23 February 2010 Tissue classification by wavelet modified generic Fourier descriptor and their recognition using hybrid correlator
Author Affiliations +
Abstract
Segmentation in Magnetic resonance imaging (MRI) images is a widely studied problem, and techniques (supervised and unsupervised) are discussed in the literature. The basic approaches to image segmentation are based upon: (a) boundary representation, (b) regional characteristics and (c) a combination of boundary and region-based features. In this paper, we report classification of brain tissue based objects employing one of combination of boundary and region-based features as wavelet modified generic Fourier descriptor (WGFD) technique. This technique have been applied to a database consisting of 3 different class's tissues, each class consist of 50 shapes. The Euclidean distance has been calculated as a similarity measure parameter for tissue shape classification. The classification results have been carried out and it is inferred that WGFD performs for brain tissue classification. For brain tissue recognition, a simulation experiment employing hybrid correlator architecture has been carried out. We have used Wavelet modified maximum average correlation hight (MACH) filter for hybrid correlator. Mexican-hat wavelet has used to synthesize the wavelet MACH filter for simulation experiment.
© (2010) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Raj Bahadur Yadav, Arun K. Gupta, "Tissue classification by wavelet modified generic Fourier descriptor and their recognition using hybrid correlator", Proc. SPIE 7564, Photons Plus Ultrasound: Imaging and Sensing 2010, 75642R (23 February 2010); doi: 10.1117/12.841570; https://doi.org/10.1117/12.841570
PROCEEDINGS
7 PAGES


SHARE
Back to Top