US4918633A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Digital image noise reduction method and transmission system
Assignee: Eastman Kodak Co
•
Filing: 1988-03-22
•
Granted: 1990-04-17
•
Expires: 2008-03-22
Abstract
Abstract A method for reducing noise in an image signal includes the steps of sampling an image to produce a sampled image signal representing pixel amplitude values, the sampled image signal including noise having a known probabilty density function (pdf); digitizing the pixel amplitude values to produce a digital image; forming an image difference signal from the digital image signal by taking the difference between successive pixel amplitude values; processing the image difference signal in digital processing means including a digital MAP filter incorporating a MAP estimator for estimating the signal and noise pdf's of a signal having a Laplacian pdf degraded by noise having a know pdf, to produce a MAP filtered image difference signal; and recovering a restored digital image signal from the MAP filtered image difference signal by summing successive amplitude difference values in the MAP filtered image difference signal. Since the pdf of the differential image signal is more nearly stationary than the image signal itself, a MAP estimator based on a differential image signal provides a more reliable estimate of the original signal statistics, even though the statistics of original signal may vary widely.
Inventors
- 👤 James R. Sullivan
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-27 07:28
CPC Classifications
G; G06; G06T; G06T5/00; G06T5/70; G06T5/10; G06V; G06V10/00; G06V10/20; G06V10/30; G06T2207/00; G06T2207/20; G06T2207/20076; G06T2207/20172; G06T2207/20204
Claims Analysis
Total Claims: 12
Claim 1
Independent Claim
149 words
A signal processing method for reducing noise in an image signal, comprising the steps of: (a) sampling an image to produce a sampled image signal representing pixel amplitude values, the sampled image signal including noise having a known probability density function (pdf); (b) digitizing the pixel amplitude values to produce a digital image; (c) forming an image difference signal from the digital image signal by taking the difference between successive pixel amplitude values; (d) processing the image difference signal in digital processing means including a digital MAP filter incorporating a MAP estimator for estimating the signal and noise pdf's of a signal having a Laplacian pdf degraded by noise having a known pdf, to produce a MAP filtered image difference signal; and (e) recovering a restored digital image signal from the MAP filtered image difference signal by summing successive amplitude difference values in the MAP filtered image difference signal.
Claim 2
Dependent Claim
149 words
A signal processing method for reducing noise in an image signal, comprising the steps of: (a) sampling an image to produce a sampled image signal representing pixel amplitude values, the sampled image signal including noise having a known probability density function (pdf); (b) digitizing the pixel amplitude values to produce a digital image; (c) forming an image difference signal from the digital image signal by taking the difference between successive pixel amplitude values; (d) processing the image difference signal in digital processing means including a digital MAP filter incorporating a MAP estimator for estimating the signal and noise pdf's of a signal having a Laplacian pdf degraded by noise having a known pdf, to produce a MAP filtered image difference signal; and (e) recovering a restored digital image signal from the MAP filtered image difference signal by summing successive amplitude difference values in the MAP filtered image difference signal.
Claim 3
Dependent Claim
40 words
The image signal processing method claimed in claim 1, wherein the MAP filter employs an estimate of the noise variance, and further including the step of estimating the noise variance and employing the noise variance estimate in the MAP filter.
Claim 4
Dependent Claim
53 words
The image signal processing method claimed in claim 2, wherein the MAP filter employs local variance estimates of the image difference signal, and further including the steps of: (a) segmenting the image difference signal into segments of constant variance employing a generalized-likelihood-ratio-test, and employing the variance of the segments in the MAP filter.
Claim 5
Dependent Claim
47 words
The image signal processing method claimed in claim 3, wherein the noise is signal dependent, and further including the steps of: (a) first performing a transformation on the image signal that renders the noise signal independent; and (b) performing the inverse transformation on the restored image signal.
Claim 6
Dependent Claim
81 words
The image signal processing method claimed in claim 4, wherein the image signal is derived from a photographic image degraded by film grain noise being signal dependent and having a Gaussian pdf, said transformation having the form ##EQU25## wherein C 1 and C 2 are arbitrary constants and σ n (S') is the Selwyn granularity measured as a function of photographic density S', C 1 being the estimate of the noise variance σ n that is supplied to the MAP filter.
Claim 7
Dependent Claim
47 words
The image signal processing method claimed in claim 5, wherein said MAP filter implements the operation ##EQU26## where ΔS is the map filtered image difference signal, ΔX is the noisy image difference signal, and σΔS is the estimate of the variance of the original image difference signal.
Claim 8
Dependent Claim
39 words
The image signal processing method claimed in claim 6, including the steps of: (a) determining the mean value of X in the segments when ΔS=0; and (b) employing said mean value as the processed signal value in such segments.
Claim 9
Dependent Claim
209 words
A DPCM digital image transmission system, comprising: (a) an encoder for converting a digital image signal to a differential pulse code modulated (DPCM) image signal, the encoder including, (1) a differencing means for forming the DPCM image signal by taking the difference between an incoming digital image signal and a prediction signal, (2) MAP filter means comprising a programmed digital device for, receiving the DPCM image signal, and producing a MAP filtered DPCM image signal, the MAP filter incorporating a MAP estimator for estimating the signal and noise pdf's of a signal having a Laplacian pdf degraded by noise having a known pdf, (3) a quantizer, receiving the DPCM image signal and producing a quantized DPCM image signal; (4) a feedback loop including a predictor for receiving the quantized DPCM image signal from the quantizer and generating and supplying the prediction signal to the differencing means; (b) a communication channel for the quantized DPCM image signal; (c) a decoder for decoding the DPCM image signal, including, (1) summing means for forming an image signal by summing the DPCM image signal and a prediction signal, and (2) a feedback loop for receiving the image signal from the summing means and generating and supplying the prediction signal to the summing means.
Claim 10
Dependent Claim
209 words
A DPCM digital image transmission system, comprising: (a) an encoder for converting a digital image signal to a differential pulse code modulated (DPCM) image signal, the encoder including, (1) a differencing means for forming the DPCM image signal by taking the difference between an incoming digital image signal and a prediction signal, (2) MAP filter means comprising a programmed digital device for, receiving the DPCM image signal, and producing a MAP filtered DPCM image signal, the MAP filter incorporating a MAP estimator for estimating the signal and noise pdf's of a signal having a Laplacian pdf degraded by noise having a known pdf, (3) a quantizer, receiving the DPCM image signal and producing a quantized DPCM image signal; (4) a feedback loop including a predictor for receiving the quantized DPCM image signal from the quantizer and generating and supplying the prediction signal to the differencing means; (b) a communication channel for the quantized DPCM image signal; (c) a decoder for decoding the DPCM image signal, including, (1) summing means for forming an image signal by summing the DPCM image signal and a prediction signal, and (2) a feedback loop for receiving the image signal from the summing means and generating and supplying the prediction signal to the summing means.
Claim 11
Dependent Claim
67 words
The DPCM digital image transmission system claimed in claim 8, wherein the noise is signal dependent noise, and further comprising: (a) programmed digital device means in the encoder for performing a transformation on the image signal prior to the differencing means to render noise in the signal, signal independent; and (b) programmed digital device means in the decoder for performing an inverse transformation on the image signal.
Claim 12
Dependent Claim
80 words
The DPCM digital image transmission system claimed in claim 8, wherein said MAP filter means employs local variance estimates of the image difference signal, and further comprising: (a) segment generation means comprising a programmed digital device for separating the image signal into segments having constant variance, said segment generation means including means for applying a generalized-likelihood-ratio-test to successive samples of the DPCM image signal, and for supplying the MAP filter means with estimates of the variance of the respective segments.
References Cited (Backward Citations)
US4167749A
Cited
US4330833A
Cited
US4441121A
Cited
US4654863A
Cited
US4764971A
Cited
US4774682A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1993-04-16 | 1993-10-16 | 1994-04-17 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1997-04-16 | 1997-10-16 | 1998-04-17 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2001-04-16 | 2001-10-16 | 2002-04-17 | $7700.00 | Window Closed / Lapsed |