US 2023/0055561 A1Published application

Decorrelation transform for a color filter array pattern image

AssigneeFraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V.

InventorsThomas RICHTER, Christian MINUTH, Micha NIETSCH, Christian SCHERL, Wolfgang HEPPNER, Tobias JASCHKE, Marcus FREISLEBEN, Bilal AHMED

Life of this patent

Estimating the term…

Abstract

A concept for coding a CFA pattern image is provided, wherein, for each of the sample clusters of the CFA pattern image, having first to fourth sample positions, first to fourth color coordinates are computed, so as to provide an image representation of the CFA pattern image.

Claims24

Dependent claims

An apparatus for decoding an image representation into a color filter array (CFA) pattern image,

wherein the CFA pattern image comprises first samples of a first channel, second samples of a second channel, and third samples of a third channel,
wherein the CFA pattern image is partitioned into sample clusters, each of which comprises first to fourth sample positions with one of the first samples located at the first sample position, one of the second samples located at the second sample position, and two third samples located at the third and fourth sample positions, respectively,
wherein the apparatus is configured to derive, for each sample cluster, a first color coordinate for the first sample position of the respective sample cluster, a second color coordinate for the second sample position of the respective sample cluster, a third color coordinate for the third sample position of the respective sample cluster and a fourth color coordinate for the fourth sample position of the respective sample cluster,
wherein the apparatus is configured to decode the image representation into the CFA pattern image by computing
for the third sample position of each sample cluster, the third sample at the respective third sample position by forming a difference of the third color coordinate at the third sample position and a first filtered color coordinate which is derived by forming a third weighted sum of first and second color coordinates located at first and second sample positions neighboring the third sample position in a manner so that a sum of weights of the third weighted sum is larger than one half,
for the fourth sample position of each sample cluster, the third sample at the respective fourth sample position by forming a difference of the fourth color coordinate at the fourth sample position and a second filtered color coordinate which is derived by forming a fourth weighted sum of first and second color coordinates located at first and second sample positions neighboring the fourth sample position in a manner so that a sum of weights of the fourth weighted sum is larger than one half,
for the first sample position of each sample cluster, the first sample at the respective first sample position by forming a summation of the first color coordinate at the first sample position and a first filtered third sample value derived by forming a first weighted sum of third samples computed for third and fourth sample positions neighboring the first sample position,
for the second sample position of each sample cluster, a second sample at the respective second sample position by forming a summation of the second color coordinate at the second sample position and a second filtered third sample value derived by forming a second weighted sum of third samples computed for third and fourth sample positions neighboring the second sample position.

5 more elements

An apparatus for coding a color filter array (CFA) pattern image into an image representation,

wherein the CFA pattern image comprises first samples of a first channel, second samples of a second channel, and third samples of a third channel,
wherein the CFA pattern image is partitioned into sample clusters, each of which comprises first to fourth sample positions with one of the first samples located at the first sample position, one of the second samples located at the second sample position, and two third samples located at the third and fourth sample positions, respectively, wherein the apparatus is configured to code the CFA pattern image into the image representation by computing
for the first sample position of each sample cluster, a first color coordinate so that the first color coordinate is representative of a difference of the first sample at the first sample position and a first filtered third sample value derived by forming a first weighted sum of third samples neighboring the first sample position,
for the second sample position of each sample cluster, a second color coordinate so that the second color coordinate is representative of a difference of the second sample at the second sample position and a second filtered third sample value derived by forming a second weighted sum of third samples neighboring the second sample position,
for the third sample position of each sample cluster, a third color coordinate by forming a summation of the third sample at the third sample position and a first filtered color coordinate which is derived by forming a third weighted sum of first and second color coordinates computed for first and second sample positions neighboring the third sample position in a manner so that a sum of weights of the third weighted sum is larger than one half,
for the fourth sample position of each sample cluster, a fourth color coordinate by forming a summation of the third sample at the fourth sample position and a second filtered color coordinate which is derived by a fourth weighted sum of first and second color coordinates computed for first and second sample positions neighboring the fourth sample position in a manner so that a sum of weights of the fourth weighted sum is larger than one half.

3 more elements

A method for decoding an image representation into a CFA pattern image,

wherein the CFA pattern image comprises first samples of a first channel, second samples of a second channel, and third samples of a third channel,
wherein the CFA pattern image is partitioned into sample clusters, each of which comprises first to fourth sample positions with one of the first samples located at the first sample position, one of the second samples located at the second sample position, and two third samples located at the third and fourth sample positions, respectively,
wherein the method comprises deriving, for each sample cluster, a first color coordinate for the first sample position of the respective sample cluster, a second color coordinate for the second sample position of the respective sample cluster, a third color coordinate for the third sample position of the respective sample cluster and a fourth color coordinate for the fourth sample position of the respective sample cluster,
wherein the method comprises decoding the image representation into the CFA pattern image by computing
for the third sample position of each sample cluster, the third sample at the respective third sample position by forming a difference of the third color coordinate at the third sample position and a first filtered color coordinate which is derived by forming a third weighted sum of first and second color coordinates located at first and second sample positions neighboring the third sample position in a manner so that a sum of weights of the third weighted sum is larger than one half,
for the fourth sample position of each sample cluster, the third sample at the respective fourth sample position by forming a difference of the fourth color coordinate at the fourth sample position and a second filtered color coordinate which is derived by forming a fourth weighted sum of first and second color coordinates located at first and second sample positions neighboring the fourth sample position in a manner so that a sum of weights of the fourth weighted sum is larger than one half,
for the first sample position of each sample cluster, the first sample at the respective first sample position by forming a summation of the first color coordinate at the first sample position and a first filtered third sample value derived by forming a first weighted sum of third samples computed for third and fourth sample positions neighboring the first sample position,
for the second sample position of each sample cluster, a second sample at the respective second sample position by forming a summation of the second color coordinate at the second sample position and a second filtered third sample value derived by forming a second weighted sum of third samples computed for third and fourth sample positions neighboring the second sample position.

5 more elements

A method for coding a color filter array (CFA) pattern image into an image representation,

wherein the CFA pattern image comprises first samples of a first channel, second samples of a second channel, and third samples of a third channel,
wherein the CFA pattern image is partitioned into sample clusters, each of which comprises first to fourth sample positions with one of the first samples located at the first sample position, one of the second samples located at the second sample position, and two third samples located at the third and fourth sample positions, respectively, wherein the method comprises coding the CFA pattern image into the image representation by computing
for the first sample position of each sample cluster, a first color coordinate so that the first color coordinate is representative of a difference of the first sample at the first sample position and a first filtered third sample value derived by forming a first weighted sum of third samples neighboring the first sample position,
for the second sample position of each sample cluster, a second color coordinate so that the second color coordinate is representative of a difference of the second sample at the second sample position and a second filtered third sample value derived by forming a second weighted sum of third samples neighboring the second sample position,
for the third sample position of each sample cluster, a third color coordinate by forming a summation of the third sample at the third sample position and a first filtered color coordinate which is derived by forming a third weighted sum of first and second color coordinates computed for first and second sample positions neighboring the third sample position in a manner so that a sum of weights of the third weighted sum is larger than one half,
for the fourth sample position of each sample cluster, a fourth color coordinate by forming a summation of the third sample at the fourth sample position and a second filtered color coordinate which is derived by a fourth weighted sum of first and second color coordinates computed for first and second sample positions neighboring the fourth sample position in a manner so that a sum of weights of the fourth weighted sum is larger than one half.

3 more elements

A non-transitory digital storage medium having stored thereon a computer program for performing a method for decoding an image representation into a CFA pattern image,

wherein the CFA pattern image comprises first samples of a first channel, second samples of a second channel, and third samples of a third channel,
wherein the CFA pattern image is partitioned into sample clusters, each of which comprises first to fourth sample positions with one of the first samples located at the first sample position, one of the second samples located at the second sample position, and two third samples located at the third and fourth sample positions, respectively,
wherein the method comprises deriving, for each sample cluster, a first color coordinate for the first sample position of the respective sample cluster, a second color coordinate for the second sample position of the respective sample cluster, a third color coordinate for the third sample position of the respective sample cluster and a fourth color coordinate for the fourth sample position of the respective sample cluster,
wherein the method comprises decoding the image representation into the CFA pattern image by computing
for the third sample position of each sample cluster, the third sample at the respective third sample position by forming a difference of the third color coordinate at the third sample position and a first filtered color coordinate which is derived by forming a third weighted sum of first and second color coordinates located at first and second sample positions neighboring the third sample position in a manner so that a sum of weights of the third weighted sum is larger than one half,
for the fourth sample position of each sample cluster, the third sample at the respective fourth sample position by forming a difference of the fourth color coordinate at the fourth sample position and a second filtered color coordinate which is derived by forming a fourth weighted sum of first and second color coordinates located at first and second sample positions neighboring the fourth sample position in a manner so that a sum of weights of the fourth weighted sum is larger than one half,
for the first sample position of each sample cluster, the first sample at the respective first sample position by forming a summation of the first color coordinate at the first sample position and a first filtered third sample value derived by forming a first weighted sum of third samples computed for third and fourth sample positions neighboring the first sample position,
for the second sample position of each sample cluster, a second sample at the respective second sample position by forming a summation of the second color coordinate at the second sample position and a second filtered third sample value derived by forming a second weighted sum of third samples computed for third and fourth sample positions neighboring the second sample position,
when said computer program is run by a computer.

6 more elements

A non-transitory digital storage medium having stored thereon a computer program for performing a method for coding a color filter array (CFA) pattern image into an image representation,

wherein the CFA pattern image comprises first samples of a first channel, second samples of a second channel, and third samples of a third channel,
wherein the CFA pattern image is partitioned into sample clusters, each of which comprises first to fourth sample positions with one of the first samples located at the first sample position, one of the second samples located at the second sample position, and two third samples located at the third and fourth sample positions, respectively, wherein the method comprises coding the CFA pattern image into the image representation by computing
for the first sample position of each sample cluster, a first color coordinate so that the first color coordinate is representative of a difference of the first sample at the first sample position and a first filtered third sample value derived by forming a first weighted sum of third samples neighboring the first sample position,
for the second sample position of each sample cluster, a second color coordinate so that the second color coordinate is representative of a difference of the second sample at the second sample position and a second filtered third sample value derived by forming a second weighted sum of third samples neighboring the second sample position,
for the third sample position of each sample cluster, a third color coordinate by forming a summation of the third sample at the third sample position and a first filtered color coordinate which is derived by forming a third weighted sum of first and second color coordinates computed for first and second sample positions neighboring the third sample position in a manner so that a sum of weights of the third weighted sum is larger than one half,
for the fourth sample position of each sample cluster, a fourth color coordinate by forming a summation of the third sample at the fourth sample position and a second filtered color coordinate which is derived by a fourth weighted sum of first and second color coordinates computed for first and second sample positions neighboring the fourth sample position in a manner so that a sum of weights of the fourth weighted sum is larger than one half,
when said computer program is run by a computer.

4 more elements

Dependencies are inferred from the claim text.

Facts

Application: 18/045,859Filed 12 Oct 2022
Priority: 16 Apr 2020
Published: 23 Feb 2023
Term ends: Not estimatedThe source records don’t support an estimate.

Classification5 codes

H04N 19/60
Methods or arrangements for coding, decoding, compressing or decompressing digital video signals: using transform coding
H04N 19/593
Using predictive coding: involving spatial prediction techniques
H04N 23/843
For processing colour signals: demosaicing, e.g. interpolating colour pixel values
H04N 25/134
Characterised by the spectral characteristics of the filter elements: based on three different wavelength filter elements

Family

Application18/045,859Continuation
US 12,063,391 B2Related patent
US 2023/0055561 A1This patent

From the US filing history. Foreign family members aren’t listed.

Timeline

Citations

Open elsewhere

Building an agent or pulling many patents? Use the API or MCP server instead of scraping pages. Get an API key

patentgrep

Loading patent…

Abstract

An apparatus for decoding an image representation into a color filter array (CFA) pattern image,

wherein the CFA pattern image comprises first samples of a first channel, second samples of a second channel, and third samples of a third channel,
wherein the CFA pattern image is partitioned into sample clusters, each of which comprises first to fourth sample positions with one of the first samples located at the first sample position, one of the second samples located at the second sample position, and two third samples located at the third and fourth sample positions, respectively,
wherein the apparatus is configured to derive, for each sample cluster, a first color coordinate for the first sample position of the respective sample cluster, a second color coordinate for the second sample position of the respective sample cluster, a third color coordinate for the third sample position of the respective sample cluster and a fourth color coordinate for the fourth sample position of the respective sample cluster,
wherein the apparatus is configured to decode the image representation into the CFA pattern image by computing
for the third sample position of each sample cluster, the third sample at the respective third sample position by forming a difference of the third color coordinate at the third sample position and a first filtered color coordinate which is derived by forming a third weighted sum of first and second color coordinates located at first and second sample positions neighboring the third sample position in a manner so that a sum of weights of the third weighted sum is larger than one half,
for the fourth sample position of each sample cluster, the third sample at the respective fourth sample position by forming a difference of the fourth color coordinate at the fourth sample position and a second filtered color coordinate which is derived by forming a fourth weighted sum of first and second color coordinates located at first and second sample positions neighboring the fourth sample position in a manner so that a sum of weights of the fourth weighted sum is larger than one half,
for the first sample position of each sample cluster, the first sample at the respective first sample position by forming a summation of the first color coordinate at the first sample position and a first filtered third sample value derived by forming a first weighted sum of third samples computed for third and fourth sample positions neighboring the first sample position,
for the second sample position of each sample cluster, a second sample at the respective second sample position by forming a summation of the second color coordinate at the second sample position and a second filtered third sample value derived by forming a second weighted sum of third samples computed for third and fourth sample positions neighboring the second sample position.

5 more elements

Facts

Application

18/045,859Filed 12 Oct 2022

Priority

16 Apr 2020

Published

23 Feb 2023

Term ends

Not estimatedThe source records don’t support an estimate.

Classification5 codes

H04N 19/60
Methods or arrangements for coding, decoding, compressing or decompressing digital video signals: using transform coding
H04N 19/593
Using predictive coding: involving spatial prediction techniques
H04N 23/843
For processing colour signals: demosaicing, e.g. interpolating colour pixel values
H04N 25/134
Characterised by the spectral characteristics of the filter elements: based on three different wavelength filter elements