Title: Fuzzy rule-based segmentation of CT brain images of hemorrhage for compression

Authors: V.K. Sudha; R. Sudhakar; Valentina E. Balas

Addresses: Department of Electronics and Communication Engineering, Dr. Mahalingam College of Engineering and Technology, Udumalai Road, Pollachi – 642003, Tamil Nadu, India ' Department of Electronics and Communication Engineering, Dr. Mahalingam College of Engineering and Technology, Udumalai Road, Pollachi – 642003, Tamil Nadu, India ' Department of Automatics and Applied Software, Aurel Vlaicu University of Arad, 77 B-dul Revolutiei, 310130 Arad, Romania

Abstract: This paper presents segmentation of hemorrhage from the computed tomography image. Fuzzy rule-based technique constituting three parameters namely mean difference, mode difference and grey level intensity is used. These parameters are fuzzified into four fuzzy regions with trapezoidal membership functions. The segmented arbitrary shaped region of interest information is employed in JPEG2000 encoder while encoding the image. It encodes ROI with highest priority in allocating bits as well as in transmission. 15 CT image slices with hemorrhage were used in the experiment. Experimental results shows good segmentation of hemorrhage region from CT images and allows encoding image at any desired bit rate, while maintaining lossless in ROI and lossy/lossless in non-ROI.

Keywords: computed tomography; brain hemorrhage; brain bleeding; fuzzy rule-based systems; JPEG 2000 encoder; fuzzy logic; image segmentation; brain images; image compression; mean difference; mode difference; grey level intensity; medical imaging; lossless compression.

DOI: 10.1504/IJAIP.2012.052069

International Journal of Advanced Intelligence Paradigms, 2012 Vol.4 No.3/4, pp.256 - 267

Published online: 23 Aug 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article