A Combined Model of Bayesian Network and Spatial Markov Kernel for Multiclass Image Segmentation and Categorization
Journal Title: International Journal of Modern Engineering Research (IJMER) - Year 2013, Vol 3, Issue 1
Abstract
Abstract: We propose a new Bayesian Network model that segments the region of interest from the image and categorize the segmented image. A Bayesian Network is constructed from an over segmentation to model the statistical dependencies among edge segments, vertices, corners and blobs. The proposed interactive image segmentation involves active user intervention for interactive image segmentation while existing interactive segmentation approaches are often passively depend on the user to provide exact intervention. Image categorization is done based on the Spatial Markov Kernel (SMK)that categorizes the segmented image
Authors and Affiliations
Anurekha G
Keywords
Related Articles
- EP ID EP151345
- DOI -
- Views 126
- Downloads 0
On rg-Separation Axioms
In this paper we define almost rg-normality and mild rg-normality, continue the study of further properties of rgnormality. We show that these three axioms are regular open hereditary. Also define the class of alm...
A Study of diesel engine fuelled with Madhuca Indica biodiesel and its blend with Diesel fuel
The engine emission characteristics of Mahua (Madhuca Indica) biodiesel (Mahua Oil Methyl Ester) and its blends with diesel is presented. The thermo-physical properties of all the fuel blends have been measured and prese...
Investigation of the Role of Bulkhead and Crack Stopper Strap in the Fail-Safe Design of a Wide Bodied Transport Aircraft
One of the fail-safe design features is the two-bay crack arrest capability of the airframe. In particular two-bay longitudinal and two-bay circumferential crack arrest feature is the main aspect of design for...
Palmprint Identification Based on Principle Line Using Machine Learning Techniques
In this paper, we propose principle line based Palmprint Identification method. In this method to detect principle lines of palm print is with consecutive filtering operations. Smoothing operation is used to remove i...
performance Analysis of DV-Hop Localization Using Voronoi Approach
Localization that is to be aware of position of the node in the network is an essential issue in wireless sensor network (WSN). Localization algorithm in WSN can be divided into rangebased and range-free algorithm....