Data dan Teknologi Multimedia Sesi 08 Nofriyadi Nurdam

 Introduction

 Data compression involves encoding information using fewer bits than the original representation would use.  Compression is useful to reduce the consumption of hard disk space or transmission bandwidth.  On the downside, compressed data must be decompressed to be used, and it may be detrimental to some applications.

 For instance, a compression of video may require expensive hardware for the video to be decompressed fast enough to be viewed as it is being decompressed  The option of decompressing the video in full before watching it may be inconvenient, and requires storage space for the decompressed video

 The data compression schemes involves the degree of compression, the amount of distortion, and the computational resources required to compress and uncompress the data  Compression was one of the main drivers for the growth of information during the past two decades.  There are two compression concept, lossy and lossless compression

 Lossless data compression is a class of data compression algorithms that allows the exact original data to be reconstructed from the compressed data.  The term lossless is in contrast to lossy data compression, which only allows an approximation of the original data to be reconstructed, in exchange for better compression rates.

 Lossless data compression is used in many applications, such as ZIP and gzip  It is also used as a component within lossy data compression technologies

 Lossless compression is used if it is important that the original and the decompressed data be identical  Some image file formats, like PNG or GIF, use only lossless compression, while others like TIFF and MNG may use either lossless or lossy methods  Lossless audio formats are most often used for archiving or production purposes

 Most lossless compression programs do two things in sequence, first generate a statistical model for the input data, and second use this model to map input data to bit sequences in such a way that frequently encountered data will produce shorter output than "improbable" data.  The algorithms used to produce bit sequences are Huffman coding and arithmetic coding

 Arithmetic coding achieves compression rates close to the best possible for a particular statistical mode  Huffman compression is simpler and faster but produces poor results for models that deal with symbol probabilities close to 1

 There are two primary ways of constructing statistical models: static model and adaptive model  In a static model, the data is analyzed and a model is constructed, then this model is stored with the compressed data

 This approach is simple and modular, but has the disadvantage that the model itself can be expensive to store, and also that it forces a single model to be used for all data being compressed, and so performs poorly on files containing heterogeneous data

 Adaptive models dynamically update the model as the data is compressed. Both the encoder and decoder begin with a trivial model, yielding poor compression of initial data, but as they learn more about the data, performance improves.  Most popular types of compression used in practice now use adaptive coders.

 Lossy compression is a data encoding method that compresses data by discarding (losing) some of it  The procedure aims to minimize the amount of data that need to be held, handled, and/or transmitted by a computer.  Typically, a substantial amount of data can be discarded before the result is sufficiently degraded to be noticed by the user.

 Lossy compression is most commonly used to compress multimedia data (audio, video, and still images), especially in applications such as streaming media and internet telephony  By contrast, lossless compression is required for text and data files, such as bank records and text articles

 Run-length encoding (RLE) is a very simple form of data compression in which runs of data (that is, sequences in which the same data value occurs in many consecutive data elements) are stored as a single data value and count, rather than as the original run  This is most useful on data that contains many such runs, example simple graphic images such as icons, line drawings, and animations  It is not useful with files that don't have many runs as it could greatly increase the file size.

 For example, a screen with black text on a solid white background. There are black pixel for text and white pixel  B for black pixel and W for white WWWWWWWWWWWWBWWWWWWWWWWWWBBBWWWW WWWWWWWWWWWWWWWWWWWWBWWWWWWWWWW WWWW  The RLE converts to 12W1B12W3B24W1B14W  The run-length code represents the original 67 characters in only 18.

 Run-length encoding is lossless data compression and is well suited to palette- based iconic images  It does not work well at all on continuous- tone images such as photographs

 Original text: ADA ATE APPLE  There are 7 symbols, A, D, E, L, P, T and space with frequency: 4 As, 2 Ps, 2 Es, 2 spaces, 1 D, 1 T and 1 L  The symbols are presented by 3 bits: A:000 D:001 E:010 L:011 P:100 T:101 Space:110  Encoded text needs 39 bits (compared to original text 104 bits)

 Original text: ADA ATE APPLE  There are 7 symbols, A, D, E, L, P, T and space with frequency: 4 As, 2 Ps, 2 Es, 2 spaces, 1 D, 1 T and 1 L  The symbols are presented depending on frequency A:0 P:10 E:110 Space:1110 D:11110 T: L:111111

 The Preffix Property  Encoded text needs bits (39 bits)  In general variable length encoding is better the fix length encoding  Deencoding is done with tree structure

 Variable length coding  Tree structure is built bottom up  Level paling bawah terdiri dari simbol dengan kemunculan paling sedikit