WO2000018125A1 - Procede et dispositif pour le codage d'une image numerisee, et procede et dispositif pour le decodage d'une image numerisee codee - Google Patents

Procede et dispositif pour le codage d'une image numerisee, et procede et dispositif pour le decodage d'une image numerisee codee Download PDF

Info

Publication number
WO2000018125A1
WO2000018125A1 PCT/DE1999/002741 DE9902741W WO0018125A1 WO 2000018125 A1 WO2000018125 A1 WO 2000018125A1 DE 9902741 W DE9902741 W DE 9902741W WO 0018125 A1 WO0018125 A1 WO 0018125A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
pixels
partial
determined
digitized
Prior art date
Application number
PCT/DE1999/002741
Other languages
German (de)
English (en)
Inventor
Gero BÄSE
Original Assignee
Siemens Aktiengesellschaft
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Siemens Aktiengesellschaft filed Critical Siemens Aktiengesellschaft
Publication of WO2000018125A1 publication Critical patent/WO2000018125A1/fr

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/503—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
    • H04N19/51—Motion estimation or motion compensation
    • H04N19/523—Motion estimation or motion compensation with sub-pixel accuracy

Definitions

  • the invention relates to a method and an arrangement for encoding a digitized image as well as a method and an arrangement for decoding a encoded digitized image.
  • Each image in a recorded image sequence is digitized and has pixels. Coding information is assigned to the pixels.
  • Coding information is further to be understood as luminance information (brightness value) and / or chrominance information (color value).
  • the pixels are grouped into several image areas, so-called image blocks, with usually 8x8 or 16x16 pixels.
  • each image block is subjected to a discrete cosine transformation (DCT), the spectral coefficients determined from them are quantized and subjected to entropy coding.
  • DCT discrete cosine transformation
  • the encoded information is decoded again when the digitized image is encoded, ie the quantized spectral coefficients are inversely quantized and subjected to an inverse discrete cosine transformation (IDCT).
  • IDCT inverse discrete cosine transformation
  • a motion estimation is carried out for an image to be encoded for each image area of the image to be encoded.
  • an area is determined in a temporally preceding or subsequent image in a predetermined search area which best corresponds to the image area to be encoded with regard to a predetermined criterion.
  • the determination is made in such a way that a sum over the squares of the differences in the individual luminance values corresponds to the pixels of the image block to be coded and the image area to be compared of the temporally subsequent or temporally preceding image.
  • the movement is estimated in such a way that a higher resolution than that specified by the spacing of the pixels from one another is taken into account.
  • partial pixels are inserted between the individual pixels and coding information is assigned to the partial pixels, which results from interpolation or filtering of the coding information of adjacent pixels or partial pixels.
  • a sub-pixel is usually inserted in the middle between two pixels. This procedure is also referred to as field motion estimation.
  • the result of the motion estimation is a pixel motion vector and a field pixel motion vector.
  • the pixel motion vector is used to indicate the pixels by which the determined image area must be shifted from the preceding or following image so that it corresponds to the image block to be coded.
  • the sub-pixel motion vector specifies by which sub-pixels, based on the pixel motion vector, the area of the temporally preceding or following image must be shifted in order to be able to replace the image block to be coded.
  • coding information is no longer transformed at all (DCT), but only the pixel motion vector and the partial pixel motion vector are encoded.
  • an error signal can additionally be coded and transmitted, which is formed by the difference between the coding information assigned to the pixels of the picture block to be coded and the coding information of the determined picture area from the temporally preceding or the temporally following picture.
  • An increase in the data compression achieved is achieved in the method known from [1] in that pictures in the arrangement for coding are not coded, ie are left out.
  • the omitted pictures are reconstructed by unidirectional or bidirectional prediction of intermediate pictures in the arrangement for decoding.
  • the method known from [1] and the arrangement known from [1] has in particular the disadvantage that an increased transmission rate is required to transmit the pixel motion vectors.
  • the invention is therefore based on the problem of specifying methods and arrangements for encoding and decoding a digitized image in which the data rate required for transmitting the encoded digitized image is reduced.
  • a method for coding a digitized image using a temporally preceding or temporally following image has the following steps: a) the digitized image and the temporally preceding or temporally following image have pixels which are each grouped into several image areas, b) for the Image areas of the digitized image are each subjected to a motion estimation with an increased resolution than is specified by the spacing of the image points from one another, c) the motion estimation being carried out in such a way that the following steps are carried out:
  • a pixel motion vector is determined, which is used to indicate by how many pixels the determined image area in the digitized image is shifted relative to a corresponding image area in the previous or subsequent image, and
  • a partial pixel motion vector is determined, which is used to indicate by how many partial pixels, based on the increased resolution, the determined image area in the digitized image compared to the corresponding image based on the pixel motion vector. area in the temporally preceding or temporally following image is shifted, and d) in which the sub-pixel motion vectors as well as a motion estimation indication, with which the manner in which the pixel motion vectors are formed, are encoded.
  • the following steps have: a) the coded Sub-pixel motion vectors and the coded motion estimation information are decoded, b) the digitized image and the temporally preceding or following image have pixels, each grouped into several image areas rt, and c) for the image areas of the digitized image, movement compensation is carried out in such a way that the following steps are carried out:
  • a pixel motion vector is determined using the motion estimation information, by means of which it is specified by how many pixels the determined image area in the digitized image is shifted relative to a corresponding image area in the temporally preceding or following image, and
  • An arrangement for coding a digitized picture using a temporally preceding or following temporally picture has a processor which is set up in such a way that the following steps can be carried out: a) the digitized picture and the temporally preceding or following picture have picture elements , which are each grouped into several image areas, b) for the image areas of the digitized image, a motion estimate is carried out with an increased resolution than is specified by the spacing of the image points from one another, c) the motion estimate is carried out in such a way that the following steps are carried out :
  • a pixel motion vector is determined, which is used to indicate by how many pixels the determined image area in the digitized image is shifted relative to a corresponding image area in the temporally preceding or following image, and
  • a sub-pixel motion vector is determined, with which it is specified how many sub-pixels according to the increased resolution, based on the pixel motion vector, shift the determined image area in the digitized image relative to the corresponding image area in the temporally preceding or following image d) the sub-pixel motion vectors as well as a motion estimation indication with which the manner in which the pixel motion vectors are formed are encoded.
  • An arrangement for decoding a coded digitized picture using a temporally preceding or following picture and coded partial picture Point motion vectors as well as a coded motion estimation specification with which it is specified how pixel motion vectors are formed, with the partial pixel motion vectors being used to indicate how many partial pixels according to a predetermined increased resolution based on the pixel motion vector the determined Image area in the digitized image is shifted from the corresponding image area in the temporally preceding or following image has a processor which is set up in such a way that the following steps can be carried out: a) the coded partial pixel motion vectors and the coded motion estimation Information is decoded, b) the digitized image and the temporally previous or temporally subsequent image have pixels which are each grouped into a plurality of image areas, c) for the image areas of the digitized image, motion compensation de takes place in each case This means that the following steps are carried out: • Using the motion estimation information, a pixel motion vector is determined, which is used to indicate
  • the invention achieves increased data compression for encoding or decoding the digitized image.
  • the image areas are preferably image blocks.
  • the increased resolution is preferably formed by partial pixels which are inserted between the pixels at a predetermined distance, preferably in the middle of the distance between a plurality of pixels and / or partial pixels.
  • a motion estimation with an increased resolution is carried out several times for the image areas of the digitized image. Each time a partial pixel motion vector is determined, a different starting point within the image area is assumed. In each case the best sub-pixel motion vector with regard to a predetermined criterion is selected and coded.
  • An arbitrarily predeterminable number of images preceding or following in time can be used for coding or decoding.
  • FIG. 1 shows an arrangement of two computers, a camera and a screen, with which the coding, the over- transmission as well as the decoding and presentation of the image data
  • FIG. 2 shows a sketch of an arrangement for block-based coding of a digitized image
  • FIG. 1 shows an arrangement which comprises two computers 102, 108 and a camera 101, image coding, transmission of the image data and image decoding being illustrated.
  • a camera 101 is connected to a first computer 102 via a line 119.
  • the camera 101 transmits recorded images 104 to the first computer 102.
  • the first computer 102 has a first processor 103 which is connected to an image memory 105 via a bus 118.
  • a method for image coding is carried out with the first processor 103 of the first computer 102.
  • Image data 106 encoded in this way is transmitted from the first computer 102 to a second computer 108 via a communication link 107, preferably a line or a radio link.
  • the second computer 108 contains a second processor 109 which is connected to an image memory 111 via a bus 110.
  • a method for image decoding is carried out with the second processor 109.
  • Both the first computer 102 and the second computer 108 each have a screen 112 or 113, on which the image data 104 are visualized, the visualization on the screen 112 of the first computer 102 usually being carried out only for control purposes.
  • Input units are provided for operating both the first computer 102 and the second computer 108, preferably a keyboard 114 or 115, and a computer mouse 116 or 117.
  • the image data 104 which are transmitted from the camera 101 via the line 119 to the first computer 102, are data in the time domain, while the data 106, which are from the first Computer 102 are transmitted to the second computer 108 via the communication link 107, which are image data in the spectral range.
  • the decoded image data is displayed on the screen 113.
  • FIG. 2 shows a sketch of an arrangement for carrying out a block-based image coding method according to the H.263 standard (see [1]).
  • a video data stream to be encoded with chronologically successive digitized images is fed to an image coding unit 201.
  • the digitized images are subdivided into macro blocks 202, each macro block containing 16 ⁇ 16 pixels.
  • the macro block 202 comprises 4 picture blocks 203, 204, 205 and 206, each picture block containing 8x8 picture elements to which luminance values (brightness values) are assigned.
  • each macroblock 202 comprises two chrominance blocks 207 and 208 with chrominance values (color information, color saturation) assigned to the pixels.
  • luminance value and color difference values are referred to as color values.
  • the image blocks are fed to a transformation coding unit 209.
  • a transformation coding unit 209 In the case of differential image coding, values to be coded from image blocks of temporally preceding images are subtracted from the image blocks currently to be coded; only the difference formation information 210 is supplied to the transformation coding unit (Discrete Cosine Transformation, DCT) 209.
  • the current macroblock 202 is communicated to a motion estimation unit 229 via a connection 234.
  • spectra for the picture blocks or difference picture blocks to be coded are tral coefficients 211 formed and fed to a quantization unit 212.
  • Quantized spectral coefficients 213 are supplied to both a scan unit 214 and an inverse quantization unit 215 in a reverse path.
  • entropy coding is carried out on the scanned spectral coefficients 232 in an entropy coding unit 216 provided for this purpose.
  • the entropy-coded spectral coefficients are transmitted as coded image data 217 to a decoder via a channel, preferably a line or a radio link.
  • Spectral coefficients 218 obtained in this way are fed to an inverse transformation coding unit 219 (inverse discrete cosine transformation, IDCT).
  • Reconstructed coding values (also difference coding values) 220 are supplied to an adder 221 in the differential image mode.
  • Reconstructed image blocks 222 are formed with the adder 221 and stored in an image memory 223.
  • Chrominance values 224 of the reconstructed image blocks 222 are fed from the image memory 223 to a motion compensation unit 225.
  • an interpolation takes place in an interpolation unit 227 provided for this purpose.
  • the number of brightness values contained in the respective image block is preferably quadrupled.
  • All brightness values 228 are supplied to both the motion compensation unit 225 and the motion estimation unit 229.
  • the motion estimation unit 229 also receives the picture blocks of the macro block to be coded in each case (16x16 picture elements) via the connection 234. Motion estimation unit 229, the motion estimation takes into account the interpolated brightness values ("motion estimation on a half-pixel basis").
  • the result of the motion estimation is a motion vector 230, by means of which a local shift of the selected macroblock from the temporally preceding image to the macroblock 202 to be coded is expressed.
  • Both brightness information and chrominance information relating to the macroblock determined by the motion estimation unit 229 are shifted by the motion vector 230 and subtracted from the coding values of the macroblock 202 (see data path 231).
  • the result of the motion estimation is thus the motion vector 230 with two motion vector components, a first motion vector component BV X and a second motion vector component BVy along the first direction x and the second direction y:
  • the motion vector 230 is assigned to the image block.
  • the image coding unit from FIG. 2 thus supplies a motion vector 230 for all image blocks or macro image blocks.
  • a pixel motion vector BBV which represents the shift in the image area determined with the motion estimation of the time describes the preceding or temporally subsequent image on the respective image point and a sub-image point motion vector TBV is determined, by means of which it is described how many partial image points and by which local displacement, based on the image point to which the image area has been moved, the image area is shifted , so that the entire local shift of the determined image area corresponds to the shift according to the determined motion vector.
  • the error signal and the motion vector are no longer coded, but rather the respective partial pixel motion vector TBV and a motion estimation information, with which it is specified how the motion vector BV has been determined or is to be determined.
  • the motion estimation information, the partial pixel motion vectors and the coded image information are decoded.
  • a motion estimation is carried out in the context of the reconstruction of the coded image, using previously reconstructed images which have preceded or followed in time in accordance with the procedure which is predetermined by the motion estimation information.
  • the motion vector is thus determined.
  • the pixel motion vector is determined by forming the difference between the motion vector and the partial pixel motion vector.
  • a motion compensation takes place using the determined motion vectors and the pixel motion vectors, with which the image to be reconstructed is reconstructed.
  • motion estimates can be carried out for an image block by starting a motion estimation at different starting positions in the temporally preceding or following image.
  • the determined motion vector of the motion estimation is compared with the other determined motion vectors and the motion vector is selected which is optimal in relation to a predetermined criterion.
  • the motion vector is selected which is optimal in relation to a predetermined criterion.
  • Motion vector selected for which the error signal in the context of the motion estimation is minimal is minimal.
  • the information of the start position for the motion estimation is also encoded and transmitted.
  • the motion vector can also be determined by motion estimation in the order for decoding, since it is now known for it in which pixel the motion estimation must begin.
  • the motion estimate can be determined, for example, according to the following rule:

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)

Abstract

Selon l'invention, l'évaluation d'un déplacement se fait de telle sorte qu'un vecteur de déplacement de points d'image est déterminé, vecteur à l'aide duquel il est indiqué de combien de points d'image la zone d'image déterminée est décalée dans l'image numérisée, par rapport à une zone d'image correspondante de l'image qui précède ou suit, dans le temps, l'image susmentionnée. En outre, un vecteur de déplacement de points d'image partiels est déterminé, à l'aide duquel est indiqué de combien de points d'image partiels, selon une résolution élevée prédéterminée et à partir du vecteur de déplacement de points d'image, la zone d'image déterminée, dans l'image numérisée, est décalée par rapport à la zone d'image correspondante dans l'image qui précède ou suit, dans le temps, l'image susmentionnée. Les vecteurs de déplacement de points d'image partiels ainsi qu'une indication d'évaluation de déplacement, qui indique de quelle façon les vecteurs de déplacement de points d'image sont formés, sont codés.
PCT/DE1999/002741 1998-09-23 1999-09-01 Procede et dispositif pour le codage d'une image numerisee, et procede et dispositif pour le decodage d'une image numerisee codee WO2000018125A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
DE19843618.1 1998-09-23
DE19843618 1998-09-23

Publications (1)

Publication Number Publication Date
WO2000018125A1 true WO2000018125A1 (fr) 2000-03-30

Family

ID=7881964

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/DE1999/002741 WO2000018125A1 (fr) 1998-09-23 1999-09-01 Procede et dispositif pour le codage d'une image numerisee, et procede et dispositif pour le decodage d'une image numerisee codee

Country Status (1)

Country Link
WO (1) WO2000018125A1 (fr)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0468279A2 (fr) * 1990-07-24 1992-01-29 ANT Nachrichtentechnik GmbH Procédé pour déterminer des vecteurs de mouvement pour les blocs d'une séquence d'images de source
EP0579844A1 (fr) * 1992-02-08 1994-01-26 Samsung Electronics Co., Ltd. Procede et appareil d'evaluation de mouvement
EP0652676A1 (fr) * 1993-11-08 1995-05-10 Sony Corporation Dispositif et méthode pour comprimer un signal numérique d'images en mouvement
EP0723366A2 (fr) * 1995-01-17 1996-07-24 Graphics Communications Laboratories Procédé d'estimation du mouvement et appareil pour calculer un vecteur de mouvement
EP0895425A2 (fr) * 1997-07-28 1999-02-03 Victor Company of Japan, Ltd. Codeur à compensation de mouvement avec contrÔle de la précision des vecteurs de mouvement

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0468279A2 (fr) * 1990-07-24 1992-01-29 ANT Nachrichtentechnik GmbH Procédé pour déterminer des vecteurs de mouvement pour les blocs d'une séquence d'images de source
EP0579844A1 (fr) * 1992-02-08 1994-01-26 Samsung Electronics Co., Ltd. Procede et appareil d'evaluation de mouvement
EP0652676A1 (fr) * 1993-11-08 1995-05-10 Sony Corporation Dispositif et méthode pour comprimer un signal numérique d'images en mouvement
EP0723366A2 (fr) * 1995-01-17 1996-07-24 Graphics Communications Laboratories Procédé d'estimation du mouvement et appareil pour calculer un vecteur de mouvement
EP0895425A2 (fr) * 1997-07-28 1999-02-03 Victor Company of Japan, Ltd. Codeur à compensation de mouvement avec contrÔle de la précision des vecteurs de mouvement

Similar Documents

Publication Publication Date Title
DE69530336T2 (de) Bewegungskompensation für digitale Videosignale mit Zeilensprung
DE69021500T2 (de) Bedingungsgemässe bewegungskompensierte Interpolation von digitalen Bewegtbildsignalen.
DE69733007T2 (de) Vorrichtung zur codierung und decodierung von bewegtbildern
DE69834902T2 (de) Bewegungskompensierte prädiktive bildkodierung und -dekodierung
DE19704439C2 (de) Verfahren und Vorrichtung zur Bewegungsschätzung in einem digitalen Videocodierer unter Verwendung von Trajektorien
DE69909364T2 (de) Vorrichtung und Methode zur Änderung der Grösse eines komprimiertes Bildes
DE19541457C1 (de) Verfahren zur Codierung eines Videodatenstroms einer aus Bildblöcken bestehenden Videosequenz
DE69919205T2 (de) Verfahren und gerät für die mpeg-bildkodierung von eingefügten bildern
DE69915843T2 (de) Teilbandkodierung/-dekodierung
EP1941744B1 (fr) Procede et dispositifs de determination et de reconstruction d'une zone d'image predite
EP1101196B1 (fr) Procede et dispositif d'evaluation du mouvement dans une image numerisee possedant des pixels
EP1110407B1 (fr) Procede et dispositif pour le codage et le decodage d'une image numerisee faisant appel a un vecteur de deplacement total
EP1116184B1 (fr) Procede et systeme de traitement d'une image numerisee dotee de points d'images
EP0985317B1 (fr) Procede de codage et de decodage d'une image numerisee
WO1998024239A1 (fr) Procede de codage d'une image numerisee
DE19749604A1 (de) Verfahren zum Kodieren eines Modus beim Kodieren binärer Formen
EP1285537B1 (fr) Procede et agencement pour le codage et le decodage d'une suite d'images
WO2000018125A1 (fr) Procede et dispositif pour le codage d'une image numerisee, et procede et dispositif pour le decodage d'une image numerisee codee
DE19951341A1 (de) Verfahren zur bewegungskompensierenden Prädiktion von Bewegtbildern sowie Einrichtung hierzu
WO2001049038A1 (fr) Procede, dispositif et produit programme informatique servant a effectuer une prediction lors du codage d'une image divisee en blocs d'image
EP1121809B1 (fr) Procede et dispositif pour le codage d'une image numerisee, procede et dispositif pour le decodage d'une image numerisee
EP1157557A1 (fr) Procede et dispositif pour la transformation d'une zone d'image
WO2001062009A1 (fr) Procede et dispositif de codage, ou de codage et decodage d'une suite de nombres
WO2000049525A1 (fr) Procede et dispositif pour memoriser au moins une image avec les donnees relationnelles qui s'y rapportent
WO2000065539A1 (fr) Procede et dispositif pour la lecture et la memorisation, assistees par ordinateur, de coefficients de transformation, et produits programmes informatiques et supports d'information lisibles par un ordinateur

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): JP KR US

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE

DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
121 Ep: the epo has been informed by wipo that ep was designated in this application
122 Ep: pct application non-entry in european phase