Rajbabu Velmurugan

Acoustic Multitarget Tracking Using Direction-of-Arrival Batches

In this paper, we propose a particle filter acoustic direction-of-arrival (DOA) tracker to track multiple maneuvering targets using a state space approach. The particle filter determines its state vector using a batch of DOA estimates. The filter likelihood treats the observations as an image, using template models derived from the state update equation, and also incorporates the possibility of missing data as well as spurious DOA observations. Multiple targets are handled using a partitioned state-vector approach. The particle filter solution is compared with three other methods: the extended Kalman filter, Laplacian filter, and another particle filter that uses the acoustic microphone outputs directly. In addition, we demonstrate an autonomous system for multiple target DOA tracking with automatic target initialization and deletion. The initialization system uses a track-before-detect approach and employs matching pursuit to initialize multiple targets. Computer simulations are presented to compare the performance of the algorithms

A Range-Only Multiple Target Particle Filter Tracker

We propose a particle filter tracker to track multiple maneuvering targets using a batch of range measurements. The state update is formulated through a locally linear motion model and the observability of the state vector is proved using geometrical arguments. The data likelihood treats the range observations as an image using template models derived from the state update equation, and incorporates the possibility of missing data as well as spurious range observations. The particle filter handles multiple targets, using a partitioned state-vector approach. The filter proposal function uses a Gaussian approximation of the full-posterior to cope with target maneuvers for improved efficiency. By treating the range measurements as images and using smoothness constraints, the particle filter is able to avoid the data association problems. Computer simulations demonstrate the performance of the tracking algorithm

CNT: concept-map based navigation and discovery in a repository of learning content

In this paper, we present a tool that automatically connects keywords in student generated concept maps to relevant learning components in our digital repository. Currently, there are over 6,000 heterogeneous components available in our systems, with more than 1,000 dedicated to the teaching of ECE 2025, an introductory course in Signal Processing. These components consist of captured lectures, support material, multimedia examples, worked problems and others. The CNT (concept navigation tool) connects concept map nodes to course content based on keywords embedded in the concept nodes. CNT goes beyond just integrating search techniques with a map-authoring tool. The concept maps constructed by students become the navigation tool that allows them to explore the relevant content and improve or expand their concept maps as their understanding grows. This environment was designed to increase the depth of a students conceptual understanding of course material. The paper includes details about the design and implementation of the CNT system and the supporting data systems.

Estimation of lip opening for scaling of vocal tract area function for speech training aids

For visual feedback of articulatory efforts in speech training aids, the vocal tract shape can be estimated by LPC analysis of the speech signal. The vocal tract is modelled as a concatenation of equal length sections and the ratios of the areas at section interfaces are calculated and these are scaled using the area of a reference section. The lip opening area as estimated from a video recording of the speakers face can be used as a reference area for obtaining the vocal tract shape during speech utterances with transitional tract configuration. A technique for estimating the area of the lip opening based on template matching is investigated. It satisfactorily tracked the horizontal and vertical opening of the lips in the video images of speakers with different skin hues, recorded under good lighting conditions.

Effective data association scheme for tracking closely moving targets using factor graphs

Effectiveness of tracking closely moving targets depends on the capability to resolve the ambiguity in associating measurements-to-tracks. Joint probabilistic data association (JPDA) has been shown to be very effective in tracking closely moving objects, but the approach is susceptible to track coalescence. The factor graph (FG) based association scheme developed in this paper circumvents the track coalescence by avoiding multiple hypothesis equivalence with recursive updation of likelihood values. The improvement in association using factor graph based data association scheme over JPDA has been demonstrated using a simulated scenario of closely moving targets. The steady state likelihood values obtained at the end of recursive process are shown to match the likelihoods obtained from measurements.

A Multi Target Bearing Tracking System using Random Sampling Consensus

In this paper, we present an acoustic direction-of-arrival (DOA) tracking system to track multiple maneuvering targets using a state space approach. The system consists of three blocks: beamformer, random sampling, and particle filter. The beamformer block processes the received acoustic data to output bearing batches as point statistics. The random sampling block determines temporal clustering of the bearings in a batch to determine region-of-interests (ROIs). Based on the track-before-detect approach, each ROI indicates the presence of a possible target. We describe three random sampling algorithms called RANSAC, MSAC, and NAPSAC to use in the random sampling block. The particle filter then tracks the targets via its interactions with the beamformer and the random sampling blocks. We present a computational analysis of the random sampling blocks and show tracking results with field data.

Joint MAP estimation for blind deconvolution: when does it work?

Blind deconvolution aims at reconstructing an image from its blurred and noisy version, when the blur kernel is not known. It has been acknowledged that the naive maximum aposteriori probability (MAP) algorithm favors a no-blur solution [3]. In [8] the failure of the direct MAP approach is addressed and it is proved that a simultaneous MAP estimation of the image and the point spread function (PSF) fails, providing a trivial solution. In contrast, we show that an appropriate choice of PSF prior during joint MAP estimation does provide a non-trivial solution. We provide the feasible range for the PSF regularization factor which would prevent a trivial solution.

Extensions to Orthogonal Matching Pursuit for Compressed Sensing

Compressed Sensing (CS) provides a set of mathematical results showing that sparse signals can be exactly reconstructed from a relatively small number of random linear measurements. A particularly appealing greedy-approach to signal reconstruction from CS measurements is the so called Orthogonal Matching Pursuit (OMP). We propose two modifications to the basic OMP algorithm, which can be handy in different situations.

Mixed-mode Implementation of Particle Filters

In this paper, we develop new mixed-mode implementations for particle Biters and compare them to a digital implementation. The motivation for the mixed-mode implementation is to achieve low-power implementation of particle filters. The specific application considered is a bearings-only, single-target tracking algorithm. Specifically, we develop mixed-mode implementations that use analog components to realize nonlinear functions in the particle filter algorithm. The analog implementation of nonlinear functions uses low-power multiple-input translinear element (MITE) networks. Simulation results for one mixed-mode implementation of the bearings-only tracker show that the analog errors are low enough to support accurate tracking. Redesign of the mixed-mode implementation in a second form with more analog components will result in nearly twenty times less power dissipation.

Mean LBP and modified fuzzy C-means weighted hybrid feature for illumination invariant mean-shift tracking

Object tracking is a critical task in surveillance and activity analysis. Two main issues for tracking are appearance (illumination) and structural (size of a target) variations of the object. We propose a method which is robust and addresses these issues by incorporating features that are less variant to these changes. The proposed features are mean local binary pattern (mLBP), an illumination invariant texture feature, and modified fuzzy c-means (MFCM) weighted color histogram to handle both illumination and scale changes. These features are combined to form a hybrid mean-shift (MS) vector and used in the MS vector framework for target tracking. Experimental results using standard benchmark videos show that the proposed scheme can lead to better localization and robust tracking in challenging illumination scenarios, when compared to several existing tracking algorithms.