← Back to Patent Explorer
US7206458B2 Expired / Lapsed US ⏱️ 0.0 years remaining

Method of compressing digital images acquired in CFA format

Assignee: STMicroelectronics SRL
•
Filing: 2002-12-13
•
Granted: 2007-04-17
•
Expires: 2022-12-13
📄 Original PDF 🌐 Google Patents 🔄 Sync Now

Abstract

Abstract A method of compressing digital images acquired in CFA format that utilizes optimized quantization matrices. The method, basing itself on the statistical characterization of the error introduced during the processing phase that precedes compression, appropriately modifies the coefficients of any initial quantization matrix, even of a standard type, obtaining a greater compression efficiency without introducing further quality losses.

Inventors

  • 👤 Sebastiano Battiato
  • 👤 Massimo Mancuso

Bibliographic Details

Application Number —
Jurisdiction US
Data Source google_patents
Last Synchronized 2026-11-09 10:16
CPC Classifications
H; H04; H04N; H04N19/00; H04N19/10; H04N19/134; H04N19/154; H04N19/102; H04N19/124; H04N19/169; H04N19/17; H04N19/172; H04N19/186; H04N19/60; H04N19/63

Claims Analysis

Total Claims: 18
Claim 1
Independent Claim 174 words
A method for compressing a digital image, the method comprising: a) acquiring the image in a CFA format; b) performing a full-resolution expansion of the CFA image to generate a complete image having at least one plane; c) synthesizing for the at least one plane of the complete image a quantization matrix by selecting an arbitrary starting matrix Qst, such quantization matrix having elements which are quantization coefficients associated with particular spatial frequencies on the basis of their position; d) assigning to each spatial frequency a statistical measure of the error introduced into said spatial frequency by said at least full-resolution expansion performed on the at least one plane; e) modifying the matrix Qst by multiplying each of its elements by an appropriate weighting coefficient that is a function of the statistical error measure assigned to the corresponding spatial frequency; f) transforming the at least one plane of the complete image into a two-dimensional spatial-frequency domain using a discrete transform; and g) quantizing the at least one transformed plane using the modified matrix Qst.
Claim 2
Independent Claim 174 words
A method for compressing a digital image, the method comprising: a) acquiring the image in a CFA format; b) performing a full-resolution expansion of the CFA image to generate a complete image having at least one plane; c) synthesizing for the at least one plane of the complete image a quantization matrix by selecting an arbitrary starting matrix Qst, such quantization matrix having elements which are quantization coefficients associated with particular spatial frequencies on the basis of their position; d) assigning to each spatial frequency a statistical measure of the error introduced into said spatial frequency by said at least full-resolution expansion performed on the at least one plane; e) modifying the matrix Qst by multiplying each of its elements by an appropriate weighting coefficient that is a function of the statistical error measure assigned to the corresponding spatial frequency; f) transforming the at least one plane of the complete image into a two-dimensional spatial-frequency domain using a discrete transform; and g) quantizing the at least one transformed plane using the modified matrix Qst.
Claim 3
Dependent Claim 18 words
A method in accordance with claim 1 , wherein the digital CFA image is in the Bayer format.
Claim 4
Dependent Claim 17 words
A method in accordance with claim 1 , wherein the complete image, is in the YCrCb format.
Claim 5
Dependent Claim 25 words
A method in accordance with claim 1 , wherein the starting quantization matrix Qst is modified solely for the compression of the luminance Y plane.
Claim 6
Dependent Claim 18 words
A method in accordance with claim 3 , wherein the chrominance planes are sub-sampled before they are transformed.
Claim 7
Dependent Claim 18 words
A method in accordance with claim 1 , wherein the discrete transform is the DCT (Discrete Cosine Transform).
Claim 8
Dependent Claim 21 words
A method in accordance with claim 6 , wherein the transform algorithm divides the planes into square sub-blocks with 8-pixel sides.
Claim 9
Dependent Claim 18 words
A method in accordance with claim 1 , wherein the discrete transform is the DWT (Discrete Wavelet transform).
Claim 10
Dependent Claim 20 words
A method in accordance with claim 1 , wherein the full-resolution expansion is obtained by means of an interpolation algorithm.
Claim 11
Dependent Claim 27 words
A method in accordance with claim 1 , wherein the starting matrix Q st is a quantization matrix optimized in accordance with statistical, content-related, or perceptive criteria.
Claim 12
Dependent Claim 23 words
A method in accordance with claim 1 , wherein the starting matrix Q st is a quantization matrix suggested by the JPEG standard.
Claim 13
Dependent Claim 27 words
A method in accordance with claim 1 , wherein some elements of the quantization matrix obtained by modifying the initial matrix Q st are forced to unity.
Claim 14
Dependent Claim 42 words
A method in accordance with claim 1 , wherein the only elements of the starting matrix Q st to be modified are the elements for which the statistical error measure assigned to the spatial frequency associated with them exceeds a predetermined threshold.
Claim 15
Dependent Claim 44 words
A method in accordance with claim 1 , wherein the only elements of the starting matrix Q st to be modified are the elements for which the statistical error measure assigned to the spatial frequency associated with them does not exceed a predetermined threshold.
Claim 16
Dependent Claim 21 words
A method in accordance with claim 1 , wherein the weighting coefficients are an increasing function of the statistical error measure.
Claim 17
Dependent Claim 38 words
A method in accordance with claim 1 , wherein the statistical error measure assigned to a spatial frequency is the sum of the modules of the errors measured at that particular frequency in a large sample of images.
Claim 18
Dependent Claim 13 words
A digital still camera employing a method in accordance with claim 1 .

USPTO Annuity & Maintenance Windows

Annuity Stage Window Opens Payment Deadline Surcharge Period Until Estimated Fee Status
3.5 Years (Year 4) 2010-04-16 2010-10-16 2011-04-17 $2000.00 Window Closed / Lapsed
7.5 Years (Year 8) 2014-04-16 2014-10-16 2015-04-17 $3760.00 Window Closed / Lapsed
11.5 Years (Year 12) 2018-04-16 2018-10-16 2019-04-17 $7700.00 Window Closed / Lapsed

Assigned Categories & Tags

No categories assigned to this patent yet.

Assign Category Tag