Efficient algorithms for lossless compression of 2D/3D images

N/ACitations
Citations of this article
5Readers
Mendeley users who have this article in their library.
Get full text

Abstract

We propose enhancements to the 2D lossless image compression method embodied in CALIC. The enhanced version of CALIC obtained better compression than obtained by CALIC on 37 of the 45 images in our benchmark suite. The two methods tied on 6 of the remaining 8 images. This benchmark suite includes medical, natural, and man-made images. We also propose a lossless compression method for 3D images. Our method employs motion estimation and obtained better compression than competing wavelet-based lossless compression methods on all 8 3D medical images in our benchmark suite.

Cite

CITATION STYLE

APA

Chen, F., Sahni, S., & Vemuri, B. C. (1999). Efficient algorithms for lossless compression of 2D/3D images. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1614, pp. 681–688). Springer Verlag. https://doi.org/10.1007/3-540-48762-x_84

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free