Bioinformatics-Inspired Algorithms for 2D-Image Analysis-Application to Synthetic and Medical Images Part I: Images in Rectangular Format

Bioinformatics-Inspired Algorithms for 2D-Image Analysis-Application to Synthetic and Medical Images Part I: Images in Rectangular Format

Perambur S. Neelakanta (Department of Computer and Electrical Engineering & Computer Science, Florida Atlantic University, Boca Raton, FL, USA) and Deepti Pappusetty (Department of Computer and Electrical Engineering & Computer Science, Florida Atlantic University, Boca Raton, FL, USA)
Copyright: © 2012 |Pages: 25
DOI: 10.4018/ijbce.2012010102

Abstract

To ascertain specific features in bio-/medical-images, a new avenue of using the so-called Needleman-Wunsch (NW) and Smith-Waterman (SW) algorithms (of bioinformatics) is indicated. In general, NW/SW algorithms are adopted in genomic science to obtain optimal (global and local) alignment of two linear sequences (like DNA nucleotide bases) to determine the similarity features between them and such 1D-sequence algorithms are presently extended to compare 2D-images via binary correlation. The efficacy of the proposed method is tested with synthetic images and a brain scan image. Thus, the way of finding the location of a distinct part in a synthetic image and that of a tumour in the brain scan image is demonstrated.
Article Preview

Pattern Classifiers And Binary Correlators

In the pursuits of pattern recognition, images can be specified as spatially discretized set of binary intensities (Shalkoff, 1989, 1992); hence, binary correlators are used as classifiers of patterns (expressed in digital formats) to compare and correlate the underlying image features. A binary correlator in essence, is a classifier that determines which of a set of classes is most representative of an unknown pattern. In image comparison contexts, it translates to finding the extent of correlation between two image-frames; and, if the images are denoted via binary variables 1 and 0, the corresponding classification conforms to a binary correlator.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 8: 2 Issues (2019): Forthcoming, Available for Pre-Order
Volume 7: 2 Issues (2018)
Volume 6: 2 Issues (2017)
Volume 5: 2 Issues (2016)
Volume 4: 2 Issues (2015)
Volume 3: 2 Issues (2014)
Volume 2: 2 Issues (2013)
Volume 1: 2 Issues (2012)
View Complete Journal Contents Listing