JPEG2000压缩标准详解

JPEG2000压缩标准

5星 · 超过95%的资源需积分: 10 9 浏览量更新于2024-07-30 收藏 2.27MB PDF 举报

"JPEG2000压缩标准是图像编码系统的一种先进标准，由ISO和ITU共同制定。JPEG 2000 Final Committee Draft Version 1.0发布于2000年3月16日，它标志着该标准的最终草案阶段。此标准的项目编号为1.29.15444，即JPEG2000（也称为ISO/IEC 15444-1）。文档的主要编辑者包括Martin Boliek、Charilaos Christopoulos和Eric Majani，他们隶属于ISO/IEC JTC1/SC29/WG1，即负责静止图像编码的国际组织。文档请求的行动是进行审查并根据需要采取适当行动，主要分发对象为WG1成员。JPEG 2000标准的联系人是Martin Boliek，而ISO/IEC JTC1/SC29/WG1的召集人是Daniel T. Lee。" JPEG2000压缩标准的核心知识点如下： 1. **分层编码**：JPEG2000采用了一种称为“分层编码”的技术，允许用户在不同分辨率下逐步显示图像，从而实现渐进传输。这使得在网络带宽有限的情况下，用户可以快速看到低质量的图像，随着数据的接收，图像质量逐渐提高。 2. **小波变换**：与传统的JPEG使用离散余弦变换（DCT）不同，JPEG2000使用小波变换来处理图像。小波变换提供多分辨率分析，能够更好地捕捉图像的局部特性，从而实现更高效的压缩。 3. **无损压缩**：JPEG2000支持无损压缩模式，这意味着在解压后，图像数据可以完全恢复到原始状态，而不会有任何质量损失，这对于需要保留原始数据的重要应用非常有用。 4. **区域编码**：该标准允许对图像的特定区域进行独立编码，这使得可以对图像的不同部分应用不同的压缩比，增强了灵活性和可控性。 5. **颜色空间处理**：JPEG2000支持多种颜色空间，包括YCbCr、RGB等，并且可以进行颜色空间转换，提高了压缩效率和颜色保真度。 6. **错误恢复和鲁棒性**：JPEG2000的编码结构允许在数据传输或存储过程中出现错误时进行恢复，因为每个部分都是独立编码的。这种特性提高了其在有噪声环境或不稳定的网络条件下的鲁棒性。 7. **码流和文件格式**：JPEG2000使用JP2（JPEG 2000 Part 2，ISO/IEC 15444-2）作为其核心文件格式，同时还有JPC（JPEG 2000 Codestream）码流格式，适用于流式传输和嵌入式应用。 8. **扩展性**：JPEG2000标准设计时考虑了扩展性，可以应用于其他领域，如医学影像、遥感图像和3D图像的压缩。 9. **压缩效率**：JPEG2000通常提供比传统JPEG更高的压缩效率，尤其是在处理具有大量细节或高动态范围的图像时。 10. **元数据支持**：JPEG2000支持在文件中嵌入元数据，如图像的创作信息、版权信息等，方便管理和使用。 JPEG2000标准通过引入小波变换、分层编码、区域编码和无损压缩等技术，极大地提升了图像压缩的性能和灵活性，使其成为数字图像处理领域的里程碑式标准。

ISO/IEC FCD15444-1 : 2000 (V1.0, 16 March 2000)

ITU-T Rec. T.800 (2000 FCDV1.0) 1

INTERNATIONAL STANDARD

ITU-T RECOMMENDATION

INFORMATION TECHNOLOGY –

JPEG 2000 IMAGE CODING SYSTEM

1 Scope

This Recommendation | International Standard deﬁnes a set of lossless (bit-preserving) and lossy compression methods

for coding continuous-tone, bi-level, grey-scale, or colour digital still images.

This Recommendation | International Standard

— speciﬁes decoding processes for converting compressed image data to reconstructed image data

— speciﬁes a codestream syntax containing information for interpreting the compressed image data

— speciﬁes a ﬁle format

— provides guidance on encoding processes for converting source image data to compressed image data

— provides guidance on how to implement these processes in practice

2 References

The following Recommendations and International Standards contain provisions which, through reference in this text,

constitute provisions of this Recommendation | International Standard. At the time of publication, the editions indicated

were valid. All Recommendations and Standards are subject to revision, and parties to agreements based on this

Recommendation | International Standard are encouraged to investigate the possibility of applying the most recent

edition of the Recommendations and Standards listed below. Members of IEC and ISO maintain registers of currently

valid International Standards. The Telecommunication Standardization Bureau of the ITU maintains a list of currently

valid ITU-T Recommendations.

2.1 Identical Recommendations | International Standards

— ITU-T Recommendation T.81 | ISO/IEC 10918-1:1994, Information technology - Digital compression

and coding of continuous-tone still images: Requirements and guidelines.

— ITU-T Recommendation T.88 | ISO/IEC 14492-1, Lossy/lossless coding of bi-level images

2.2 Additional references

— Coded character set—7 bit, American Standard Code for Information Interchange, ANSI X3.4–1986.

— ISO/IEC 646:1991, ISO 7-bit coded character set for information interchange.

— ITU-T Recommendation T.83 | ISO/IEC 10918-2:1995, Information technology - Digital compression

and coding of continuous-tone still images: Compliance testing.

— ITU-T Recommendation T.84 | ISO/IEC 10918-3:1996, Information technology - Digital compression

and coding of continuous-tone still images: Extensions.

— ITU-T Recommendation T.84 | ISO/IEC 10918-3 Amd 1 (In preparation), Information technology -

Digital compression and coding of continuous-tone still images: Extensions - Amendment 1.

ISO/IEC FCD15444-1 : 2000 (V1.0, 16 March 2000)

2 ITU-T Rec. T.800 (2000 FCDV1.0)

— ITU-T Recommendation T.86 | ISO/IEC 10918-4, Information technology - Digital compression and

coding of continuous-tone still images: Registration of JPEG Proﬁles, SPIFF Proﬁles, SPIFF Tags,

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.82 | ISO/IEC 11544:1994, Information technology - Coded representation

of picture and audio information — Progressive bi-level image compression

— 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

3 Deﬁnitions

For the purposes of this Recommendation | International Standard, the following deﬁnitions apply.

, ﬂoor function: This indicates the largest integer not exceeding x.

, ceiling function: This indicates the smallest integer not exceeded by x.

arithmetic coder: An entropy coder that converts variable length strings to variable length codes.

auxiliary component: A component from the codestream that is used by the application outside the scope of

colourspace conversion. For example, an opacity component or a depth component would be an auxiliary

component.

box: A building block deﬁned by a unique box type and length. Some particular boxes may contain other

boxes.

box contents: Refers to the data wrapped within the box structure. The contents of a particular box are stored

within the DBox ﬁeld within the Box data structure as deﬁned in Annex I.6

box type: Speciﬁes the kind of information that shall be stored with the box. The type of a particular box is

stored within the TBox ﬁeld within the Box data structure as deﬁned in Annex I.6.

bit-plane: A two dimensional array of bits. In this Recommendation | International Standard a bit-plane refers

to all the bits of the same magnitude in all coefﬁcients or samples. This could refer to a bit-plane in a

component, tile-component, 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 markers or marker segments in the main and tile-part headers. It does include any packet headers and in

stream markers and marker segments not found in the main or tile-part headers.

big endian: The bits occur in order from most signiﬁcant to least signiﬁcant.

byte: Eight-bit octet.

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. The overhead information is restricted to that required for the

expansion into image data 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.

ISO/IEC FCD15444-1 : 2000 (V1.0, 16 March 2000)

ITU-T Rec. T.800 (2000 FCDV1.0) 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. The result of each pass (after arithmetic coding) is a stream of compressed data.

colour component: A component from the codestream that functions as an input to a colour transformation

system. For example, a red component or a greyscale component would be a colour component.

colour image: An image that has more than one component.

component: A two-dimensional array of samples. A colour image typically consists of several components

from a speciﬁed colour space, for instance representing red, green, and blue.

compressed data: Any data that is part of the bit stream except for packet headers and in stream markers and

marker segments.

conforming reader: An application that reads and interprets a JP2 ﬁle correctly as deﬁned by Annex J of this

Recommendation | International Standard.

container box: An 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

container box. When used as part of a relationship between two boxes, the term container box refers to the box

which directly contains the other box.

context: Function of coefﬁcients previously decoded and 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 from the context the probability distribution of the predicted bit.

context vector: The binary vector consisting of the signiﬁcance states of its context 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 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 lowest resolution decomposition level, LL

sub-bands. In this speciﬁcation, only the LL sub-band can be further decomposed.

delimiting markers and marker segments: Markers and marker segments that give information about

beginning and ending points of structures in the codestream.

discrete wavelet transform (DWT): A transformation that iteratively transforms one signal into two or more

ﬁltered and decimated signals corresponding to different frequency bands. This transform operates on spatially

discrete samples.

encoder: An embodiment of an encoding process.

encoding process: A process, that 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 include 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.

ﬁxed information markers and marker segments: Markers and marker segments that offer information

about the original image.

functional markers and marker segments: Markers and marker segments that offer information about the

decoding procedures to be used

ISO/IEC FCD15444-1 : 2000 (V1.0, 16 March 2000)

4 ITU-T Rec. T.800 (2000 FCDV1.0)

header: Part of the codestream that contains only markers and marker segments. There are two types of

headers. The main header is found at the beginning of the codestream and tile-part headers are found at the

beginning of each tile-part.

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 vertical high-pass synthesis

ﬁltering and horizontal high-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 vertical low-pass synthesis ﬁltering

and horizontal high-pass synthesis ﬁltering.

image: The set of all components.

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 grid origin where the origin of

the image area can be found.

image data: Either source image data or reconstructed image data.

in bit stream markers and marker segments: Markers and marker segments that provide error resilience

functionality.

informational markers and marker segments: Markers and marker segments that offer ancillary

information.

irreversible: A transformation, progression, system, or quantization that, due to systemic or quantization error,

disallows lossless recovery. An irreversible process can only lead to lossy compression.

JP2 ﬁle: The name of ﬁle in the ﬁle format described in this speciﬁcation. Structurally, a JP2 ﬁle is a

contiguous sequence of boxes.

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 vertical high-pass synthesis ﬁltering

and horizontal 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 vertical low-pass synthesis ﬁltering

and horizontal low-pass synthesis ﬁltering.

layer: A collection of coding pass compressed data 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. Distortion free restoration can be assured. 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 that are not lossless. Distortion free restoration

is not assured. 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 coding parameters that can apply to every tile and tile-component.

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).

ISO/IEC FCD15444-1 : 2000 (V1.0, 16 March 2000)

ITU-T Rec. T.800 (2000 FCDV1.0) 5

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 code-

block segment lengths.

packet partition: A division of one tile-component by a rectangular grid. One packet partition size is speciﬁed

for each resolution level.

packet partition location: One rectangular region of a packet partition.

pointer markers and marker segments: Markers and marker segments that offer information about the

location of structures in the codestream.

precinct: A sub-division of a tile-component, within a each resolution, used for limiting the size of packets.

precision: Number of bits allocated to a particular sample, coefﬁcient, or other binary numerical

representation.

progressive: The order of 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.

raster order: A particular sequential order of data of any type within an array. The raster order starts with the

top left data point and moves to the immediate right data point, and so on, to the end of the line. After the end

of the line is reached the next data point in the sequence is the left-most data point immediately below the

current line. This order is continued to the end of the array.

reconstructed image (data): An image, that 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 images, components, tiles, sub-bands, etc. are

associated. Reference grid units or points are used to describe the mapping of the tiles and the components.

reference tile: A rectangular sub-grid of any size associated with the reference grid.

region of interest (ROI): A deﬁned area of the image, component, or tile-component that is considered of

particular relevance by some user deﬁned measure.

resolution: The spatial mapping of samples to a physical space. In this Recommendation | International

Standard the decomposition levels of the wavelet transform relate to each other with relative resolutions

differing by powers of two.

reversible: A transformation, progression, system, or quantization that does not suffer systemic or

quantization error and, therefore, allows lossless signal recovery. The result of reversible process may be lossy

or lossless depending on the quantization and other factors in the system.

sample: One element in the two-dimensional array that comprises a component.

segmentation symbol: A special symbol coded with a uniform context at the end of each coding pass for error

resilience.

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.

0 zx<≤

剩余204页未读，继续阅读

xhp0235

粉丝: 4
资源: 15

JPEG2000压缩标准详解

JPEG2000图像压缩基础、标准和实践——部分2.pdf

JPEG2000源码的使用手册

Jpeg2000压缩标准实现kakadu

JPEG2000压缩标准详解：新一代图像处理技术

matlab_jpeg2000.zip_JPEG2000压缩_Jpeg2000Gui_jpeg 压缩_jpeg2000 in m

Jpeg2000.rar_JPEG 2000_JPEG2000_JPEG2000压缩_image compression_jpe

jpeg2000.zip_-baijiahao_JPEG2000_jpeg2000 matlab_jpeg2000压缩率_图像压

jpeg图片压缩标准

JPEG_JPEG2000 压缩解压源码

JPEG2000压缩算法研究

最新资源