Image CompressionStandards : Image CompressionStandards By
A.Raju
NIT Warangal
What’s Image Compression? : What’s Image Compression? Reduction of the amount of data removal of redundant data
Transforming a 2-D pixel array into
a statistically uncorrelated data set
Why Compression? : Why Compression? Important in data storage and data transmission
Examples:
– Progressive transmission of images/videos
– Video coding (HDTV, teleconferencing)
– Digital libraries and image databases
– Remote sensing
– Medical imaging
Why Standardization? : Why Standardization? Compression is one of the technologies that enable the multimedia revolution to occur.
However for technology to be effective there has to be some degree of standardization so that the equipment designed by different vendors can talk to each other.
These standards are accepted by ISO (International Standards Organization)and ITU(International Telecommunications Union) once called CCITT (Consultative committee of the International Telephone and Telegraph).
Image Compression Standards : Image Compression Standards
Bi-Level Still Images : Bi-Level Still Images CCITT Group3:Designed for transmitting binary documents over telephone. (FAX)
CCITT Group4:Simplified version of CCITT Group 3
JBIG or JBIG1:Joint Bi-level Image Experts Group, for lossless compression of bi level images
JBIG2:For internet and FAX applications, internet, desktop
TIFF:Tagged Image File Format. Flexible file format supporting variety of compression standards including JPEG, JPEG2000
Continuous Tone Still Images : Continuous Tone Still Images JPEG: Joint Photographic Experts Group. Popular for compression of images on Internet.
JPEG-2000: Follow on to JPEG for increased compression of photographic quality images
PDF: Portable Document Format. For representing 2-D documents
PNG : Portable Network Graphic. For Losslessly compression of full color images.
Video : Video DV:Digital Video. For semiprofessional video production, camcorders
H.261:2 way video conferencing standard. DCT compression mode.
H.262:Designed for DVDs with 15Mb/s
H.263:Enhanced version of H.261 for ordinary telephone modems.
H.264:For video conferencing, Internet streaming, and TV Broadcasting Uses variable block size integer transforms, adaptive arithmetic coding
MPEG-1:Motion Pictures Expert Group. For CD-ROM applications.
MPEG-2:Extension for MPEG-1, for DVDs with higher transfer rate
MPEG-4:Extension to MPEG-2, supports HDTV, variable block sizes
MPEG-4 AVC:Uses Advanced Video Coding (AVC). Identical to H.264
Compression Scheme: : Compression Scheme: In any compression scheme there are:
Step 1- Removal of redundancy based on implicit assumption about the structure in the data
Step 2- Assignment of binary code words to the information deemed non redundant.
Transformer
Applies a one-to-one transformation to the input image data. Output of the transformer is an image representation which is more efficient compression than the raw image data.
Slide 10: Quantizer
Generates a limited number of symbols that can be used in the representation of the compressed image.
Coder
Coder assigns a code word, a binary bit stream , to each symbol at the output of Quantization.
The coder may employ a Fixed-Length or Variable-
Length codes.VLC, also known as Entropy Coding, assigns a codeword in such a way as to minimize the average length of the binary representation of the symbols
JPEG standard : JPEG standard JPEG : Joint Photographic Experts Group Standard for continuous tone still images
Widely used standard
New Standard : JPEG 2000
The standard describes 29 distinct coding systems for compression of images.
Four Modes:
Sequential (Baseline) JPEG Progressive JPEG
Hierarchical JPEG Lossless JPEG
Major Steps in JPEG Image Compression
• Discrete Cosine Transform (DCT) / IDCT
• Quantization / Dequantization
• Differential Pulse Code Modulation (DPCM)
• Entropy Coding of DC Coefficients
• Run-Length Coding (RLC)
• Entropy Coding of AC Coefficients
Block diagram for JPEG encoder. : Block diagram for JPEG encoder.
Block diagram for JPEG Decoder. : Block diagram for JPEG Decoder.
Baseline JPEG : Baseline JPEG DC Coefficient :
First coefficient in every 8 x 8 block
Represents the average value of pixels in block
AC Coefficients : Remaining 63 coefficients in every 8 x 8 block
DC Coefficients: treated separately from the AC Coefficients
Differential values of DC coeffs. of all blocks are derived and encoded.
The entries of Q(u,v) tend to have larger values towards the lower right corner. This aims to introduce more loss at the higher spatial frequencies
Example: : Example: Original 8x8 image Shifting and DCT Quantization matrix Q(u,v) Quantized DCT coefficient matrix 15
JPEG coefficient coding categories : JPEG coefficient coding categories
Suggested Huffman code for DC differences : Suggested Huffman code for DC differences JPEG Standard, Table K3 – Luminance
JPEG suggested AC code for luminance : JPEG suggested AC code for luminance
JPEG suggested AC code for luminance(cont.) : JPEG suggested AC code for luminance(cont.)
Slide 20: Progressive: Spectral Selection Spectral selection: Takes advantage of the “spectral” (spatial frequency spectrum) characteristics of the DCT coefficients: higher AC components provide detail information.
Scan 1: Encode DC and first few AC components, e.g., AC1, AC2.
Scan 2: Encode a few more AC components, e.g., AC3, AC4, AC5.
...
Scan n: Encode the last few ACs, e.g., AC61, AC62, AC63.
Slide 21: Progressive: Successive Approximation Successive approximation: Instead of gradually encoding spectral bands, all DCT coefficients are encoded simultaneously but with their most significant bits (MSBs) first.
Scan 1: Encode the first few MSBs, e.g., Bits 7, 6, 5, 4.
Scan 2: Encode a few more less significant bits, e.g., Bit 3.
... Scan n: Encode the least significant bit (LSB), Bit 0.
Example:
150(10)=10010110(2)
144(10)=10010000(2)
144(10)=10010000(2)
148(10)=10010100(2)
150(10)=10010110(2)
150(10)=10010110(2)
Slide 22: Hierarchical JPEG The encoded image at the lowest resolution is basically a compressed lowpass filtered image, whereas the images at successively higher resolutions provide additional details (differences from the lower resolution images).
Provides Progressive Representations
Provides Multiple Resolutions
Requires more space
JPEG 2000 Family of Standard : JPEG 2000 Family of Standard JPEG(1992)
lossy and lossless, DCT based
JPEG-LS(1997)
lossy and near-lossless prediction-based
JPEG2002 part1(2002)
lossy and lossless, wavelet-based
JPEG2002 part2(2002)
Provision for 3D extension
JPEG (Started in 2002)
Multicomponent and Volumetric images
The JPEG 2000 Standard : The JPEG 2000 Standard JPEG 2000 PART 1
Released in 2002
High coding efficiency ( 20% more than JPEG)
Wavelet-based compression
Seamless lossy-to-lossless compression
Highly scalable ( in quality, resolution, component)
Easy alteration of the progression order
Region of interest coding
Random access to code stream portions
The JPEG 2000 Standard : The JPEG 2000 Standard JPEG 2000 part 2
Supports arbitrary wavelet filters
Supports multi component trasformations
Multi component support:
Linear block transform ( DCT,KLT,….)
3D wavelet transform
Predictive coding
Must specify inverse transform, then store 2D wavelet-compressed components
The JPEG-LS Standard : The JPEG-LS Standard Released in 1997
Based on non-linear prediction and context-based Golomb-Rice coding
Has low complexity
Lossless compression is better than JPEG 2000
Provides support for near-lossless compression
Does not provide scalability
Good choice for on-board compression
Various standards : Various standards Film 8mm, 16mm, 35mm, IMAX
TV NTSC, PAL, SECAM, HDTV
VideoTape VHS, S-VHS, Hi-8, Dig-8, DV
Compression MJPEG, MPEG (1,2,3,4)
File Formats AVI, MPG, VCD, DVD, MOV
Slide 28: THANK YOU