations requiring only 0(N2log 2 N) computations for N x N images. The filter algorithm is a hybrid algorithm where the image is transformed along one dimension and spatially filtered, recursively, in the other. Examples on a 255 X 255 image are given.

Detection and selective smoothing of transmission errors in linear PCM

Abstract: We consider detection of transmission errors in pcm by means of statistical hypothesis testing of the received quantized sequence. When errors are detected, a median filter is used to smooth waveform discontinuities. We describe two error detectors, one (cdc), based on correlation measurements, and the other (ddc), based on sample-to-sample difference measurements. While both offer s/n advantages over conventional pcm in the presence of errors, ddc is more promising both in terms of performance and simplicity of implementation.

Short Papers Optimum Filter for Minimization of Image Registration Error Variance

Abstract: It is shown that when two images are registered by cross-correlating one image with a functional of the other image the minimum registration error variance results when the operation performed cofresponds to that of matched filtering.

Design Considerations for Digital Image Processing Systems

Abstract: The brightness levels in a scene or image, together with the spatial relationships among these levels, comprise the total input and output data available to the digital image processor. Indeed, in any mathematical image processing operation, these are the elements of the image which are being manipulated and which represent the major concern of the image processing professional. However, many other factors affect image brightness relations and can distort or obscure the outcome of any image processing experiment. These factors, which consist of a long chain of transmitters, transducers, signal conditioners and processors, are in aggregate commonly called the image chain. An understanding of the image chain is essential to the design of image processing systems.

Elimination of limit cycles in digital filters with very low increase in the quantization noise

Abstract: In recursive digital filters sustained parasitic oscillations are possible. These limit cycles can be suppressed by "random rounding" such that for any relevant input signal the output signal is superimposed only by an aperiodic quantization noise. The mean power of this unavoidable noise can be reduced by a circuit modification.

Noise problems in optical image processing

Abstract: Noise problems in image processing by optical filtering are examined A comparison between coherent and incoherent illumination is made

Sectioning techniques for processing signals with signal-dependent noise

Abstract: In a previous paper, it was shown that a non-linear Bayesian approach to deconvolution could be implemented by sectioned digital filtering. The method of solution used an iterative scheme to maximize the a-posteriori probability in a Bayesian estimator, and it was demonstrated that the iteration computations could be carried out by using sectional digital filtering. In this paper, it is shown that the same technique can be used to solve a wider class of problems of a nonlinear Bayesian estimation with noise which is dependent upon the signal. The analysis is general enough to encompass signal deconvolution, as well as noise smoothing, in both the linear and nonlinear cases.

An adaptive digital filter for improving the signal-to-noise ratio of a signal with additive noise (Corresp.)

Abstract: An adaptive digital filtering scheme is developed to deal with the problem of improving the signal-to-noise ratio of a signal corrupted by noise when only general {\em a priori} assumptions regarding the signal and the noise are possible. Specifically, the noise is assumed to be white zero-mean and uncorrelated; while the signal is considered to be band-limited, possibly with slowly varying spectrum. The proposed adaptive digital filtering scheme is based upon a class of variable wave digital filters. Adaptation of the digital filter multipliers is accomplished through the use of an identification procedure based on an adaptive spectral estimation method.

A statistical, coarse-sampled data filter

Abstract: A digital filter design may be implemented with only one analog summing device, two shift registers with respective lengths equal to the orders of numerator and denominator, a generator of uniformly distributed analog noise, and either a simple low-pass RC filter or a counter. The principle of operation is based on polarity correlation. It has an economic advantage due to lack of storage and digital multipliers, but is applicable only when the data sampling rate is fairly low.

Application Of The K-L Transform To Spatial Domain Filtering Of Multiband Images

Abstract: Conventional two-dimensional low-pass filtering of images for wideband noise suppression invariably, if not by definition, degrades the spatial definition of objects within the image. Thus, SNR improvement is achieved at the expense of edge fidelity. We present in this paper a technique which offers improved noise suppression relative to conventional filtering, and at the same time tends to preserve the important high frequency content of the target image.