Multiple-valued and spectral approach to lossless compression of binary, gray scale and color biomedical images

• Published in: Proceedings - International Symposium on Multiple-Valued Logic pp. 136 - 142
• Main Authors: Falkowski, B.J., Olejnicka, B.T.
• Format: Conference Proceeding Journal Article
• Language: English
• Published: Los Alamitos CA IEEE 2002
• Subjects: Algebra; Applied sciences; Artificial intelligence; Biomedical engineering; Biomedical imaging; Computer science; control theory; systems; Digital images; Exact sciences and technology; Image coding; Image color analysis; Image segmentation; Image storage; Logical, boolean and switching functions; Material storage; Pattern recognition. Digital image processing. Computational geometry; Surgery; Theoretical computing; Spectral function; Multivalued function; Grey level image; Image databank; Binary image; Lossless compression; Medical imagery; Image compression; Color image; Biomedical engineering
• DOI: 10.1109/ISMVL.2002.1011081
• EISSN: 2378-2226
• ISBN: 9780769514628; 0769514626
• ISSN: 0195-623X

In this paper, we propose a new method based on multiple-valued functions and spectral analysis for the lossless compression of binary, gray scale and color biomedical images. Preprocessing of the raw image data is first applied to make the data more amenable for compression. The preprocessed image is then segmented into blocks to which various coding and compression is applied. Detailed description of segmentation and compression schemes is presented. The proposed scheme has been implemented in the C language and compared with other state-of-the-art techniques. Experimental results indicate significant improvements in compression ratios.