WO2000030360A1 - Compresseur de donnees a vitesse fixe - Google Patents

Compresseur de donnees a vitesse fixe Download PDF

Info

Publication number
WO2000030360A1
WO2000030360A1 PCT/US1999/025571 US9925571W WO0030360A1 WO 2000030360 A1 WO2000030360 A1 WO 2000030360A1 US 9925571 W US9925571 W US 9925571W WO 0030360 A1 WO0030360 A1 WO 0030360A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
value
predetermined number
compression method
data compression
Prior art date
Application number
PCT/US1999/025571
Other languages
English (en)
Inventor
Honary Honary
Original Assignee
Conexant Systems, Inc.
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 Conexant Systems, Inc. filed Critical Conexant Systems, Inc.
Publication of WO2000030360A1 publication Critical patent/WO2000030360A1/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
    • H04N19/90—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using coding techniques not provided for in groups H04N19/10-H04N19/85, e.g. fractals
    • H04N19/98—Adaptive-dynamic-range coding [ADRC]
    • 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
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/186—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being a colour or a chrominance component

Definitions

  • the present invention relates generally to data compression in data processing systems; and, more particularly, it relates to a platform independent, fixed rate data compressor used in data processing systems, such as imaging systems, which maximizes characteristics such as available storage capacity or available data transmission bandwidth while maintaining high-frequency content.
  • Various traditional methods employing data compression seek to encode a plurality of data using the least amount of data possible. Certain traditional methods group the plurality of data into a number of sub-groups. Each of the sub-groups is analyzed and a single value is used to represent the entire sub-group.
  • One specific problem associated with such a method is that high frequency information can be lost during the compression step. That is to say, while the data is compressed in that a single value is used to represent a larger plurality of data, any large variations in the data can be undesirably smoothed over. For example, in image processing systems where there may be a very distinct edge within an image, if the compression method seeks to compress the portion of the image having the distinct edge, then the edge itself could lose image resolution or undesirable artifacts may be generated during reproduction of the image.
  • a traditional method that compressing chroma information of color images is to partition the image into a number of groups, for example, having a 2x2 dimension (i.e., four pixels arranged in a square orientation). An average color is commonly calculated of the four pixels. This average value is then applied to the four pixels themselves. While such a method does reduce the amount of information needed to encode the 2x2 group, any distinct image information contained within the 2x2 group will be lost. If this conventional method of applying an average to the 2x2 group happens to be applied to a group having a steep transition between black and white, the transition itself will be smeared across the group. A substantially gray color will result in the 2x2 group and the distinctive information will be lost.
  • finite impulse response filtering is applied at both an encoder and a decoder portion of an image processing system. While such filtering can be used to reduce the loss of high frequency information of the image, it can be have a high computational cost. Often, data sample rate conversion may be required employing both up and down sampling at both the encoder and decoder.
  • up sampling may take place at only one of the encoder and decoder, but this still requires a high computational investment to accommodate the high speed calculations required to perform the finite impulse response filtering in near real time.
  • a high computational investment to accommodate the high speed calculations required to perform the finite impulse response filtering in near real time.
  • such an investment is simply prohibitive and an alternate method of performing the data compression is employed.
  • this traditional averaging compression scheme invariably is employed during times when high speed data transmission is required.
  • the traditional averaging compression scheme is employed exclusively.
  • This traditional averaging compression scheme can be applied to other data processing systems besides image processing systems.
  • Voice data processing systems may similarly perform an averaging method to reduce the amount of data that must be transmitted between a voice encoder and a voice decoder. Particularly within wireless communication systems where the limits on available bandwidth may be severe, any method by which data may be compressed is desirable.
  • a competing interest is to maintain the audible perceptual character of the voice data. This competing interest is analogous to the desire to preserve the visible perceptual character of image data as much as possible.
  • the undesirable effect of averaging segments of the voice data is the smoothing of distinct sound. High frequency noises will be lost and the listener in such a voice data processing system will have difficulty in distinguishing what is being spoken.
  • a data compression system having processing circuitry that processes a plurality of data into a more compressed plurality of data. For a group of the plurality of data, a certain value is calculated that is representative of the group of the plurality of data. This value may correspond to the intensity, color, or some other such characteristic of the group of the plurality of data.
  • the data are then encoded as a function relative to that value. When appropriate, the encoded data are then used to reconstruct a representation of the given number of the plurality of data. When this process of encoding and decoding various groups of data, the entire plurality of data can be reconstructed.
  • the value is the average value of the characteristic of the group of the plurality of data. If the characteristic is the intensity of the data, then the value corresponds to the average intensity of the group of the plurality of data. If the characteristic is the color of the data, then the value corresponds to the average color of the group of the plurality of data. Also, the plurality of data may be partitioned into a specific number of groups, appropriately chosen for that specific plurality of data, in which the processing circuit may operate in a most efficient manner.
  • the group may be further divided into a sub-groups which are appropriately chosen as a function of the value. For example, if the value is the average of the group of the plurality of data, then the group is further divided into two sub-groups, one sub-group whose individual data are above the average, and the other sub-group whose individual data are below the average. In other embodiments of the invention, these sub-groups may be processed in a similar fashion as the original group. Each sub-group may be further divided thereby providing any number of sub-groups within a given group of the plurality of data.
  • the plurality of data itself may be divided into a number of ranges.
  • color image data may be divided into a predetermined number of ranges of the color spectrum wherein each datum of the group is encoded as a function of where it lies within that color spectrum.
  • the plurality of data could be divided as a function of the predetermined number of ranges of the color spectrum.
  • each sub-group could be processed to calculate the value of that sub-group.
  • Fig. 1 is a system diagram depicting a data compression system built in accordance with the present invention which, in one embodiment, inputs uncompressed data into a data compression system and outputs compressed data.
  • Fig. 2 is a functional block diagram illustrating another embodiment of the present invention, wherein an operational method is employed to encode a plurality of data using an average of the plurality of data.
  • Fig. 3 is a functional block diagram illustrating another embodiment of the present invention, wherein an operational method is employed to encode a plurality of data using an average of the plurality of data to generate at least two sub-groups of the plurality of data.
  • Fig. 4 is a system diagram illustrating a compression/decompression system that utilizes the operational method of Fig. 3.
  • Fig. 5 is a system diagram of a further embodiment of the present invention illustrating a representation of a group of the plurality of data before compression and after compression.
  • Fig. 6 is a system diagram of a partitioned image spectrum illustrating another embodiment of the present invention.
  • Fig. 7 is a system diagram illustrating a plurality of data partitioned into a number of sub-groups wherein each sub-group has its own sub-average.
  • Fig. 1 is a system diagram depicting a data compression system 100 built in accordance with the present invention which inputs uncompressed data 150 into a data compression system 100 and outputs compressed data 160.
  • the data compression system 100 comprises a processing circuitry 120.
  • processing circuitry 120 comprises a single processing unit such as a digital signal processor (DSP).
  • DSP digital signal processor
  • it includes at least one additional processing element such a CPU that interacts with the DSP to perform compression of the uncompressed data 150. Additional elements may be included without departing from the scope or spirit of the invention.
  • a memory 110 is included in the data compression system 100.
  • the memory 100 assists the processing circuitry 120 in the processing of the uncompressed data 150.
  • a peripheral circuitry 130 and an interface circuitry 140 are also included in the data compression system 100.
  • the peripheral circuitry 130 interfaces with the processing circuitry 120.
  • the interface circuitry 140 is used to communicate with processing resources outside of the data compression system 100.
  • the interface circuitry 140 interfaces with a local area network to communicate the compressed data 160 to various devices.
  • Such various devices include color photo-copy machines, color facsimiles, color printers, black and white printers, digital cameras, and digital printers.
  • the data compression system 100 may be used generate the compressed data 160 for a wide variety of applications including storage, reproduction (either audio or visual data), and transmission through limited bandwidth communication media.
  • Fig. 2 is a functional block diagram illustrating another embodiment of the present invention.
  • an operational method 200 is employed to encode a plurality of data using an average of the plurality of data.
  • a certain group of data is identified.
  • processing circuitry 120 of Fig. 1 is used to perform the identification of the plurality of data.
  • the plurality of data is intelligently partitioned to assist in subsequent processing of the data.
  • the plurality of data is broken into sub-groups that are more amenable to data transfer and processing using the available processing resources of the data processing system.
  • an average value is calculated from the plurality of data.
  • the AVG is representative of at least one characteristic of the plurality of data.
  • the characteristic is pixel intensity in certain embodiments employed in image data processing systems.
  • the characteristic is color in other embodiments.
  • the characteristic is the spectral content of the speech signal. Any number of characteristics may be used to calculate an average in various embodiments of the present invention without departing from the scope and spirit of the invention.
  • the plurality of data is encoded as a function relative to the average calculated in the block 220.
  • the encoding divides the plurality of data into two groups. One group consists of data whose individual value of the characteristic is above the average calculated in the block 220. The other group consists of data whose individual value of the characteristic is below the average calculated in the block 220.
  • the plurality of data into two groups is partitioned into multiple groups. One group consists of data whose individual value of the characteristic is one standard deviation below the average calculated in the block 220. Another group consists of data whose individual value of the characteristic is two standard deviation below the average calculated in the block 220.
  • any number of groups can be encoded as a function relative to the average of the plurality of data.
  • the resultant is a plurality of data wherein each datum of the plurality of data is encoded as a function relative to the average of the characteristic of the plurality of data.
  • Fig. 3 is a functional block diagram illustrating another embodiment of the present invention.
  • an operational method 300 is employed by the invention to encode a plurality of data using an average of the plurality of data to generate at least two sub-groups of the plurality of data.
  • a plurality of data is analyzed.
  • the analysis involves the identification of various patterns within the plurality of data. For example, the analysis identifies those portions of data that have substantially similar characteristics.
  • the plurality of data is sub-grouped according to the similarity of the characteristic in certain portions of the data.
  • the plurality of data is sub-grouped in an optimal manner.
  • the optimal sub-grouping included maintaining a fixed number of data in each sub-group to maximize the available processing resources of the data compression system 100.
  • a decision is performed in a block 340 of whether or not to perform compression on a specific sub-group.
  • a given sub-group may consist of completely uniform data.
  • Such a subgroup may be represented with the characteristics of any one datum within the subgroup.
  • image data an example of such a uniform distribution is a portion of the plurality of data having a constant color and intensity. If in the block 340 the decision is made not to perform data compression on the sub-group of data, then the method terminates.
  • an average value is calculated of at least one characteristic of the sub-group of data in a block 350.
  • the characteristic from which the average is calculated may be chosen from a variety of characteristics.
  • the actual value associated with a given datum of the plurality of data is compared to the average value calculated in the block 350.
  • the average value is compared to a specific value of a given pixel (PIX) selected from a plurality of image data in the block 360.
  • the given pixel When the actual value of the given pixel is greater than or equal to the average value calculated in the block 350, then the given pixel is encoded with a value of zero in a block 364. When the actual value of the given pixel is less than the average value calculated in the block 350, then the given pixel is encoded with a value of one in a block 368.
  • the values of zero and one include data bits in certain embodiments of the invention.
  • an average low value AVG_LOW
  • an average high value For those pixels that are encoded with a value of one in the block 368, an average high value (AVG_HIGH) is calculated in a block 370.
  • the average high and average low values correspond only to the data that have been partitioned into the respective groups of being above or below the average calculated for the sub-group.
  • Fig. 4 is a system diagram illustrating another embodiment of the present invention.
  • a compression/decompression system 400 utilizes the operational method of Fig. 3.
  • compression is performed on the plurality of data.
  • the calculated values of PIX, AVG HIGH, and AVG LOW are either transmitted across communication media or stored in storage media.
  • Examples of communication media include local area networks, internet, and wireless communication systems.
  • Examples of storage media include computer hard drives and optical data storage devices.
  • the compressed data is then decompressed in a block 430.
  • the decompression is performed to reproduce the plurality of data in a visible manner.
  • the reproduced plurality of data in substantially perceptually indistinguishable from the original plurality of data before compression.
  • all the data in a sub-group that have been encoded with AVG_HIGH or AVG_LOW will be reproduced to posses the same characteristic.
  • AVG HIGH or AVG LOW values correspond to the color of pixels within image data
  • all the data that were encoded with the value AVG_HIGH will be reproduced with the same value AVG_HIGH for color upon decompression.
  • all the data that were encoded with the value AVG_LOW will be reproduced with the same value AVG_LOW for color upon decompression.
  • FIG. 5 is a system diagram illustrating a representation 500 of a group of the plurality of data before compression and after compression.
  • An unprocessed pixel group 510 is compressed and decompressed in a block 520 resulting in a processed pixel group 530.
  • each of the data having a value of 'x' lies above the average of the entire unprocessed pixel group 510.
  • Each of the data having a value of 'y' lies below the average of the entire unprocessed pixel group 510.
  • those data having a value of 'x' are encoded to a value of 'w' in the processed pixel group 530.
  • Those data having a value of 'y' are encoded to a value of 'z' in the processed pixel group 530.
  • Fig. 6 is a system diagram illustrating a partitioned image spectrum 600 in another embodiment of the present invention.
  • An image data spectrum 610 represents the entirety required to represent a plurality of image data.
  • the image data spectrum 610 is a spectrum of light intensity of the plurality of data. In other embodiments, it is a spectrum of the color variation of the plurality of data. In still other embodiments, it is a spectrum of the plurality of data itself.
  • the image data spectrum 610 is partitioned into a plurality of sub-spectra, wherein each sub-spectrum has a sub-average.
  • the image data spectrum 610 is partitioned into 'N' sub-spectra (see sub-spectrum N that has Sub-AvgN 620).
  • the individual data of the plurality of data are represented by the plurality of sub-spectra, wherein each sub-spectrum has a sub- average.
  • Fig. 7 is a system diagram that illustrates a plurality of data 700 partitioned into a number of sub-groups (see Sub-Group N 710) wherein each sub-group has its own sub-average as described in Fig. 6. in another embodiment of the present invention.
  • Fig. 7 shows the plurality of data 700 having a two-dimensional distribution, those having skill in the art appreciate that the plurality of data 700 may take any shape including a plurality of one dimensional arrays.
  • the method in which the plurality of data 700 is partitioned into various sub-groups may be performed in any manner described above.
  • the data is partitioned into sub-groups to maximize the available processing resources of the data compression system 100. In other embodiments, the data is partitioned into sub-groups as a function of the characteristics of the data itself as described above with respect to Fig. 3.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Compression Of Band Width Or Redundancy In Fax (AREA)

Abstract

Un compresseur de données à vitesse fixe est utilisé dans des systèmes de traitement de données comprenant des systèmes de traitement de données d'image et des systèmes de traitement de voix-données. Les données sont comprimées pour la maximisation des contraintes de système, telles que l'espace de rangement disponible et la largeur de bande. Dans un mode de réalisation de système de traitement d'image, un groupe de 4x4 pixels est comprimé par le calcul d'une valeur moyenne d'une caractéristique du groupe et par le codage de chaque pixel du groupe en fonction de la moyenne. Le groupe est ensuite divisé en deux sous-groupes dans lesquels chaque pixel est codé comme étant au-dessus ou au-dessous de la moyenne du groupe. Le compresseur de données à vitesse fixe peut être utilisé dans divers systèmes de traitement d'image fonctionnant comme un périphérique ou un dispositif autonome, dont, entre autres, des photocopieuses couleur, des fac-similés couleur, des imprimantes couleur, des imprimantes noir et blanc, des caméras numériques et des imprimantes numériques. La teneur en hautes fréquence des données est conservée dans le schéma de compression utilisé. Dans les données d'image, la conservation des données de haute fréquence permet le maintien de la résolution marginale maximum et la conservation de la qualité de perception visible des données d'image. Dans les voix-données, la préservation des données haute fréquence permet la conservation des sons haute fréquence et la qualité de perception sonore des voix-données.
PCT/US1999/025571 1998-11-16 1999-10-29 Compresseur de donnees a vitesse fixe WO2000030360A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US19281998A 1998-11-16 1998-11-16
US09/192,819 1998-11-16

Publications (1)

Publication Number Publication Date
WO2000030360A1 true WO2000030360A1 (fr) 2000-05-25

Family

ID=22711156

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US1999/025571 WO2000030360A1 (fr) 1998-11-16 1999-10-29 Compresseur de donnees a vitesse fixe

Country Status (1)

Country Link
WO (1) WO2000030360A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103428485A (zh) * 2012-05-24 2013-12-04 富士通株式会社 图像编码方法及图像解码方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5414527A (en) * 1991-08-14 1995-05-09 Fuji Xerox Co., Ltd. Image encoding apparatus sensitive to tone variations
EP0708551A2 (fr) * 1994-10-18 1996-04-24 Fuji Xerox Co., Ltd. Appareil de codage signal d'image qui change entre codage à longueur variable et codage à longueur fixe
EP0741489A1 (fr) * 1992-11-26 1996-11-06 Mitsubishi Denki Kabushiki Kaisha Procédé de codage et décodage d'images

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5414527A (en) * 1991-08-14 1995-05-09 Fuji Xerox Co., Ltd. Image encoding apparatus sensitive to tone variations
EP0741489A1 (fr) * 1992-11-26 1996-11-06 Mitsubishi Denki Kabushiki Kaisha Procédé de codage et décodage d'images
EP0708551A2 (fr) * 1994-10-18 1996-04-24 Fuji Xerox Co., Ltd. Appareil de codage signal d'image qui change entre codage à longueur variable et codage à longueur fixe

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
CHUNG-WOEI CHAO ET AL: "MODIFIED BLOCK TRUNCATION CODING FOR IMAGE COMPRESSION", PATTERN RECOGNITION LETTERS,NL,NORTH-HOLLAND PUBL. AMSTERDAM, vol. 17, no. 14, 30 December 1996 (1996-12-30), pages 1499 - 1506, XP000682585, ISSN: 0167-8655 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103428485A (zh) * 2012-05-24 2013-12-04 富士通株式会社 图像编码方法及图像解码方法

Similar Documents

Publication Publication Date Title
JP2669707B2 (ja) ２次元情報の符号化方法
JP2756515B2 (ja) 可聴信号の知覚符号化方法および音声信号伝送方法
EP0791273B1 (fr) Codage et decodage, temporel et modelise, de signaux d'images
US6993198B2 (en) Encoding apparatus and encoding method
EP0619059A1 (fr) Procede et appareil destines a la compression de donnees d'image et recourant a un codage combine luminance/chrominance
JPH0646403A (ja) マルチレベル信号符号化方法および画像シーケンス統合方法
US5719961A (en) Adaptive technique for encoder and decoder signal transformation
JP5318256B2 (ja) 高ビット深度画像の圧縮
JPH09182069A (ja) 画像圧縮方法および画像圧縮装置
JP2000115548A (ja) デジタル画像の圧縮方法及び圧縮解除方法
JP2583627B2 (ja) デジタルカラービデオ用の統計学的にエンコードされたデータを圧縮および伸張させるための方法および装置
JPH08275156A (ja) 映像信号符号化装置
US8428116B2 (en) Moving picture encoding device, method, program, and moving picture decoding device, method, and program
JPH066829A (ja) 画像データ圧縮方法及び装置
US5978513A (en) Picture encoding method and apparatus and picture decoding method and apparatus
WO2000030360A1 (fr) Compresseur de donnees a vitesse fixe
JPH06225153A (ja) 画像符号化装置
US6029129A (en) Quantizing audio data using amplitude histogram
JP3691180B2 (ja) 符号化装置、符号化方法
JPH10126778A (ja) 画像圧縮方法および装置
EP0427653A1 (fr) Système de compression d'images numériques appartenant à une séquence d'images, à quantification adaptative en fonction d'une information psychovisuelle
JP2000148191A (ja) ディジタルオーディオ信号の符号化装置
JP4441851B2 (ja) 符号化装置および符号化方法、復号装置および復号方法、並びにプログラムおよび記録媒体
JP3057258B2 (ja) ディジタルデータの高能率符号化装置及び方法
JP2002208860A (ja) データ圧縮装置とそのデータ圧縮方法及びデータ圧縮用プログラムを記録したコンピュータ読み取り可能な記録媒体、並びにデータ伸長装置とそのデータ伸長方法

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): JP

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

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