Fast Graph Partitioning Active Contours for Image Segmentation Using Histograms. (26th January 2010)

• Record Type: Journal Article
• Authors: Nath, Sumit K.; Palaniappan, Kannappan
• Other Names: Nikolaidis Nikos Academic Editor.
• Is Part Of: EURASIP journal on image and video processing. Volume 2009(2009)
• Publication Date: 2010-01-26
• Subjects: Image processing -- Digital techniques -- Periodicals; Digital video -- Periodicals; Traitement d'images; Vidéo numérique; Digital video; Image processing -- Digital techniques; Periodicals; Electronic journal; Electronic journals; 621.367
• DOI: 10.1155/2009/820986
• Languages: English
• ISSNs: 1687-5176

Abstract

Abstract : We present a method to improve the accuracy and speed, as well as significantly reduce the memory requirements, for the recently proposed Graph Partitioning Active Contours (GPACs) algorithm for image segmentation in the work of Sumengen and Manjunath (2006). Instead of computing an approximate but still expensive dissimilarity matrix of quadratic size, ( N s 2 M s 2 ) / ( n s m s ), for a 2D image of sizeN s × M s and regular image tiles of sizen s × m s, we use fixed length histograms and an intensity-based symmetric-centrosymmetric extensor matrix to jointly compute terms associated with the completeN s M s × N s M s dissimilarity matrix. This computationally efficient reformulation of GPAC using a very small memory footprint offers two distinct advantages over the original implementation. It speeds up convergence of the evolving active contour and seamlessly extends performance of GPAC to multidimensional images.