US4910694A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Method for approximating a value which is a nonlinear function of the linear average of pixel data
Assignee: Eastman Kodak Co
•
Filing: 1988-01-11
•
Granted: 1990-03-20
•
Expires: 2008-01-11
Abstract
Abstract A system for weighting nonlinear, e.g. logarithmic, pixel data as part of a missing pixel interpolation. This is accomplished to generate a weighted average of the logarithmic intensity values for those pixels which surround the missing pixel in order to approximate the logarithm of the linear average of these pixel values. Two weighting coefficients of unequal value are selected so that their sum equals the quantity one. The larger of the logarithmic values is weighted by multiplying the larger logarithmic pixel value by the larger weighting coefficient, and then by adding to this quantity the smaller logarithmic pixel value which has been multiplied by the smaller weighting coefficient. The result is a weighted logarithmic average which approximates the log of the linear average.
Inventors
- 👤 Lawrence E. Walther
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-03 23:35
CPC Classifications
H; H04; H04N; H04N23/00; H04N23/80; H04N23/84; H04N23/843
Claims Analysis
Total Claims: 16
Claim 1
Independent Claim
249 words
In an imaging system having image signal inputs and image signal outputs, a method of averaging first and second image signal outputs corresponding to first and second different pixels of an image in order to approximate a third image signal output corresponding to a third different pixel of the image, the method comprising the steps of: a. receiving the first and second image signal outputs wherein said first and second image signal outputs vary from their respective first and second image signal inputs in a nonlinear manner so that at lower levels of the image signal inputs a change in the levels of the image signal inputs generates a first change in the levels of the image signal outputs, and at higher levels of the image signal inputs, a corresponding change in the levels of the image signal inputs generates a second change in the levels of the image signal outputs which is difference from the first change; b. determining a difference between the level of the first image signal output and the level of the second image signal output; and c. generating a weighted average of the nonlinear first and second image signal outputs in order to approximate said third image signal output from an average of the first and second image signal inputs by changing an unweighted average of the first and second image signal outputs by an amount which is a function of the difference between the levels of the first and second image signal outputs.
Claim 2
Dependent Claim
249 words
In an imaging system having image signal inputs and image signal outputs, a method of averaging first and second image signal outputs corresponding to first and second different pixels of an image in order to approximate a third image signal output corresponding to a third different pixel of the image, the method comprising the steps of: a. receiving the first and second image signal outputs wherein said first and second image signal outputs vary from their respective first and second image signal inputs in a nonlinear manner so that at lower levels of the image signal inputs a change in the levels of the image signal inputs generates a first change in the levels of the image signal outputs, and at higher levels of the image signal inputs, a corresponding change in the levels of the image signal inputs generates a second change in the levels of the image signal outputs which is difference from the first change; b. determining a difference between the level of the first image signal output and the level of the second image signal output; and c. generating a weighted average of the nonlinear first and second image signal outputs in order to approximate said third image signal output from an average of the first and second image signal inputs by changing an unweighted average of the first and second image signal outputs by an amount which is a function of the difference between the levels of the first and second image signal outputs.
Claim 3
Dependent Claim
43 words
The method as set forth in claim 1 wherein: a. the first change in the levels of the first and second signal outputs is greater than the second change; and b. the unweighted average of the first and second signal outputs is increased.
Claim 4
Dependent Claim
33 words
The method as set forth in claim 1 wherein the unweighted average is increased by an amount which is proportional to the difference between the levels of the first and second signal outputs.
Claim 5
Dependent Claim
57 words
The method as set forth in claim 1 wherein: a. the unweighted average of the first and second signal outputs is an average of the logarithms of the first and second signal outputs; and b. the average of the first and second signal inputs is a logarithm of the average of the first and second signal inputs.
Claim 6
Dependent Claim
109 words
The method as set forth in claim 1 wherein the generating step is accomplished by: a. generating a first weighting coefficient X and a second weighting coefficient Y in a manner that X is greater than Y; b. determining the larger and smaller of the nonlinear first and second signal outputs; c. multiplying (i) the larger of the first and second signal outputs by the X coefficient to generate a first weighted output, and (ii) the smaller of the nonlinear first and second signal levels by the Y coefficient to generate a second weighted output; and d. combining the first and second weighted outputs to generate the weighted average.
Claim 7
Dependent Claim
73 words
The method as set forth in claim 1 wherein: a. during the receiving step the first and second signal outputs are generated by a nonlinear amplifier having a dynamic range; and b. the generating step includes the steps of: (1) determining a difference X-Y between the weighting coefficients, and (2) selecting the weighting coefficients X and Y so that the difference X-Y is a function of the dynamic range of the nonlinear amplifier.
Claim 8
Dependent Claim
35 words
The method as set forth in claim 6 wherein during the generating step the coefficients X and Y are selected so that at greater dynamic ranges of the nonlinear amplifier, the difference X-Y is greater.
Claim 9
Dependent Claim
45 words
The method as set forth in claim 7 wherein during the generating step the first weighting coefficient is selected so that 1/2<X<1, and the second weighting coefficient is selected so that 0<Y<1/2, and further wherein the sum of X and Y is equal to one.
Claim 10
Dependent Claim
78 words
The method as set forth in claim 8 wherein the generating and multiplying steps are accomplished by: a. determining an absolute value difference between the first and second signal output levels; b. multiplying the absolute value difference by a selected weighting factor which is greater than 0 and less than 1/2 to generate a third output; and c. subtracting the third output from the larger of the first and second signal output levels to generate the weighted average.
Claim 11
Dependent Claim
111 words
The method as set forth in claim 9 wherein: a. the selected weighting factor is -1/Z; b. the third output equals -(1/Z)(|log A-log B|), where log A is the value of the first signal level and log B is the value of the second signal level; c. when log A is greater than log B, the weighted average equals the sum of the third output and log A, which in turn is equal to log A-1/Z(|log A-log B|); and d. when log B is greater than log A, the weighted average equals the sum of the third output and log B, which in turn is equal to log B-1/Z(|log B-log A|).
Claim 12
Dependent Claim
41 words
The method as set forth in claim 10 wherein: a. during the receiving step the first and second signal levels are received from a logarithmic amplifier having a dynamic range of about 8 octaves: and b. Z is equal to 4.
Claim 13
Independent Claim
205 words
A method for estimating missing color pixel information from nonmissing color pixel information which is output from an image sensor, the method comprising the steps of: a. receiving image pixel date from the image sensor which represents levels of primary colors and which has a missing color pixel, the image pixel data being in a logarithmic format; b. selecting data representing logarithmic levels of first and second different nonmissing image pixels from the image pixel data, the first and second logarithmic pixel data representing a primary color which is missing from the missing color pixel data; c. generating a first weighting coefficient X wherein 1/2<x<1, and a second weighting coefficient Y wherein 0<Y<1/2, and further wherein the sum of X and Y is substantially equal to one; d. determining the larger and smaller of the first and second logarithmic pixel levels; e. multiplying (i) the larger of the logarithmic first and second pixel levels by the x coefficient to generate a first output, and (ii) the smaller of the first and second pixel levels by the X coefficient to generate a second output; and f. combining the first and second outputs to generate an output which is the estimate of the missing color pixel information.
Claim 14
Independent Claim
205 words
A method for estimating missing color pixel information from nonmissing color pixel information which is output from an image sensor, the method comprising the steps of: a. receiving image pixel date from the image sensor which represents levels of primary colors and which has a missing color pixel, the image pixel data being in a logarithmic format; b. selecting data representing logarithmic levels of first and second different nonmissing image pixels from the image pixel data, the first and second logarithmic pixel data representing a primary color which is missing from the missing color pixel data; c. generating a first weighting coefficient X wherein 1/2<x<1, and a second weighting coefficient Y wherein 0<Y<1/2, and further wherein the sum of X and Y is substantially equal to one; d. determining the larger and smaller of the first and second logarithmic pixel levels; e. multiplying (i) the larger of the logarithmic first and second pixel levels by the x coefficient to generate a first output, and (ii) the smaller of the first and second pixel levels by the X coefficient to generate a second output; and f. combining the first and second outputs to generate an output which is the estimate of the missing color pixel information.
Claim 15
Dependent Claim
25 words
The method as set forth in claim 12 wherein the correction term increases as the difference between the first and second image signal levels increases.
Claim 16
Dependent Claim
41 words
The method as set forth in claim 13 wherein the correction term is defined by the equation (1/Z)(|log A-log B|), where A represents the first image signal level, B represents the second image signal level, and Z is greater than two.
References Cited (Backward Citations)
US3737781A
Cited
US4242704A
Cited
US4454546A
Cited
US4463375A
Cited
US4468704A
Cited
US4656500A
Cited
US4642678A
Cited
US4649568A
Cited
US4663655A
Cited
US4774565A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1993-03-19 | 1993-09-18 | 1994-03-20 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1997-03-19 | 1997-09-18 | 1998-03-20 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2001-03-19 | 2001-09-18 | 2002-03-20 | $7700.00 | Window Closed / Lapsed |