Patent US5003618A successfully refreshed from Google Patents.
US5003618A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Automatic adaptive anisotropic digital filtering and biasing of digitized images
Assignee: University of Pittsburgh
•
Filing: 1989-07-14
•
Granted: 1991-03-26
•
Expires: 2009-07-14
Abstract
Abstract Adaptive anisotropic digital filtering is automatically applied to the graycale values of substantially of all the pixels in at least a portion of a digital image by, for each such pixel, selecting and angularly orienting the matrix of coefficients of an anisotropic filter kernel as a weighting function of the gray scale values of pixels in eight, 45 degree sectors under the kernel, and then performing a convolution of the gray scale values of pixels under the kernel with the rotated kernel to generate a filtered value for the center pixel. Local brightness of the image is adjusted either in conjunction with anisotropic filtering or separately by comparing for each pixel the average gray scale value of the pixels under the kernel with a reference value and applying a bias function to the center pixel grey scale value to brighten or darken the image dependent upon the results of the comparison.
Inventors
- 👤 Frank Meno
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-05 13:03
CPC Classifications
G; G06; G06T; G06T5/00; G06T5/20
Claims Analysis
Total Claims: 23
Claim 1
Independent Claim
146 words
A method of adaptive anisotropic filtering of gray scale values for pixels of a digital image comprising the steps of: (a) providing a kernel comprising a matrix of coefficients of a selected anisotropic filter function covering a predetermined number of pixels; (b) selecting a pixel and centering said kernel on said selected pixel; (c) angularly orienting said matrix of coefficients of said kernel as a weighting function of pixel gray scale values for a plurality of sectors of pixels under said kernel angularly distributed around said selected pixel; (d) performing a convolution of said pixel gray scale values under said kernel with said angularly oriented matrix of coefficients of the selected anisotropic filter function to generate a modified gray scale value for said selected pixel; and (e) successively repeating steps (b) through (e) for substantially all pixels in at least a portion of said digital image.
Claim 2
Independent Claim
146 words
A method of adaptive anisotropic filtering of gray scale values for pixels of a digital image comprising the steps of: (a) providing a kernel comprising a matrix of coefficients of a selected anisotropic filter function covering a predetermined number of pixels; (b) selecting a pixel and centering said kernel on said selected pixel; (c) angularly orienting said matrix of coefficients of said kernel as a weighting function of pixel gray scale values for a plurality of sectors of pixels under said kernel angularly distributed around said selected pixel; (d) performing a convolution of said pixel gray scale values under said kernel with said angularly oriented matrix of coefficients of the selected anisotropic filter function to generate a modified gray scale value for said selected pixel; and (e) successively repeating steps (b) through (e) for substantially all pixels in at least a portion of said digital image.
Claim 3
Dependent Claim
86 words
The method of claim 1 wherein said step of angularly orienting a matrix of coefficients of said kernel comprises: grouping equal numbers of pixels under said kernel into said sectors of pixels which are spaced from the selected pixel by sector axes equiangularly spaced about and passing adjacent to said selected pixel; selecting a sector as a function of the gray scale values for the pixels in said sectors; and orienting said matrix of coefficients in said kernel to the sector axis of the selected sector.
Claim 4
Dependent Claim
45 words
The method of claim 2 wherein said step of selecting a sector comprises: determining the average gradient between the gray scale values of the pixels in each sector and in its opposite sector; and selecting the selected sector as a function of said average gradients.
Claim 5
Dependent Claim
43 words
The method of claim 3 wherein said step of selecting a sector as a function of the average gradients between the gray scale values of pixels in each sector and in the opposite sector comprises selecting the sector with the greatest average gradient.
Claim 6
Dependent Claim
43 words
The method of claim 3 wherein the step of selecting a sector as a function of the average gradients between the gray scale values of pixels in each sector and in the opposite sector comprises selecting the sector with the smallest gradient value.
Claim 7
Dependent Claim
74 words
The method of claim 3 wherein the step of determining the average gradient between the gray scale values of the pixels in each sector and its opposite sector comprises for each sector: determining the average gray scale value of the pixels in the sector and in its opposite sector, and subtracting the average gray scale value of the pixels in the opposite sector from the average gray scale value of pixels in the sector.
Claim 8
Dependent Claim
63 words
The method of claim 3 wherein the step of determining the average gradient between the gray scale values of the pixels in each sector and in its opposite sector comprises: selecting a gradient kernel, successively rotating the gradient kernel into alignment with the axis of each sector, and performing for each rotation a convolution between the kernel and the pixel gray scale values.
Claim 9
Dependent Claim
66 words
The method of claim 3 wherein said step of providing a kernel comprises providing a kernel having a matrix or coefficients constituting a low pass filter function in one direction and a high pass filter function in a direction orthogonal to said one direction, and wherein the step of angularly orienting said matrix of coefficients comprises aligning said one direction parallel to the selected sector axis.
Claim 10
Dependent Claim
65 words
The method of claim 3 wherein said step of providing a kernel comprises providing at least two kernels each comprising a matrix of coefficients of a different anisotropic filter function, and wherein said step of angularly orienting said matrix of coefficients includes first selecting one of said at least two provided kernels as a function of the sums of gray scale values in each sector.
Claim 11
Dependent Claim
72 words
The method claim 3 including biasing said pixel gray scale values to adjust brightness without substantially effecting contrast comprising: adding said sums of gray scale values for all of said sectors to generate a total gray scale value, comparing said total gray scale value to a selected reference value, and adjusting said modified gray scale value to a biased gray scale value for said selected pixel as a function of said comparison.
Claim 12
Dependent Claim
69 words
A method of claim 10 wherein said step of adjusting the modified gray scale value comprises reducing the modified gray scale value of the selected pixel by a selected bias function when said total gray scale value exceeds said reference value, and increasing the modified gray scale value of the selected pixel by said selected bias function when said total gray scale value is less than said reference value.
Claim 13
Dependent Claim
75 words
The method of claim 10 wherein said step of adjusting the modified gray scale value of said selected pixel comprises increasing the modified gray scale value of the selected pixel by a selected bias function when the total gray scale value is more than said reference value and reducing the modified gray scale value of the selected pixel by said selected bias function when the total gray scale value is less than said reference value.
Claim 14
Dependent Claim
58 words
The method of claim 12 including storing limits for said adjusted gray scale values, comparing the adjusted gray scale value for the selected pixel to said limits, selecting an alternate bias function when a biased gray scale value is outside said limits and readjusting the modified gray scale values of each selected pixel by said alternate bias function.
Claim 15
Dependent Claim
128 words
Apparatus for adaptive anisotropic filtering of gray scale values for pixels of a digital image comprising: storage means storing a selected kernel comprising a matrix of coefficients of a selected anisotropic filter function covering a predetermined number of pixels; means successively selecting substantially all pixels in at least a portion of said digital image; orienting means centering said kernel on said selected pixel, and angularly orienting said matrix of coefficients of said kernel as a function of the distribution of pixel gray scale values of pixels under said kernel; and convolution means performing a convolution of said pixel gray scale values under said kernel with said angularly oriented matrix of coefficients of the selected anisotropic filter function to generate a modified gray scale value for said selected pixel.
Claim 16
Dependent Claim
128 words
Apparatus for adaptive anisotropic filtering of gray scale values for pixels of a digital image comprising: storage means storing a selected kernel comprising a matrix of coefficients of a selected anisotropic filter function covering a predetermined number of pixels; means successively selecting substantially all pixels in at least a portion of said digital image; orienting means centering said kernel on said selected pixel, and angularly orienting said matrix of coefficients of said kernel as a function of the distribution of pixel gray scale values of pixels under said kernel; and convolution means performing a convolution of said pixel gray scale values under said kernel with said angularly oriented matrix of coefficients of the selected anisotropic filter function to generate a modified gray scale value for said selected pixel.
Claim 17
Dependent Claim
59 words
The apparatus of claim 14 wherein said storage means stores at least two kernels each comprising a matrix of coefficients of a different anisotropic filter function, and wherein said orienting means includes means for first selecting one of said two stored kernels as a function of the distribution of pixel gray scale values of pixels under the selected kernel.
Claim 18
Dependent Claim
86 words
The apparatus of claim 14 wherein said orienting means include means grouping equal numbers of pixels under said kernel into sectors of pixels spaced from the selected pixel by sector axes equiangularly spaced about and passing adjacent to said selected pixel, means summing for each sector the gray scale values for all of the pixels in the sector, means selecting a sector as a function of said sums, and means aligning said matrix of coefficients of said kernel to the sector axis of the selected sector.
Claim 19
Dependent Claim
96 words
The apparatus of claim 16 including biasing means biasing said modified gray scale values of the selected pixels to adjust brightness without substantially effecting contrast comprising: means storing a reference value, means adding the sums of pixel gray scale values for all of said pixels under the kernel to generate a total pixel value, means comparing said total pixel value to said stored reference value and means adjusting the modified gray scale value of the selected pixel to an adjusted gray scale value by a selected bias function in a sense determined by said comparison means.
Claim 20
Dependent Claim
19 words
The apparatus of claim 16 wherein said grouping means groups pixels under said kernel into eight equiangularly spaced sectors.
Claim 21
Dependent Claim
57 words
The apparatus of claim 16 wherein said storing means stores a matrix of coefficients constituting a low pass filter function in one direction and a high pass filter function in a direction orthogonal to said one direction, and wherein said aligning means aligns said matrix of coefficients with said one direction parallel to the selected sector axis.
Claim 22
Dependent Claim
20 words
The apparatus of claim 18 wherein said grouping means groups said pixels under said kernel into eight equally spaced sectors.
Claim 23
Dependent Claim
93 words
The apparatus of claim 20 including biasing means for the modified pixel gray scale values to adjust brightness without substantially effecting contrast comprising: means storing a reference value, means adding the sums of pixel gray scale values for all of said pixels under the kernel to generate a total pixel value, means comparing said total pixel value to said stored reference value, and means adjusting the modified gray scale value of the selected pixel to an adjusted gray scale value by a selected bias function in a sense determined by said comparison means.
References Cited (Backward Citations)
US4547895A
Cited
US4691366A
Cited
US4541116A
Cited
US4724544A
Cited
US4618990A
Cited
US4760541A
Cited
US4792900A
Cited
US4908872A
Cited
US4811414A
Cited
US4783840A
Cited
US4791679A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1994-03-25 | 1994-09-24 | 1995-03-26 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1998-03-25 | 1998-09-24 | 1999-03-26 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2002-03-25 | 2002-09-24 | 2003-03-26 | $7700.00 | Window Closed / Lapsed |