Search

Video based vehicle detection is an active research area in Intelligent Transportation System. The aim is to efficiently extract the required data from large amount of videos captured by surveillance cameras by detecting, tracking identifying objects of interest and analyze their activities. This paper provides a review of literature on vehicle detection, recognition and tracking. Many approaches have been proposed for detection which includes appearance based models and motion based models .Vehicle recognition based on various attributes such as color, license plate, logos are discussed. Region based and feature based tracking along with various tracking algorithms are surveyed. Multi-view methods for vehicle detection are also discussed.

Graphology assists in interpreting a person's character and identity attributes by analysing the handwriting. Graphology is the investigation of the sensible state of mind at the time of writing. These thoughtfully give understanding into the morality, hidden talents and enthusiastic conditions of the writer or artist. The handwriting is exceptional. Composing is a physical procedure. The cerebrum state is reflected in the handwriting, as the signals moves to fingers, where together they control the handwriting. Graphology can be utilized to analyse the characteristic features of an individual. The various approaches used in personality analysis using graphology are discussed in this paper.

Dynamic PET brain studies are common in research and are becoming common in clinical applications. Simulation of dynamic PET is an important step to validate techniques and methods. This study compares the CPU and GPU version of a GATE simulation running on a highperformance computing system. We simulated the Siemens 3T MR-BrainPET with dynamic brain activity. The results show close agreement in the number of coincidence events, including phantom scatter, which we made available in the output of GPU simulation through code changes. The code will be made available to the GATE community and the simulation data is available to the interested researchers.

Silicon tip fabrication by wet anisotropic etching using Potassium Hydroxide (KOH) and Tetra Methyl Ammonium Hydroxide (TMAH) was studied under varying process conditions using thermal oxide as a mask. The fabrication was carried out using only wet etching with a single layer of optical lithography. The mask structures used for the formation of tips were circles of different dimeter ranging from 30 μm to 120 μm to fabricate the tips with diameter from 40nm to 400nm, without post oxidation sharpening method. Experiments were carried out under two different conditions: a) Beaker containing small volume with constant stirring and b) Bulk volume MEMS bath with and without recirculation and filtration. The results were analyzed to understand the nature of alkali solution and the effect of temperature and method of recirculation/filtration on the dimensions and nature of the tips formed, along with the behavior of the oxide mask in the two different solutions.

Deep vertical etching of Si using hydrogen peroxide as oxidant and Au as catalyst has been studied in terms of relative reactant concentrations and catalyst morphology. The concentrations of hydrogen peroxide and hydrofluoric acid are highly important and the ratio of the same can be varied to achieve isotropic to highly anisotropic vertical etching. Catalyst morphology and positioning of the substrate was also seen to affect vertical etch profiles, with high catalyst deposition rates and horizontal placement of the sample giving deep etching without defects.