Compression of medical images by using artificial neural networks

Zümray Dokur*

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Citations (Scopus)

Abstract

This paper presents a novel lossy compression scheme for medical images by using an incremental self-organized map (ISOM). Three neural networks for lossy compression scheme are comparatively examined: Kohonen map, multi-layer perceptron (MLP) and ISOM. In the compression process of the proposed method, the image is first decomposed into blocks of 8×8 pixels. Two-dimensional discrete cosine transform (2D-DCT) coefficients are computed for each block. The dimension of DCT coefficients vectors (codewords) is reduced by low-pass filtering. Huffman coding is applied to the indexes of codewords obtained by the ISOM. In the decompression process, inverse operations of each stage of the compression are performed in the opposite way. It is observed that the proposed method gives much better compression rates.

Original languageEnglish
Title of host publicationInternational Conference on Intelligent Computing, ICIC 2006, Proceedings
PublisherSpringer Verlag
Pages337-344
Number of pages8
ISBN (Print)3540372717, 9783540372714
DOIs
Publication statusPublished - 2006
EventInternational Conference on Intelligent Computing, ICIC 2006 - Kunming, China
Duration: 16 Aug 200619 Aug 2006

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume4113 LNCS - I
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

ConferenceInternational Conference on Intelligent Computing, ICIC 2006
Country/TerritoryChina
CityKunming
Period16/08/0619/08/06

Fingerprint

Dive into the research topics of 'Compression of medical images by using artificial neural networks'. Together they form a unique fingerprint.

Cite this