Back to results
US 6,636,637 B1Granted
Method and arrangement for coding and decoding a digitized image
AssigneeSiemens Aktiengesellschaft
InventorAndre Kaup
Life of this patent
Estimating the term…
Abstract
The method and apparatus have a shape-adapted transformation coding is in which in the intra-image coding and the inter-image coding different shape-adapted transformation codings are carried out. A first shape-adapted transformation coding takes place in the intra-image coding, and a second shape-adapted transformation coding, different from the first shape-adapted image coding, takes place in the inter-image coding.
Claims20
A method for coding a digitized image that have image objects with a number of image points, comprising the steps of:
- performing image coding in one of an intra-image coding mode or in an inter-image coding mode;
- transforming in the intra-image coding mode the image information of the image points;
- transforming in the inter-image coding mode difference image information of image information of two successive images;
- carrying out in the inter-image coding mode a first shape-adapted transformation coding;
- and
- carrying out in the intra-image coding mode a second shape-adapted transformation coding, different from the first shape-adapted transformation coding.
3 more elements
A method for coding a digitized image that have image objects with a number of image points, comprising the steps of:
- performing image coding in one of an intra-image coding mode or in an inter-image coding mode;
- transforming in the intra-image coding mode the image information of the image points;
- transforming in the inter-image coding mode difference image information of image information of two successive images;
- carrying out in the inter-image coding mode a first shape-adapted transformation coding;
- and
- carrying out in the intra-image coding mode a second shape-adapted transformation coding, different from the first shape-adapted transformation coding;
- wherein transformation coefficients cj of the image points xj transformed according to the first shape-adapted transformation coding are formed according to the following rule:
- cj=((2/N))DCT−N(p,k)·xj,
- whereby
- N designates a size of an image vector to be transformed in which the transformed image points are contained,
- DCT−N designates a transformation matrix of size N×N,
- p, k designate indices, with p, k ε(0, N−1).
9 more elements
A method for decoding a digitized image that have image objects with a number of image points, comprising the steps of:
- performing the image decoding in at least one of an intra-image decoding mode or in an inter-image decoding mode;
- transforming in the intra-image decoding mode the image information of the image points;
- transforming in the inter-image decoding mode difference image information of image information of two successive images;
- carrying out in the inter-image decoding mode a first inverse shape-adapted transformation coding;
- and
- carrying out in the intra-image decoding mode a second inverse shape-adapted transformation coding, different from the first inverse shape-adapted transformation coding.
3 more elements
A method for decoding a digitized image that have image objects with a number of image points, comprising the steps of:
- performing the image decoding in at least one of an intra-image decoding mode or in an inter-image decoding mode;
- transforming in the intra-image decoding mode the image information of the image points;
- transforming in the inter-image decoding mode difference image information of image information of two successive images;
- carrying out in the inter-image decoding mode a first inverse shape-adapted transformation coding;
- and
- carrying out in the intra-image decoding mode a second inverse shape-adapted transformation coding, different from the first inverse shape-adapted transformation coding;
- wherein, in the first inverse shape-adapted transformation coding, transformed image points xj are formed from transformation coefficients cj according to the following rule:
- xj=((2/N))(DCT−N(p,k))−1·cj,
- whereby
- N designates a size of the image vector to be transformed in which the transformed image points are contained,
- DCT−N designates a transformation matrix of size N×N,
- p, k designate indices, with p, k ε(0, N−1),
- ( ) − designates an inversion of a matrix.
10 more elements
An arrangement for coding a digitized image having image objects with a number of image points, comprising:
- device for receiving image point of the digitized image;
- and
- a processor unit connected to said device and structured such that
- the image coding takes place in one of an intra-image coding mode or in an inter-image coding mode,
- in the intra-image coding mode the image information of the image points is transformed,
- in the inter-image coding mode difference image information of image information of two successive images is transformed,
- in the inter-image coding mode a first shape-adapted transformation coding is carried out, and
- in the intra-image coding mode a second shape-adapted transformation coding, different from the first shape-adapted transformation coding, is carried out.
5 more elements
A arrangement for coding a digitized image having image objects with a number of image points, comprising:
- device for receiving image point of the digitized image;
- and
- a processor unit connected to said device and structured such that
- the image coding takes place in one of an intra-image coding mode or in an inter-image coding mode,
- in the intra-image coding mode the image information of the image points is transformed,
- in the inter-image coding mode difference image information of image information of two successive images is transformed,
- in the inter-image coding mode a first shape-adapted transformation coding is carried out, and
- in the intra-image coding mode a second shape-adapted transformation coding, different from the first shape-adapted transformation coding, is carried out;
- wherein the processor unit is set up such that transformation coefficients cj of the image points xj transformed according to the first shape-adapted transformation coding are formed according to the following rule:
- cj=((2/N))DCT−N(p,k)·xj,
- whereby
- N designates a size of an image vector to be transformed in which the transformed image points are contained,
- DCT−N designates a transformation matrix of size N×N,
- p, k designate indices, with p, k ε(0, N−1).
11 more elements
An arrangement for decoding a digitized image having image objects having an arbitrary number of image points, comprising:
- device for receiving image points of the digitized image;
- and
- a processor unit connected to the device and structured such that
- the image decoding takes place in one of an intra-image decoding mode or in an inter-image decoding mode,
- in the intra-image decoding mode the image information of the image points is transformed,
- in the inter-image decoding mode difference image information of image information of two successive images is transformed,
- in the inter-image decoding mode a first inverse shape-adapted transformation coding is carried out, and
- in the intra-image decoding mode a second inverse shape-adapted transformation coding, different from the first inverse shape-adapted transformation coding, is carried out.
5 more elements
A arrangement for decoding a digitized image having image objects having an arbitrary number of image points, comprising:
- device for receiving image points of the digitized image;
- and
- a processor unit connected to the device and structured such that
- the image decoding takes place in one of an intra-image decoding mode or in an inter-image decoding mode,
- in the intra-image decoding mode the image information of the image points is transformed,
- in the inter-image decoding mode difference image information of image information of two successive images is transformed,
- in the inter-image decoding mode a first inverse shape-adapted transformation coding is carried out, and
- in the intra-image decoding mode a second inverse shape-adapted transformation coding, different from the first inverse shape-adapted transformation coding, is carried out;
- wherein the processor unit is structured such that image points xj transformed in the first inverse shape-adapted transformation coding are formed from transformation coefficients cj according to the following rule:
- xj=((2/N))(DCT−N(p,k))−1·cj,
- whereby
- N designates a size of the image vector to be transformed in which the transformed image points are contained,
- DCT−N designates a transformation matrix of size N×N,
- p, k designate indices, with p, k ε(0, N−1),
- ( ) − designates an inversion of a matrix.
12 more elements
Dependencies are inferred from the claim text.