A simple unsupervised MRF model based image segmentation approach

A simple technique has been suggested to obtain optimal segmentation based on tonal and textural characteristics of an image using the Markov random field (MRF) model. The technique takes an initially over segmented image as well as the original image as its inputs and defines an MRF over the region adjacency graph (RAG) of the initially segmented regions. A tonal-region based segmentation technique due to Kartikeyan and Sarkar (1989) has been used for initial segmentation. The energy function has been defined over the first order cliques of the MRF. The essence of this approach is primarily based on quantitative values of the second order statistics, on region characteristics and consequently deciding upon the action of merging neighboring regions using the F-statistic. The effectiveness of our approach is demonstrated with wide variety of real life examples viz., indoor, outdoor and satellite and a comparison of its output with that of a previous work in the literature has been provided.     

Review and comparative evaluation of symbolic dynamic filtering for detection of anomaly patterns

Symbolic dynamic filtering (SDF) has been recently reported in literature as a pattern recognition tool for early detection of anomalies (i.e., deviations from the nominal behavior) in complex dynamical systems. This paper presents a review of SDF and its performance evaluation relative to other classes of pattern recognition tools, such as Bayesian Filters and Artificial Neural Networks, from the perspectives of: (i) anomaly detection capability, (ii) decision making for failure mitigation and (iii) computational efficiency. The evaluation is based on analysis of time series data generated from a nonlinear active electronic system. This work has been supported in part by the U.S. Army Research Laboratory and the U.S. Army Research Office under Grant No. W911NF-07-1-0376, by the U.S. Office of Naval Research under Grant No. N00014-08-1-380, and by NASA under Cooperative Agreement No. NNX07AK49A.     

Landcover classification in MRF context using Dempster-Shafer fusion for multisensor imagery.

This work deals with multisensor data fusion to obtain landcover classification. The role of feature-level fusion using the Dempster-Shafer rule and that of data-level fusion in the MRF context is studied in this paper to obtain an optimally segmented image. Subsequently, segments are validated and classification accuracy for the test data is evaluated. Two examples of data fusion of optical images and a synthetic aperture radar image are presented, each set having been acquired on different dates. Classification accuracies of the technique proposed are compared with those of some recent techniques in literature for the same image data.     

Partial replacement of metakaolin with red ceramic waste in geopolymer

Metakaolin was incrementally replaced (33.3%, 50% and 66.6%) by red ceramic waste in geopolymer formulation to study the effect on geopolymerisation and its resultant properties. The geopolymer binders composed of two calcined aluminosilicates (viz. Metakaolin and Red ceramic waste), NaOH and sodium silicate. In the experimental compositions, metakaolin was replaced gradually up to 66.6% in the clay fraction, the Si/Al increased from 3.36 to 5.16 and Na/Al increased from 0.93 to 1.38. The FTIR spectroscopic studies of geopolymer pastes along with XRD analysis indicated that the red ceramic waste partly reacts with alkali and takes part in geopolymer formation. Replacement of 33.3% metakaolin by the red ceramic waste in geopolymer binder did not reduce the compressive strength with respect to the pure metakaolin geopolymer here. Additional replacement resulted in a drastic decrease in the compressive strength of the geopolymer binder. However, the compressive strength of geopolymer mortars revealed interesting synergy between the amount of binder and particle packing in the mortar. Despite having a lower amount of binder phase, mortars with 33% and 50% red ceramic waste exhibited maximum compressive strength values. This has been attributed to improved particle packing through incorporation of red ceramic waste particles.     

A view of the thermal wave behaviors in thin plates

The objective of this research is to study the temperature variation in thin dielectric materials. The thermal wave model is being used as the classical Fourier law of heat conduction breaks down when a dielectric material of sub-micron geometry is heated rapidly. The first part of the work reviews the temperature distribution due to thermal wave in semi-infinite bodies. The main emphasis of this paper is to accurately determine the temperature profile in a finite plate made up from a dielectric material. The boundary conditions of the first and the second kind are selected for this study. Using the classical Laplace transform technique, analytical solutions are obtained for finite bodies with different boundary conditions the first kind and the second kind. Due to complexity of the problem, a symbolic algebra provided the solution for examining the thermal behavior of dielectric materials during rapid heating.     

Text and non-text separation in offline document images: a survey

International Journal on Document Analysis and Recognition (IJDAR) - Separation of text and non-text is an essential processing step for any document analysis system. Therefore, it is important to...     

First-principles study of CaCu3B4O12 (B=Co, Rh, Ir)

Employing first-principles density functional theory based calculations we investigated the change in electronic structure of CaCu₃B₄O₁₂ compounds as one moves from 3d (Co) to 4d (Rh) to 5d (Ir) element at B site. Our study sheds light on valences of Cu and B ions as one moves from 3d to 4d to 5d based compounds. The valence of Cu in Co and Rh compound turn out to be that of less known 3+ state, while that in Ir compound turn out to be commonly known 2+ state. Our first-principles study provide microscopic understanding of these different valences of Cu, in terms of changes in the mixing of Cu x ² − y ² and B-a _1g states, driven by changes in the crystal field and spin splitting. The stronger crystal field splitting for 4d and 5d elements compared to 3d at B site drive the low-spin state at Rh and Ir site as opposed to intermediate spin in case of Co.