ISO/ECCD15444-1:1999(v1.0,9 December1999 SPIFF colour Spaces, APPn Markers, SPIFF, Compression types and Registration authorities (REGAUT) TU-T Recommendation T87 ISO/EC 14495-1, Lossless and near-lossless compression of continuous-tone still images-baseline ITU-T Recommendation T88 ISO/IEC 14492-1, Lossy/lossless coding of bi-level images Additional references ISO/EC 646: 1991, ISo 7-bit coded character set for information interchange ISo 5807: 1985, Information processing- Documentation symbols and conventions for data, program and system flowcharts, program network charts and system resources chart n.ICC.1:199809 International Electrotechnical Commission. Color management in multimedia systems: Part 2: Colour Management, Part 2-1: Default RGB colour space-SRGB IEC 61966-2-1 1998.9 October 1998 W3C, Extensible Markup Language(XML 1.0), Rec-xm1-19980210 Digital Imaging Group, Flashpix digital image file format. Version 1.0.1. 10 July 1997 Definitions For the purposes of this Recommendation International Standard, the following definitions apply Lx], floor function, Indicates the largest integer not exceeding x. Txl, ceiling function, Indicates the smallest integer not exceeded by x arithmetic coder: An entropy coder used to represent the data compactly auxiliary component: A component from the codestream that is used by the application outside the scope of colour space conversion. For example, an opacity component or a depth component would be an auxiliary bit-plane: A two dimensional array of bits. In this Recommendation Specification a bit-plane refers to all the bits of the same magnitude in all coefficients or samples Could refer to a bit-plane in a component, tile- component, code-block, region of interes bit stream: The actual sequence of bits resulting from the coding of a sequence of symbols. It does not include the marker segments in the d tile headers box: a building block defined by a unique type identifier and length. Some particular boxes may contain other boxes box contents: Refers to the data wrapped within the box structure and stored within the DBox field within the Box data structure box type: Specifies the kind of information that shall be stored with the box and stored within the TBox field within the box data structure cell An optional subdivision of a tile used for low-memory encoding and decoding cleanup pass: A coding pass performed on a single bit-plane of a code-block of coefficients. It is the first pass and only coding pass for the first significant bit-plane; the third and the last pass of all the remaining bit-planes codestream: A collection of one or more bit streams and associated (overhead) information required for their decoding and expansion into image data. Such overhead information is restricted to that required for expansion and may include, but is not limited to, marker segments indicating locations of particular bit streams ndicating transform, quantization and coding types, etc. code-block: A rectangular grouping of coefficients from the same sub-band of a tile-component ITU-T Rec. T 800(1999 CDV10)
ISO/IEC CD15444-1 : 1999 (V1.0, 9 December 1999) 2 ITU-T Rec. T.800 (1999 CDV1.0) SPIFF colour Spaces, APPn Markers, SPIFF, Compression types and Registration authorities (REGAUT). — ITU-T Recommendation T.87 | ISO/IEC 14495-1, Lossless and near-lossless compression of continuous-tone still images-baseline. — ITU-T Recommendation T.88 | ISO/IEC 14492-1, Lossy/lossless coding of bi-level images 2.2 Additional references — ISO/IEC 646:1991, ISO 7-bit coded character set for information interchange. — ISO 5807:1985, Information processing - Documentation symbols and conventions for data, program and system flowcharts, program network charts and system resources charts. — International Color Consortium, ICC profile format specification. ICC.1:1998–09 — International Electrotechnical Commission. Color management in multimedia systems: Part 2: Colour Management, Part 2–1: Default RGB colour space—sRGB. IEC 61966–2–1 1998. 9 October 1998. — W3C, Extensible Markup Language (XML 1.0), Rec-xml-19980210 — Digital Imaging Group, Flashpix digital image file format. Version 1.0.1. 10 July 1997 3 Definitions For the purposes of this Recommendation | International Standard, the following definitions apply. , floor function, Indicates the largest integer not exceeding x. , ceiling function, Indicates the smallest integer not exceeded by x. arithmetic coder: An entropy coder used to represent the data compactly. auxiliary component: A component from the codestream that is used by the application outside the scope of colour space conversion. For example, an opacity component or a depth component would be an auxiliary component. bit-plane: A two dimensional array of bits. In this Recommendation | Specification a bit-plane refers to all the bits of the same magnitude in all coefficients or samples. Could refer to a bit-plane in a component, tilecomponent, code-block, region of interest, or other. bit stream: The actual sequence of bits resulting from the coding of a sequence of symbols. It does not include the marker segments in the main and tile headers. box: A building block defined by a unique type identifier and length. Some particular boxes may contain other boxes. box contents: Refers to the data wrapped within the box structure and stored within the DBox field within the Box data structure box type: Specifies the kind of information that shall be stored with the box and stored within the TBox field within the Box data structure. cell: An optional subdivision of a tile used for low-memory encoding and decoding. cleanup pass: A coding pass performed on a single bit-plane of a code-block of coefficients. It is the first pass and only coding pass for the first significant bit-plane; the third and the last pass of all the remaining bit-planes. codestream: A collection of one or more bit streams and associated (overhead) information required for their decoding and expansion into image data. Such overhead information is restricted to that required for expansion and may include, but is not limited to, marker segments indicating locations of particular bit streams, indicating transform, quantization and coding types, etc. code-block: A rectangular grouping of coefficients from the same sub-band of a tile-component. x x
ISO/EC CD15444-1: 1999(V1.0,9 December 1999) code-block scan: The order in which the coefficients within a code-block are visited during a coding pass. The code-block is processed in stripes, each consisting of four rows and spanning the width of the code-block. Each stripe is processed column by column from left to right and from top to bottom coder: An embodiment of either an encoding or decoding process coding pass: A complete pass through a code-block where the appropriate coefficient values and context are applied. There are three types of coding passes: significance propagation pass, magnitude refinement pass and cleanup pass colour image: An image that has more than one component component: An image consists of one or more components. Each component is a two-dimensional array of samples. A colour image typically consists of several components such as RGB compressed data: Portion of the codestream, tile, or packet that contains the bit stream description. conforming reader: An application that reads and interprets a JP2 file correctly as defined by this context: Function of samples previously decoded used to condition the coding of the present sample. context label: The arbitrary index used to distinguish different context values. The labels are used as a convenience of notation rather than being normative context modelling: Procedure determining the probability distribution of difference with predicted bit from context vector: The binary vector consisting of the significance states of its context neighbor coefficients decoder: An embodiment of a decoding process, and optionally a colour transformation process. decoding process: A process which takes as its input compressed data and outputs a reconstructed image data. decomposition level: A collection of wavelet sub-bands where each coefficient has the same span with respect to the original samples. These include HL, LH, HH and, for the highest decomposition level, LL sub-bands. In this specification, only the ll sub-band can be further decomposed. discrete wavelet transform: DWT. a wavelet transformation performed on, and resulting in, spatiall encoder: An embodiment of an encoding process encoding process: A process, which takes as its input a source image and outputs compressed image data file format: This consists of a codestream and additional support data and information not explicitly required for the decoding of image data. Examples of such support data are text fields providing titling, security and historical information, markers to support placement of multiple codestreams within a given data file, and markers to support exchange between platforms or conversion to other file formats HH sub-band: The sub-band obtained by forward horizontal high-pass analysis filtering and vertical high pass analysis filtering. This sub-band contributes to reconstruction with inverse horizontal low-pass synthesis ltering and vertical low-pass synthesis filtering HL sub-band: The sub-band obtained by forward horizontal high-pass analysis filtering and vertical low-pass nalysis filtering. This sub-band contributes to reconstruction with inverse horizontal low-pass synthesis filtering and vertical high-pass synthesis filtering image: A set of two-dimensional arrays(components)of data image area: A rectangular part of the reference grid, registered by offsets from the origin and having the size of the image. The components are contained within this area and are related to the reference grid with respect image area offset: The width and height down and to the right of the reference origin where the origin of the mage area can be found ITU-T Rec T 800(1999 CDV10) 3
ISO/IEC CD15444-1 : 1999 (V1.0, 9 December 1999) ITU-T Rec. T.800 (1999 CDV1.0) 3 code-block scan: The order in which the coefficients within a code-block are visited during a coding pass. The code-block is processed in stripes, each consisting of four rows and spanning the width of the code-block. Each stripe is processed column by column from left to right and from top to bottom. coder: An embodiment of either an encoding or decoding process. coding pass: A complete pass through a code-block where the appropriate coefficient values and context are applied. There are three types of coding passes: significance propagation pass, magnitude refinement pass and cleanup pass. colour image: An image that has more than one component. component: An image consists of one or more components. Each component is a two-dimensional array of samples. A colour image typically consists of several components such as RGB. compressed data: Portion of the codestream, tile, or packet that contains the bit stream description. conforming reader: An application that reads and interprets a JP2 file correctly as defined by this Recommendation | International Standard. context: Function of samples previously decoded used to condition the coding of the present sample. context label: The arbitrary index used to distinguish different context values. The labels are used as a convenience of notation rather than being normative. context modelling: Procedure determining the probability distribution of difference with predicted bit from the context. context vector: The binary vector consisting of the significance states of its context neighbor coefficients decoder: An embodiment of a decoding process, and optionally a colour transformation process. decoding process: A process which takes as its input compressed data and outputs a reconstructed image data. decomposition level: A collection of wavelet sub-bands where each coefficient has the same span with respect to the original samples. These include HL, LH, HH and, for the highest decomposition level, LL sub-bands. In this specification, only the LL sub-band can be further decomposed. discrete wavelet transform: DWT. A wavelet transformation performed on, and resulting in, spatially discrete coefficients. encoder: An embodiment of an encoding process. encoding process: A process, which takes as its input a source image and outputs compressed image data. file format: This consists of a codestream and additional support data and information not explicitly required for the decoding of image data. Examples of such support data are text fields providing titling, security and historical information, markers to support placement of multiple codestreams within a given data file, and markers to support exchange between platforms or conversion to other file formats HH sub-band: The sub-band obtained by forward horizontal high-pass analysis filtering and vertical highpass analysis filtering. This sub-band contributes to reconstruction with inverse horizontal low-pass synthesis filtering and vertical low-pass synthesis filtering. HL sub-band: The sub-band obtained by forward horizontal high-pass analysis filtering and vertical low-pass analysis filtering. This sub-band contributes to reconstruction with inverse horizontal low-pass synthesis filtering and vertical high-pass synthesis filtering. image: A set of two-dimensional arrays (components) of data. image area: A rectangular part of the reference grid, registered by offsets from the origin and having the size of the image. The components are contained within this area and are related to the reference grid with respect to this area. image area offset: The width and height down and to the right of the reference origin where the origin of the image area can be found
ISO/ECCD15444-1:1999(v1.0,9 December1999 image data: Either source image data or reconstructed image data irreversible: A transformation, progression, system, or quantization that, due to systemic or quantization error, JPEG 2000: Used to refer globally to the encoding and decoding processes in this recommendation International Standard and their embodiment in applications LH sub-band: The sub-band obtained by forward horizontal low-pass analysis filtering and vertical high-pass nalysis filtering. This sub-band contributes to reconstruction with inverse horizontal high-pass synthesis Itering and vertical low-pass synthesis filtering. LL sub-band: The sub-band obtained by forward horizontal low-pass analysis filtering and vertical low-pa analysis filtering. This sub-band contributes to reconstruction with inverse horizontal high-pass synthesis iltering and vertical high-pass synthesis filtering layer: A collection of coding passes from one, or more, code-blocks of a tile-component. Layers have an order for encoding and decoding that must be preserved lossless: A descriptive term for the encoding and decoding processes in which the output of the decoding process is identical to the input to the encoding process. Lossless processes require reversible systems lossless coding: The mode of operation that refers to any one of the coding processes defined in this Recommendation International Standard in which all of the procedures are lossless lossy: A descriptive term for encoding and decoding processes which are not lossless. This includes both systems that are irreversible and those that include quantization magnitude refinement pass: A coding pass performed on a single bit-plane of a code-block of coefficients main header: A group of markers and marker segments at the beginning of the codestream that describe the image parameters and universal coding parameters marker: A two-byte code in which the first byte is hexadecimal FF(OxFF) and the second byte is a value between 1(0xo1)and hexadecimal FE(OxFE) marker segment: A marker and associated set of parameters mod: modb, x)=z, where z is an integer such that 0s:<x, and such that y-z is a multiple of x. packet: A part of the bit stream comprising a packet header and the coded data from one layer of one decomposition level of one component of a tile packet header: Portion of the packet that describes the layer, decomposition level, component, and the code block segment lengths progressive: The order a codestream where the decoding of each successive bit contributes to a"better reconstruction of the image. What metrics make the reconstruction"better" is a function of the application Some examples of progression are increasing resolution or improved pixel fidelity quantization: A method of reducing the precision of the individual coefficients to reduce the number of bits reconstructed image(data): An image, which is the output of a decode reconstructed sample (value): The sample value reconstructed by the decoder. This always equals the original sample value in lossless coding but may differ from the original sample value in lossy coding reference grid: A regular rectangular array of points to which all other planes (image, component, tile, sub- reference tile: A rectangular sub-grid of any size associated with the reference grid. egion of interest: A part of the image, component, or tile-component that is considered more important than the rest ITU-T Rec. T 800(1999 CDV10)
ISO/IEC CD15444-1 : 1999 (V1.0, 9 December 1999) 4ITU-T Rec. T.800 (1999 CDV1.0) image data: Either source image data or reconstructed image data. irreversible: A transformation, progression, system, or quantization that, due to systemic or quantization error, disallows lossless recovery signal. JPEG 2000: Used to refer globally to the encoding and decoding processes in this Recommendation | International Standard and their embodiment in applications. LH sub-band: The sub-band obtained by forward horizontal low-pass analysis filtering and vertical high-pass analysis filtering. This sub-band contributes to reconstruction with inverse horizontal high-pass synthesis filtering and vertical low-pass synthesis filtering. LL sub-band: The sub-band obtained by forward horizontal low-pass analysis filtering and vertical low-pass analysis filtering. This sub-band contributes to reconstruction with inverse horizontal high-pass synthesis filtering and vertical high-pass synthesis filtering. layer: A collection of coding passes from one, or more, code-blocks of a tile-component. Layers have an order for encoding and decoding that must be preserved. lossless: A descriptive term for the encoding and decoding processes in which the output of the decoding process is identical to the input to the encoding process. Lossless processes require reversible systems. lossless coding: The mode of operation that refers to any one of the coding processes defined in this Recommendation | International Standard in which all of the procedures are lossless. lossy: A descriptive term for encoding and decoding processes which are not lossless. This includes both systems that are irreversible and those that include quantization. magnitude refinement pass: A coding pass performed on a single bit-plane of a code-block of coefficients. main header: A group of markers and marker segments at the beginning of the codestream that describe the image parameters and universal coding parameters. marker: A two-byte code in which the first byte is hexadecimal FF (0xFF) and the second byte is a value between 1 (0x01) and hexadecimal FE (0xFE). marker segment: A marker and associated set of parameters. mod: mod(y,x) = z, where z is an integer such that , and such that y-z is a multiple of x. packet: A part of the bit stream comprising a packet header and the coded data from one layer of one decomposition level of one component of a tile. packet header: Portion of the packet that describes the layer, decomposition level, component, and the codeblock segment lengths. precision: Number of bits allocated to a particular sample. progressive: The order a codestream where the decoding of each successive bit contributes to a “better” reconstruction of the image. What metrics make the reconstruction “better” is a function of the application. Some examples of progression are increasing resolution or improved pixel fidelity. quantization: A method of reducing the precision of the individual coefficients to reduce the number of bits used to entropy code them. reconstructed image (data): An image, which is the output of a decoder. reconstructed sample (value): The sample value reconstructed by the decoder. This always equals the original sample value in lossless coding but may differ from the original sample value in lossy coding. reference grid: A regular rectangular array of points to which all other planes (image, component, tile, subband, etc.) are associated. reference tile: A rectangular sub-grid of any size associated with the reference grid. region of interest: A part of the image, component, or tile-component that is considered more important than the rest. 0 ≤ z x <
ISO/EC CD15444-1: 1999(V1.0,9 December 1999) resolution: The spatial mapping of samples to a physical space. In this Recommendation Specification the decomposition levels of the wavelet transform relate to each other with different relative resolutions reversible: A transformation, progression, system, or quantization that does not suffer systemic or uantization error and, therefore, allows lossless signal recovery. sample: One element in the two-dimensional array, which comprises a component selective arithmetic coding bypass: A coding style where some of the code-block passes are not cod shift: Multiplication or division of a binary number by factors of two sign-magnitude notation: A binary representation of an integer number where the distance from the origin is expressed with a positive number and the direction from the origin(positive or negative)is expressed with a significance propagation pass: A coding pass performed on a single bit-plane of a code-block of coefficients. significance state: Binary state variable that indicates whether the most significant 1 bit of a target coefficient in the current or more significant bit-planes or not source image(data): An image used as input to an encoder. sub-band: A group of transform coefficients resulting from the same sequence of low-pass and high-pass filtering operation, both vertically and horizontally b-band sub-band decomposition: A transformation of an image tile component into sub-bands sub-band decomposition level: The number of decompositions performed on the original tile-component sub-band recomposition: The inverse of sub-band decomposition sub-band recomposition level: The remaining number of recompositions needed to reconstruct the original superbox: A box that itself contains a contiguous sequence of boxes(and only a contiguous sequence of boxes). As the JP2 file contains only a contiguous sequence of boxes, the JP2 file is itself considered a superbox. When used as part of a relationship between two boxes, the term superbox refers to the box which directly contains the other box tile: An array of rectangular points on the reference grid, registered with and offset from the reference grid origin and defined by a base width and height. This tile array overlaps the image area and is used to define tile-component: A rectangular sub-component of any size which is part of a larger component. There is a tile- component for every component and every tile tile-part header: A group of markers and marker segments at the beginning of each tile-part codestream that describe the tile-part coding parameters tile number: The index of the current tile ranging from zero to the number of tiles minus one ile-part: A portion of the codestream that make up some, or all, of a tile. The tile-part includes at least one and up to all, of the packets that make up the tile tile-part number: The tile number of the tile with which the tile-part is associated. transform: A mathematical mapping from one signal space to another transform coefficient. a value that is the result of a transformation wavelet transform: A transformation which iteratively transforms one signal into two or more filtered and decimated signals corresponding to different frequency bands XOR: Exclusive OR logical operator. ITU-T Rec. T 800(1999 CDV10)
ISO/IEC CD15444-1 : 1999 (V1.0, 9 December 1999) ITU-T Rec. T.800 (1999 CDV1.0) 5 resolution: The spatial mapping of samples to a physical space. In this Recommendation | Specification the decomposition levels of the wavelet transform relate to each other with different relative resolutions. reversible: A transformation, progression, system, or quantization that does not suffer systemic or quantization error and, therefore, allows lossless signal recovery. sample: One element in the two-dimensional array, which comprises a component. selective arithmetic coding bypass: A coding style where some of the code-block passes are not coded by the arithmetic coder. shift: Multiplication or division of a binary number by factors of two. sign-magnitude notation: A binary representation of an integer number where the distance from the origin is expressed with a positive number and the direction from the origin (positive or negative) is expressed with a separate single bit. significance propagation pass: A coding pass performed on a single bit-plane of a code-block of coefficients. significance state: Binary state variable that indicates whether the most significant 1 bit of a target coefficient in the current or more significant bit-planes or not. source image (data): An image used as input to an encoder. sub-band: A group of transform coefficients resulting from the same sequence of low-pass and high-pass filtering operation, both vertically and horizontally. sub-band coefficient: A transform coefficient within a given sub-band. sub-band decomposition: A transformation of an image tile component into sub-bands. sub-band decomposition level: The number of decompositions performed on the original tile-component samples. sub-band recomposition: The inverse of sub-band decomposition. sub-band recomposition level: The remaining number of recompositions needed to reconstruct the original image tile component samples. superbox: A box that itself contains a contiguous sequence of boxes (and only a contiguous sequence of boxes). As the JP2 file contains only a contiguous sequence of boxes, the JP2 file is itself considered a superbox. When used as part of a relationship between two boxes, the term superbox refers to the box which directly contains the other box. tile: An array of rectangular points on the reference grid, registered with and offset from the reference grid origin and defined by a base width and height. This tile array overlaps the image area and is used to define image tiles. tile-component: A rectangular sub-component of any size which is part of a larger component. There is a tilecomponent for every component and every tile. tile-part header: A group of markers and marker segments at the beginning of each tile-part codestream that describe the tile-part coding parameters. tile number: The index of the current tile ranging from zero to the number of tiles minus one. tile-part: A portion of the codestream that make up some, or all, of a tile. The tile-part includes at least one, and up to all, of the packets that make up the tile. tile-part number: The tile number of the tile with which the tile-part is associated. transform: A mathematical mapping from one signal space to another. transform coefficient: A value that is the result of a transformation. wavelet transform: A transformation which iteratively transforms one signal into two or more filtered and decimated signals corresponding to different frequency bands. XOR: Exclusive OR logical operator
ISO/ECCD15444-1:1999(1.0,9 December1999 Abbreviations For the purposes of this Recommendation International Standard, the following abbreviations apply CCITT: International Telegraph and Telephone Consultative Committee, now ITU-T ICC: International Colour Consortium IEC: International Electrotechnical Commission ISO: International Organization for Standardization ITTF: Information Technology Task Force ITU: International Telecommunication union ITU-T: International Telecommunication Union-Telecommunication Standardization Sector(formerly the CCITT JPEG: Joint Photographic Experts Group- The joint Iso/TU committee responsible for developing standards for continuous-tone still picture coding. It also refers to the standards produced by this committee: ITU T T81 ISO/EC 10918-1, ITU-T T83 ISO/IEC 10918-2, ITU-T T84 ISO/EC 10918-3 and T87 ISO/EC 14495 JURA: JPEG Utilities Registration Authority ID-DWT: One dimensional discrete wavelet transform DWT: Discrete wavelet transform FDWT: Forward discrete wavelet transform IDWT: Inverse of the forward discrete wavelet transform LSB: Least significant bit. MSB: Most significant bit PCS: Profile Connection Space SNR: Signal to noise ratio. In this context it usually refers to reconstructed sample fidelity Symbols For the purposes of this Recommendation International Standard, the following symbols apply Ox- Denotes a hexadecimal numb CME: Comment and extension marker COC: Coding style component marker COD: Coding style default maker EOl: End of image marker IEM: Packet, main marker IET. Packet tile marker QCC: Quantization component marker QCD: Quantization default marker RIM: Region of interest marker SIZ: Size of image marker SOC: Start of image(codestream) marker ITU-T Rec. T 800(1999 CDV10)
ISO/IEC CD15444-1 : 1999 (V1.0, 9 December 1999) 6 ITU-T Rec. T.800 (1999 CDV1.0) 4Abbreviations For the purposes of this Recommendation | International Standard, the following abbreviations apply. CCITT: International Telegraph and Telephone Consultative Committee, now ITU-T ICC: International Colour Consortium IEC: International Electrotechnical Commission ISO: International Organization for Standardization ITTF: Information Technology Task Force ITU: International Telecommunication Union ITU-T: International Telecommunication Union – Telecommunication Standardization Sector (formerly the CCITT) JPEG: Joint Photographic Experts Group - The joint ISO/ITU committee responsible for developing standards for continuous-tone still picture coding. It also refers to the standards produced by this committee: ITUT T.81 | ISO/IEC 10918-1, ITU-T T.83 | ISO/IEC 10918-2, ITU-T T.84 | ISO/IEC 10918-3 and T.87 | ISO/IEC 14495. JURA: JPEG Utilities Registration Authority 1D-DWT: One dimensional discrete wavelet transform DWT: Discrete wavelet transform FDWT: Forward discrete wavelet transform IDWT: Inverse of the forward discrete wavelet transform LSB: Least significant bit. MSB: Most significant bit. PCS: Profile Connection Space SNR: Signal to noise ratio. In this context it usually refers to reconstructed sample fidelity. 5 Symbols For the purposes of this Recommendation | International Standard, the following symbols apply. 0x-: Denotes a hexadecimal number. CME: Comment and extension marker COC: Coding style component marker COD: Coding style default maker EOI: End of image marker IEM: Packet, main marker IET: Packet, tile marker QCC: Quantization component marker QCD: Quantization default marker RIM: Region of interest marker SIZ: Size of image marker SOC: Start of image (codestream) marker