Patent US6801672B1 successfully refreshed from Google Patents.
US6801672B1
Expired / Lapsed
US
⏱️ 0.0 years remaining
Removing noise from a color image using wavelets
Assignee: Individual
•
Filing: 2001-06-01
•
Granted: 2004-10-05
•
Expires: 2021-06-01
Abstract
Abstract A color image is enhanced by removing noise. A wavelet transform is applied to a color image to generate sets of wavelet transform coefficients for multiple channels. Edge maps are produced for the channels from the sets of wavelet transform coefficients. An edge likelihood map is generated based on at least one edge map. The edge likelihood map is applied to the sets of wavelet transform coefficients to generate sets of filtered wavelet transform coefficients. A de-noised color image is reconstructed from the sets of filtered wavelet transform coefficients.
Inventors
- 👤 Bruce A. Thomas
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-01 08:01
CPC Classifications
G; G06; G06T; G06T5/00; G06T5/10; G06T5/70; G06T7/00; G06T7/10; G06T7/13; G06T2207/00; G06T2207/10; G06T2207/10024; G06T2207/20; G06T2207/20048; G06T2207/20064; G06T2207/20172; G06T2207/20192
Claims Analysis
Total Claims: 41
Claim 1
Independent Claim
87 words
A method of removing noise from a color image, comprising: applying a wavelet transform to a color image to generate sets of wavelet transform coefficients for multiple color channels; producing edge maps for the multiple color channels from the sets of wavelet transform coefficients; generating a multiplicative mask based on at least two edge maps; applying the multiplicative mask to the sets of wavelet transform coefficients to generate sets of filtered wavelet transform coefficients; and reconstrucing a de-noised image from the sets of filtered wavelet transform coefficients.
Claim 2
Independent Claim
87 words
A method of removing noise from a color image, comprising: applying a wavelet transform to a color image to generate sets of wavelet transform coefficients for multiple color channels; producing edge maps for the multiple color channels from the sets of wavelet transform coefficients; generating a multiplicative mask based on at least two edge maps; applying the multiplicative mask to the sets of wavelet transform coefficients to generate sets of filtered wavelet transform coefficients; and reconstrucing a de-noised image from the sets of filtered wavelet transform coefficients.
Claim 3
Dependent Claim
21 words
The method of claim 1 wherein said generating generates the multiplicative mask based on at least two chromatically distinct edge maps.
Claim 4
Dependent Claim
22 words
The method of claim 1 wherein said generating generates the multiplicative mask based on at least two edge maps of differing scale.
Claim 5
Dependent Claim
56 words
The method of claim 1 wherein the edge maps include chromatically (i.e. color channel) distinct sets of edge maps; and said generating generates the multiplicative mask based on at least two chromatically (i.e. color channel) distinct sets of edge maps, each chromatically distinct set of edge maps including at least two edge maps of differing scale.
Claim 6
Dependent Claim
34 words
The method of claim 1 wherein the edge maps specify a location and magnitude of an edge associated with a pixel; and said generating generates the multiplicative mask based on the location and magnitude.
Claim 7
Dependent Claim
34 words
The method of claim 1 wherein the edge maps specify a location and angle of an edge associated with a pixel; and said generating generates the multiplictive mask based on the location and angle.
Claim 8
Dependent Claim
14 words
The method of claim 1 wherein the wavelet transform is an undecimated wavelet transform.
Claim 9
Dependent Claim
14 words
The method of claim 1 wherein the wavelet transform is a decimated wavelet transform.
Claim 10
Dependent Claim
17 words
The method of claim 1 wherein the multiple color channels include red, green and blue color values.
Claim 11
Dependent Claim
20 words
The method of claim 1 wherein the multiple color channels are color-space transformations of red, green and blue color values.
Claim 12
Independent Claim
78 words
A method of removing noise from a color image, comprising; applying a wavelet transform to a color image to generate sets of wavelet transform coefficients for multiple color channels; producing edge maps for the multiple color channels from the sets of wavelet transform coefficients; generating a multiplicative mask based on at least two edge maps; applying the multiplicative mask to each edge map to generate filtered edge maps; and reconstructing a de-noised image from the filtered edge maps.
Claim 13
Independent Claim
78 words
A method of removing noise from a color image, comprising; applying a wavelet transform to a color image to generate sets of wavelet transform coefficients for multiple color channels; producing edge maps for the multiple color channels from the sets of wavelet transform coefficients; generating a multiplicative mask based on at least two edge maps; applying the multiplicative mask to each edge map to generate filtered edge maps; and reconstructing a de-noised image from the filtered edge maps.
Claim 14
Dependent Claim
21 words
The method of claim 11 wherein said generating generates the multiplicative mask based on at least two chromatically distinct edge maps.
Claim 15
Dependent Claim
22 words
The method of claim 11 wherein said generating generates the multiplicative mask based on at least two edge maps of differing scale.
Claim 16
Dependent Claim
50 words
The method of claim 11 wherein the edge maps include chromatically distinct sets of edge maps, and said generating generates the multiplicative mask based on at least two chromatically distinct sets of edge maps, each chromatically distinct set of edge maps including at least two edge maps of differing scale.
Claim 17
Dependent Claim
22 words
The method of claim 11 wherein said generating the multiplicative mask includes adding all the edge maps associated with a predetermined scale.
Claim 18
Dependent Claim
18 words
The method of claim 11 wherein said generating the multiplicative mask includes adding at least two edge maps.
Claim 19
Dependent Claim
35 words
The method of claim 11 wherein said generating the multiplicative mask includes producing an edge likelihood map by combining at least two edge maps, the multiplicative mask being generated based on the edge likelihood map.
Claim 20
Dependent Claim
17 words
The method of claim 17 further comprising scaling the edge likelihood map to generate the multiplicative mask.
Claim 21
Dependent Claim
30 words
The method of claim 17 further comprising: limiting values of the edge likelihood map that exceed a first predetermined value to a second predetermined value to generate the multiplicative mask.
Claim 22
Dependent Claim
14 words
The method of claim 19 wherein said scaling includes scaling the edge likelihood map.
Claim 23
Dependent Claim
17 words
The method of claim 19 wherein said first predetermined value is equal to the second predetermined value.
Claim 24
Dependent Claim
26 words
The method of claim 19 , further comprising: generating a histogram of the edge likelihood map; and determining the first predetermined value based on the histogram.
Claim 25
Dependent Claim
47 words
The method of claim 17 wherein the edge maps include descriptors of the size and location of edges for the color channel, wherein said producing the edge likelihood map is based on the descriptors of the size and location of edges of at least two edge maps.
Claim 26
Dependent Claim
47 words
The method of claim 17 wherein the edge maps include descriptors of the location and angle of edges for the color channel, wherein said producing the edge likelihood map is based on the descriptors of the location and angle of edges of at least two edge maps.
Claim 27
Dependent Claim
14 words
The method of claim 11 wherein the wavelet transform is an undecimated wavelet transform.
Claim 28
Dependent Claim
13 words
The method of claim 11 wherein the wavelet transform is a decimated transform.
Claim 29
Dependent Claim
42 words
The method of claim 11 wherein said reconstructing includes: generating a set of wavelet transform reconstruction coefficients for each filtered edge map; and applying an inverse wavelet transform to each set of wavelet transform reconstruction coefficients to generate the de-noised color image.
Claim 30
Dependent Claim
17 words
The method of claim 11 wherein the multiple color channels are red, green and blue color values.
Claim 31
Dependent Claim
16 words
The method of claim 11 wherein the multiple color channels are luminance, saturation and hue values.
Claim 32
Dependent Claim
20 words
The method of claim 11 wherein the multiple color channels are color-space transformations of red, green and blue color values.
Claim 33
Dependent Claim
116 words
A computer program product for use in conjunction with a computer system, the computer program product comprising a computer readable storage medium and a computer program mechanism embedded therein, the computer program mechanism comprising: a wavelet transform procedure that is applied to color image data to generate sets of wavelet transform coefficients for multiple color channels; an edge detector procedure that produces edge maps for the multiple color channels from the sets of wavelet transform coefficients; instructions that generate a multiplicative mask based on at least two edge maps; instructions that multiply the edge maps by the multiplicative mask to produce filtered edge maps; and instructions that reconstruct a de-noised image from the filtered edge maps.
Claim 34
Dependent Claim
116 words
A computer program product for use in conjunction with a computer system, the computer program product comprising a computer readable storage medium and a computer program mechanism embedded therein, the computer program mechanism comprising: a wavelet transform procedure that is applied to color image data to generate sets of wavelet transform coefficients for multiple color channels; an edge detector procedure that produces edge maps for the multiple color channels from the sets of wavelet transform coefficients; instructions that generate a multiplicative mask based on at least two edge maps; instructions that multiply the edge maps by the multiplicative mask to produce filtered edge maps; and instructions that reconstruct a de-noised image from the filtered edge maps.
Claim 35
Independent Claim
87 words
A method of removing noise from a color image, comprising: applying a wavelet transform to a color image to generate sets of wavelet transform coefficients for multiple channels; producing edge maps for the channels from the sets of wavelet transform coefficients; generating an edge likelihood map based on at least one edge map; applying the edge likelihood map to the sets of wavelet transform coefficients to generate sets of filtered wavelet transform coefficients; and reconstructing a de-noised color image from the sets of filtered wavelet transform coefficients.
Claim 36
Independent Claim
87 words
A method of removing noise from a color image, comprising: applying a wavelet transform to a color image to generate sets of wavelet transform coefficients for multiple channels; producing edge maps for the channels from the sets of wavelet transform coefficients; generating an edge likelihood map based on at least one edge map; applying the edge likelihood map to the sets of wavelet transform coefficients to generate sets of filtered wavelet transform coefficients; and reconstructing a de-noised color image from the sets of filtered wavelet transform coefficients.
Claim 37
Dependent Claim
38 words
The method of claim 32 wherein the channels include luminance values, one of the edge maps is associated with the luminance values, and the edge likelihood map is based on the edge map associated with the luminance values.
Claim 38
Independent Claim
97 words
A method of removing noise from a color image, comprising: applying a wavelet transform to a color image to generate sets of wavelet transform coefficients for multiple color channels; producing edge maps for the color channels from the sets of wavelet transform coefficients; generating an edge likelihood map for each color channel based on the edge maps; applying the edge likelihood maps to the sets of wavelet transform coefficients associated with the respective color channels to generate sets of filtered wavelet transform coefficients; and reconstructing a de-noised color image from the sets of filtered wavelet transform coefficients.
Claim 39
Independent Claim
97 words
A method of removing noise from a color image, comprising: applying a wavelet transform to a color image to generate sets of wavelet transform coefficients for multiple color channels; producing edge maps for the color channels from the sets of wavelet transform coefficients; generating an edge likelihood map for each color channel based on the edge maps; applying the edge likelihood maps to the sets of wavelet transform coefficients associated with the respective color channels to generate sets of filtered wavelet transform coefficients; and reconstructing a de-noised color image from the sets of filtered wavelet transform coefficients.
Claim 40
Independent Claim
103 words
A method of removing noise from a color image, comprising: applying a wavelet transform to a color image to generate sets of wavelet transform coefficients for multiple color channels; producing edge maps for the color channels from the sets of wavelet transform coefficients; generating an edge likelihood map based on the edge maps; proportionally applying the edge likelihood map to the sets of wavelet transform coefficients in accordance with a contribution of a respective color channel to the edge likehood map to generate sets of filtered wavelet transform coefficients; and reconstructing a de-noised color image from the sets of filtered wavelet transform coefficients.
Claim 41
Independent Claim
103 words
A method of removing noise from a color image, comprising: applying a wavelet transform to a color image to generate sets of wavelet transform coefficients for multiple color channels; producing edge maps for the color channels from the sets of wavelet transform coefficients; generating an edge likelihood map based on the edge maps; proportionally applying the edge likelihood map to the sets of wavelet transform coefficients in accordance with a contribution of a respective color channel to the edge likehood map to generate sets of filtered wavelet transform coefficients; and reconstructing a de-noised color image from the sets of filtered wavelet transform coefficients.
References Cited (Backward Citations)
US6002794A
Cited
US5892850A
Cited
US6614847B1
Cited
US6195456B1
Cited
US6711299B2
Cited
US6125201A
Cited
US6075886A
Cited
US6360021B1
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2007-10-05 | 2008-04-05 | 2008-10-05 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2011-10-05 | 2012-04-05 | 2012-10-05 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2015-10-05 | 2016-04-05 | 2016-10-05 | $7700.00 | Window Closed / Lapsed |