SPSA Based Feature Relevance Estimation For Video Retrieval

With the availability of a huge amount of video data on various sources, efficient video retrieval tools are increasingly in demand. Video being a multi-modal data, the perceptions of ``relevance'' between the user provided query video (in case of Query-By-Example type of video search) and retrieved video clips are subjective in nature. We present an efficient video retrieval method that takes user's feedback on the relevance of retrieved videos and iteratively reformulates the input query feature vectors (QFV) for improved video retrieval. The QFV reformulation is done by a simple, but powerful feature weight optimization method based on Simultaneous Perturbation Stochastic Approximation (SPSA) technique. A video retrieval system with video indexing, searching and relevance feedback (RF) phases is built for demonstrating the performance of the proposed method. The query and database videos are indexed using the conventional video features like color, texture, etc. However, we use the comprehensive and novel methods of feature representations, and a spatio-temporal distance measure to retrieve the top M videos that are similar to the query. In feedback phase, the user activated iterative on the previously retrieved videos is used to reformulate the QFV weights (measure of importance) that reflect the user's preference, automatically. It is our observation that a few iterations of such feedback are generally sufficient for retrieving the desired video clips. The novel application of SPSA based RF for user-oriented feature weights optimization makes the proposed method to be distinct from the existing ones. The experimental results show that the proposed RF based video retrieval exhibit good performance.

A Jacobian-Based Algorithm for Planning Attitude Maneuvers Using Forward and Reverse Rotations

Algorithms for planning quasistatic attitude maneuvers based on the Jacobian of the forward kinematic mapping of fully-reversed (FR) sequences of rotations are proposed in this paper. An FR sequence of rotations is a series of finite rotations that consists of initial rotations about the axes of a body-fixed coordinate frame and subsequent rotations that undo these initial rotations. Unlike the Jacobian of conventional systems such as a robot manipulator, the Jacobian of the system manipulated through FR rotations is a null matrix at the identity, which leads to a total breakdown of the traditional Jacobian formulation. Therefore, the Jacobian algorithm is reformulated and implemented so as to synthesize an FR sequence for a desired rotational displacement. The Jacobian-based algorithm presented in this paper identifies particular six-rotation FR sequences that synthesize desired orientations. We developed the single-step and the multiple-step Jacobian methods to accomplish a given task using six-rotation FR sequences. The single-step Jacobian method identifies a specific FR sequence for a given desired orientation and the multiple-step Jacobian algorithm synthesizes physically feasible FR rotations on an optimal path. A comparison with existing algorithms verifies the fast convergence ability of the Jacobian-based algorithm. Unlike closed-form solutions to the inverse kinematics problem, the Jacobian-based algorithm determines the most efficient FR sequence that yields a desired rotational displacement through a simple and inexpensive numerical calculation. The procedure presented here is useful for those motion planning problems wherein the Jacobian is singular or null.

A non-linear analogue-to-digital converter for linearization of transducer input-output relationships

Analogue and digital techniques for linearization of non-linear input-output relationship of transducers are briefly reviewed. The condition required for linearizing a non-linear function y = f(x) using a non-linear analogue-to-digital converter, is explained. A simple technique to construct a non-linear digital-to-analogue converter, based on ' segments of equal digital interval ' is described. The technique was used to build an N-DAC which can be employed in a successive approximation or counter-ramp type ADC to linearize the non-linear transfer function of a thermistor-resistor combination. The possibility of achieving an order of magnitude higher accuracy in the measurement of temperature is shown.

Classification and rating of strong-motion earthquake records

Ninety-two strong-motion earthquake records from the California region, U.S.A., have been statistically studied using principal component analysis in terms of twelve important standardized strong-motion characteristics. The first two principal components account for about 57 per cent of the total variance. Based on these two components the earthquake records are classified into nine groups in a two-dimensional principal component plane. Also a unidimensional engineering rating scale is proposed. The procedure can be used as an objective approach for classifying and rating future earthquakes.

A cryptographic system based on finite field transforms

In this paper, we develop a cipher system based on finite field transforms. In this system, blocks of the input character-string are enciphered using congruence or modular transformations with respect to either primes or irreducible polynomials over a finite field. The polynomial system is shown to be clearly superior to the prime system for conventional cryptographic work.

A conditional random field-based downscaling method for assessment of climate change impact on multisite daily precipitation in the Mahanadi basin

Downscaling to station-scale hydrologic variables from large-scale atmospheric variables simulated by general circulation models (GCMs) is usually necessary to assess the hydrologic impact of climate change. This work presents CRF-downscaling, a new probabilistic downscaling method that represents the daily precipitation sequence as a conditional random field (CRF). The conditional distribution of the precipitation sequence at a site, given the daily atmospheric (large-scale) variable sequence, is modeled as a linear chain CRF. CRFs do not make assumptions on independence of observations, which gives them flexibility in using high-dimensional feature vectors. Maximum likelihood parameter estimation for the model is performed using limited memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS) optimization. Maximum a posteriori estimation is used to determine the most likely precipitation sequence for a given set of atmospheric input variables using the Viterbi algorithm. Direct classification of dry/wet days as well as precipitation amount is achieved within a single modeling framework. The model is used to project the future cumulative distribution function of precipitation. Uncertainty in precipitation prediction is addressed through a modified Viterbi algorithm that predicts the n most likely sequences. The model is applied for downscaling monsoon (June-September) daily precipitation at eight sites in the Mahanadi basin in Orissa, India, using the MIROC3.2 medium-resolution GCM. The predicted distributions at all sites show an increase in the number of wet days, and also an increase in wet day precipitation amounts. A comparison of current and future predicted probability density functions for daily precipitation shows a change in shape of the density function with decreasing probability of lower precipitation and increasing probability of higher precipitation.

Design of multimedia surveillance systems

This article addresses the problem of how to select the optimal combination of sensors and how to determine their optimal placement in a surveillance region in order to meet the given performance requirements at a minimal cost for a multimedia surveillance system. We propose to solve this problem by obtaining a performance vector, with its elements representing the performances of subtasks, for a given input combination of sensors and their placement. Then we show that the optimal sensor selection problem can be converted into the form of Integer Linear Programming problem (ILP) by using a linear model for computing the optimal performance vector corresponding to a sensor combination. Optimal performance vector corresponding to a sensor combination refers to the performance vector corresponding to the optimal placement of a sensor combination. To demonstrate the utility of our technique, we design and build a surveillance system consisting of PTZ (Pan-Tilt-Zoom) cameras and active motion sensors for capturing faces. Finally, we show experimentally that optimal placement of sensors based on the design maximizes the system performance.

Block-based feature detection and matching for mosaicing of camera-captured document images

In this paper, we present a new feature-based approach for mosaicing of camera-captured document images. A novel block-based scheme is employed to ensure that corners can be reliably detected over a wide range of images. 2-D discrete cosine transform is computed for image blocks defined around each of the detected corners and a small subset of the coefficients is used as a feature vector A 2-pass feature matching is performed to establish point correspondences from which the homography relating the input images could be computed. The algorithm is tested on a number of complex document images casually taken from a hand-held camera yielding convincing results.

OCR Prediction Using Support Vector Machine Based on Piezocone Data

The determination of the overconsolidation ratio (OCR) of clay deposits is an important task in geotechnical engineering practice. This paper examines the potential of a support vector machine (SVM) for predicting the OCR of clays from piezocone penetration test data. SVM is a statistical learning theory based on a structural risk minimization principle that minimizes both error and weight terms. The five input variables used for the SVM model for prediction of OCR are the corrected cone resistance (qt), vertical total stress (sigmav), hydrostatic pore pressure (u0), pore pressure at the cone tip (u1), and the pore pressure just above the cone base (u2). Sensitivity analysis has been performed to investigate the relative importance of each of the input parameters. From the sensitivity analysis, it is clear that qt=primary in situ data influenced by OCR followed by sigmav, u0, u2, and u1. Comparison between SVM and some of the traditional interpretation methods is also presented. The results of this study have shown that the SVM approach has the potential to be a practical tool for determination of OCR.

A model based factorization approach for dense 3D recovery from monocular video

Feature track matrix factorization based methods have been attractive solutions to the Structure-front-motion (Sfnl) problem. Group motion of the feature points is analyzed to get the 3D information. It is well known that the factorization formulations give rise to rank deficient system of equations. Even when enough constraints exist, the extracted models are sparse due the unavailability of pixel level tracks. Pixel level tracking of 3D surfaces is a difficult problem, particularly when the surface has very little texture as in a human face. Only sparsely located feature points can be tracked and tracking error arc inevitable along rotating lose texture surfaces. However, the 3D models of an object class lie in a subspace of the set of all possible 3D models. We propose a novel solution to the Structure-from-motion problem which utilizes the high-resolution 3D obtained from range scanner to compute a basis for this desired subspace. Adding subspace constraints during factorization also facilitates removal of tracking noise which causes distortions outside the subspace. We demonstrate the effectiveness of our formulation by extracting dense 3D structure of a human face and comparing it with a well known Structure-front-motion algorithm due to Brand.

Supervisory Control for Real-Time Systems Based on Conflict-Tolerant Controllers

This paper addresses the problem of detecting and resolving conflicts due to timing constraints imposed by features in real-time and hybrid systems. We consider systems composed of a base system with multiple features or controllers, each of which independently advise the system on how to react to input events so as to conform to their individual specifications. We propose a methodology for developing such systems in a modular manner based on the notion of conflict-tolerant features that are designed to continue offering advice even when their advice has been overridden in the past. We give a simple priority-based scheme forcomposing such features. This guarantees the maximal use of each feature. We provide a formal framework for specifying such features, and a compositional technique for verifying systems developed in this framework.

Handling sectional views in volume-based approach to automatically construct 3D solid from 2D views

This paper presents an algorithm for solid model reconstruction from 2D sectional views based on volume-based approach. None of the existing work in automatic reconstruction from 2D orthographic views have addressed sectional views in detail. It is believed that the volume-based approach is better suited to handle different types of sectional views. The volume-based approach constructs the 3D solid by a boolean combination of elementary solids. The elementary solids are formed by sweep operation on loops identified in the input views. The only adjustment to be made for the presence of sectional views is in the identification of loops that would form the elemental solids. In the algorithm, the conventions of engineering drawing for sectional views, are used to identify the loops correctly. The algorithm is simple and intuitive in nature. Results have been obtained for full sections, offset sections and half sections. Future work will address other types of sectional views such as removed and revolved sections and broken-out sections. (C) 2004 Elsevier Ltd. All rights reserved.

Site characterization model using least-square support vector machine and relevance vector machine based on corrected SPT data (N-c)

Statistical learning algorithms provide a viable framework for geotechnical engineering modeling. This paper describes two statistical learning algorithms applied for site characterization modeling based on standard penetration test (SPT) data. More than 2700 field SPT values (N) have been collected from 766 boreholes spread over an area of 220 sqkm area in Bangalore. To get N corrected value (N,), N values have been corrected (Ne) for different parameters such as overburden stress, size of borehole, type of sampler, length of connecting rod, etc. In three-dimensional site characterization model, the function N-c=N-c (X, Y, Z), where X, Y and Z are the coordinates of a point corresponding to N, value, is to be approximated in which N, value at any half-space point in Bangalore can be determined. The first algorithm uses least-square support vector machine (LSSVM), which is related to aridge regression type of support vector machine. The second algorithm uses relevance vector machine (RVM), which combines the strengths of kernel-based methods and Bayesian theory to establish the relationships between a set of input vectors and a desired output. The paper also presents the comparative study between the developed LSSVM and RVM model for site characterization. Copyright (C) 2009 John Wiley & Sons,Ltd.

Main chain and segmental dynamics of semi interpenetrating based on polyisoprene and poly(methyl methacrylate)

Main chain and segmental dynamics of polyisoprene (PI) and poly(methyl methacrylate)(PMMA) chains in semi IPNs were systematically studied over a wide range of temperatures (above and below T-g of both polymers) as a function of composition, crosslink density, and molecular weight. The immiscible polymers retained most of its characteristic molecular motion; however, the semi IPN synthesis resulted in dramatic changes in the motional behavior of both polymers due to the molecular level interpenetration between two polymer chains. ESR spin probe method was found to be sensitive to the concentration changes of PMMA in semi IPNs. Low temperature spectra showed the characteristics of rigid limit spectra, and in the range of 293-373 K.complex spectra were obtained with the slow component mostly arisingout of the PMMA rich regions and fast component from the PI phase. We found that the rigid PMMA chains closely interpenetrated into thehighly mobile PI network imparts motional restriction in nearby PI chains, and the highly mobile PI chains induce some degree of flexibility in highly rigid PMMA chains. Molecular level interchain mixing was found to be more efficient at a PMMA concentration of 35 wt.%. Moreover, the strong interphase formed in the above mentionedsemi IPN contributed to the large slow component in the ESR spectra at higher temperature. The shape of the spectra along with the data obtained from the simulations of spectra was correlated to the morphology of the semi IPNs. The correlation time measurement detected the motional region associated with the glass transition of PI and PMMA, and these regions were found to follow the same pattern of shifts in a-relaxation of PI and PMMA observed in DMA analysis. Activation energies associated with the T-g regions were also calculated. T-50G was found to correlate with the T-g of PMMA, and the volume of polymer segments undergoing glass transitional motion was calculated to be 1.7 nm(3).C-13 T-1 rho measurements of PMMA carbons indicate that the molecular level interactions were strong in semi IPN irrespective of the immiscible nature of polymers. The motional characteristics of H atoms attached to carbon atoms in both polymers were analyzed using 2D WISE NMR. Main relaxations of both components shifted inward, and both SEM and TEM analysis showed the development of a nanometer sized morphology in the case of highly crosslinked semi IPN. (C) 2010 Elsevier Ltd. All rights reserved.

Can biological motion be a biometric?

Biological motion has successfully been used for analysis of a person's mood and other psychological traits. Efforts are made to use human gait as a non-invasive mode of biometric. In this reported work, we try to study the effectiveness of biological gait motion of people as a cue to biometric based person recognition. The data is 3D in nature and, hence, has more information with itself than the cues obtained from video-based gait patterns. The high accuracies of person recognition using a simple linear model of data representation and simple neighborhood based classfiers, suggest that it is the nature of the data which is more important than the recognition scheme employed.