A new pattern representation scheme using data compression

Author

Watanabe, Toshinori ; Sugawara, Ken ; Sugihara, Hiroshi

Author_Institution

Graduate Sch. of Inf. Syst., Univ. of Electro-Commun., Tokyo, Japan

Volume

24

Issue

5

fYear

2002

fDate

5/1/2002 12:00:00 AM

Firstpage

579

Lastpage

590

Abstract

We propose the PRDC (Pattern Representation based on Data Compression) scheme for media data analysis. PRDC is composed of two parts: an encoder that translates input data into text and a set of text compressors to generate a compression-ratio vector (CV). The CV is used as a feature of the input data. By preparing a set of media-specific encoders, PRDC becomes widely applicable. Analysis tasks - both categorization (class formation) and recognition (classification) - can be realized using CVs. After a mathematical discussion on the realizability of PRDC, the wide applicability of this scheme is demonstrated through the automatic categorization and/or recognition of music, voices, genomes, handwritten sketches and color images

Keywords

data analysis; data compression; data structures; encoding; multimedia computing; pattern classification; text analysis; vectors; PRDC; automatic categorization; class formation; color image recognition; compression ratio vector; data compression; feature space; generality; genome recognition; handwritten sketch recognition; input data translation; media data analysis; media-specific encoders; multimedia; music recognition; pattern classification; pattern recognition; pattern representation scheme; realizability; text compressors; text encoder; vector quantization; voice recognition; Data compression;

fLanguage

English

Journal_Title

Pattern Analysis and Machine Intelligence, IEEE Transactions on

Publisher

ieee

ISSN

0162-8828

Type

jour

DOI

10.1109/34.1000234

Filename

1000234