Human Behaviour Recognition based on Trajectory Analysis using Neural Networks

Automated human behaviour analysis has been, and still remains, a challenging problem. It has been dealt from different points of views: from primitive actions to human interaction recognition. This paper is focused on trajectory analysis which allows a simple high level understanding of complex human behaviour. It is proposed a novel representation method of trajectory data, called Activity Description Vector (ADV) based on the number of occurrences of a person is in a specific point of the scenario and the local movements that perform in it. The ADV is calculated for each cell of the scenario in which it is spatially sampled obtaining a cue for different clustering methods. The ADV representation has been tested as the input of several classic classifiers and compared to other approaches using CAVIAR dataset sequences obtaining great accuracy in the recognition of the behaviour of people in a Shopping Centre.

Power of the joint segregation analysis method for testing mixed major-gene and polygene inheritance models of quantitative traits

Understanding the genetic architecture of quantitative traits can greatly assist the design of strategies for their manipulation in plant-breeding programs. For a number of traits, genetic variation can be the result of segregation of a few major genes and many polygenes (minor genes). The joint segregation analysis (JSA) is a maximum-likelihood approach for fitting segregation models through the simultaneous use of phenotypic information from multiple generations. Our objective in this paper was to use computer simulation to quantify the power of the JSA method for testing the mixed-inheritance model for quantitative traits when it was applied to the six basic generations: both parents (P-1 and P-2), F-1, F-2, and both backcross generations (B-1 and B-2) derived from crossing the F-1 to each parent. A total of 1968 genetic model-experiment scenarios were considered in the simulation study to quantify the power of the method. Factors that interacted to influence the power of the JSA method to correctly detect genetic models were: (1) whether there were one or two major genes in combination with polygenes, (2) the heritability of the major genes and polygenes, (3) the level of dispersion of the major genes and polygenes between the two parents, and (4) the number of individuals examined in each generation (population size). The greatest levels of power were observed for the genetic models defined with simple inheritance; e.g., the power was greater than 90% for the one major gene model, regardless of the population size and major-gene heritability. Lower levels of power were observed for the genetic models with complex inheritance (major genes and polygenes), low heritability, small population sizes and a large dispersion of favourable genes among the two parents; e.g., the power was less than 5% for the two major-gene model with a heritability value of 0.3 and population sizes of 100 individuals. The JSA methodology was then applied to a previously studied sorghum data-set to investigate the genetic control of the putative drought resistance-trait osmotic adjustment in three crosses. The previous study concluded that there were two major genes segregating for osmotic adjustment in the three crosses. Application of the JSA method resulted in a change in the proposed genetic model. The presence of the two major genes was confirmed with the addition of an unspecified number of polygenes.

A JOINT EXPERIMENTAL ANALYSIS OF INVESTOR BEHAVIOR IN IPO PRICING METHODS

This article jointly examines the differences of laboratory versions of the Dutch clock open auction, a sealed-bid auction to represent book building, and a two-stage sealed bid auction to proxy for the “competitive IPO”, a recent innovation used in a few European equity initial public offerings. We investigate pricing, seller allocation, and buyer welfare allocation efficiency and conclude that the book building emulation seems to be as price efficient as the Dutch auction, even after investor learning, whereas the competitive IPO is not price efficient, regardless of learning. The competitive IPO is the most seller allocative efficient method because it maximizes offer proceeds. The Dutch auction emerges as the most buyer welfare allocative efficient method. Underwriters are probably seeking pricing efficiency rather than seller or buyer welfare allocative efficiency and their discretionary pricing and allocation must be important since book building is prominent worldwide.

From correspondence analysis to multiple and joint correspondence analysis

The generalization of simple (two-variable) correspondence analysis to more than two categorical variables, commonly referred to as multiple correspondence analysis, is neither obvious nor well-defined. We present two alternative ways of generalizing correspondence analysis, one based on the quantification of the variables and intercorrelation relationships, and the other based on the geometric ideas of simple correspondence analysis. We propose a version of multiple correspondence analysis, with adjusted principal inertias, as the method of choice for the geometric definition, since it contains simple correspondence analysis as an exact special case, which is not the situation of the standard generalizations. We also clarify the issue of supplementary point representation and the properties of joint correspondence analysis, a method that visualizes all two-way relationships between the variables. The methodology is illustrated using data on attitudes to science from the International Social Survey Program on Environment in 1993.

Multicamera trajectory analysis for semantic behaviour characterisation

In this paper we propose an innovative approach for behaviour recognition, from a multicamera environment, based on translating video activity into semantics. First, we fuse tracks from individual cameras through clustering employing soft computing techniques. Then, we introduce a higher-level module able to translate fused tracks into semantic information. With our proposed approach, we address the challenge set in PETS 2014 on recognising behaviours of interest around a parked vehicle, namely the abnormal behaviour of someone walking around the vehicle.

Joint vibrations analysis in asymptomatic volunteers and symptomatic patients

The joint sound is a common sign in TMD, the diagnosis is important to establish the treatment of pathological alterations which occur in the TMJ. In this study, two groups were selected: 1, Asymptomatic volunteers; and 2, Symptomatic patients who were diagnosed in a clinical examination. After the initial examination, they were submitted to evaluation using electrovibratography (SonoPAK II, BioResearch Assoc., Inc., Milwaukee, Wisconsin). The analysis of results indicated that the averages of the vibratory energy in the symptomatic group presented higher values in all stages of the mandibular movement when compared to the averages of vibratory energy registered in the asymptomatic group.

Joint Vibration Analysis in Patients with Articular Inflammation

The study of articular sounds using a computerized system (SonoPAK) in patients with temporomandibular disorders (TMD) of inflammatory origin revealed an increase of vibratory energy when compared to asymptomatic individuals. The following conclusions were reached: 1. The amount of vibratory energy registered in these patients ranged from 8.50 to 57.61 Hz. The major vibrations occurred in the middle of the mandibular opening cycle; 2. The mean vibratory energy measured at less than 300 Hz was between 5.70 and 48.64 Hz and at higher than 300 Hz was between 3.70 and 8.99 Hz; 3. The peak amplitude in the patients with inflammation ranged from 0.35 to 3.96 Pascal and the peak of frequency from 83.20 to 120.20 Hz.

Operational modal analisys using joint statistical analysis of multiple records

In Operational Modal Analysis (OMA) of a structure, the data acquisition process may be repeated many times. In these cases, the analyst has several similar records for the modal analysis of the structure that have been obtained at di�erent time instants (multiple records). The solution obtained varies from one record to another, sometimes considerably. The differences are due to several reasons: statistical errors of estimation, changes in the external forces (unmeasured forces) that modify the output spectra, appearance of spurious modes, etc. Combining the results of the di�erent individual analysis is not straightforward. To solve the problem, we propose to make the joint estimation of the parameters using all the records. This can be done in a very simple way using state space models and computing the estimates by maximum-likelihood. The method provides a single result for the modal parameters that combines optimally all the records.

Joint statistical analysis of multiple measurement setups for the estimation of vibrational modes

Computing the modal parameters of large structures in Operational Modal Analysis often requires to process data from multiple non simultaneously recorded setups of sensors. These setups share some sensors in common, the so-called reference sensors that are fixed for all the measurements, while the other sensors are moved from one setup to the next. One possibility is to process the setups separately what result in different modal parameter estimates for each setup. Then the reference sensors are used to merge or glue the different parts of the mode shapes to obtain global modes, while the natural frequencies and damping ratios are usually averaged. In this paper we present a state space model that can be used to process all setups at once so the global mode shapes are obtained automatically and subsequently only a value for the natural frequency and damping ratio of each mode is computed. We also present how this model can be estimated using maximum likelihood and the Expectation Maximization algorithm. We apply this technique to real data measured at a footbridge.

Abnormal Pedestrian Trajectory analysis based on arbitrary-length clustering

This papers examines the use of trajectory distance measures and clustering techniques to define normal
and abnormal trajectories in the context of pedestrian tracking in public spaces. In order to detect abnormal
trajectories, what is meant by a normal trajectory in a given scene is firstly defined. Then every trajectory
that deviates from this normality is classified as abnormal. By combining Dynamic Time Warping and a
modified K-Means algorithms for arbitrary-length data series, we have developed an algorithm for trajectory
clustering and abnormality detection. The final system performs with an overall accuracy of 83% and 75%
when tested in two different standard datasets.

Developmental trajectories of marijuana use and psychological distress : exploring the co-occurrence of these phenomena in early adolescence

Le début de l’adolescence est une période de changements rapides où la détresse psychologique et l’expérimentation de la marijuana sont choses fréquentes. Certaines études longitudinales ont démontré que ces deux phénomènes ont tendance à se manifester conjointement tandis que d’autres n’ont pu observer de tel lien. Ces résultats divergents suggèrent que plusieurs questions persistent concernant la nature de cette relation. Cette thèse a pour objectif d’explorer la consommation de marijuana et la détresse psychologique en début d’adolescence afin de mieux saisir les changements à travers le temps, ainsi que d’examiner si ces deux problématiques évoluent conjointement et s’influencent réciproquement. Un échantillon de 448 adolescents garçons et filles fréquentant deux écoles secondaires de Montréal, ont été suivi de secondaire I à secondaire III. De 1999 à 2001, les participants ont complété un questionnaire à chaque année de l’étude incluant des mesures portant sur la consommation de marijuana et la détresse psychologique (IDPESQ-14). Un modèle de mixture semi-paramétrique (Nagin, 2005) a été utilisé afin d’identifier les trajectoires développementales de la consommation de marijuana et de détresse psychologique. Des analyses ont également été effectuées afin d’établir les liens d’appartenance entre chacune des trajectoires de consommation identifiées et la détresse psychologique lors de la première année de l`étude, ainsi qu’entre chacune des trajectoires de détresse psychologique et la consommation de marijuana en première année du secondaire. Finalement, des analyses de trajectoires jointes ont été effectuées afin de déterminer l’interrelation entre la consommation de marijuana et la détresse psychologique. Les résultats de notre étude suggèrent qu’il existe une grande hétérogénéité au niveau de la consommation de marijuana et la détresse psychologique. Trois trajectoires développementales ont été identifiées pour la consommation de marijuana: consommation légère, consommation grandissante et consommation élevée et stable. Trois trajectoires ont également été observées pour la détresse psychologique : basse, moyenne et élevée. Nos résultats démontrent la présence d’un lien entre la détresse psychologique rapportée lors de la première année de l’étude et les trajectoires de consommation problématiques. Ce lien a également été observé entre la consommation de marijuana rapportée lors de première année de l’étude et les trajectoires problématiques de détresse psychologique. Les analyses de trajectoires jointes démontrent la présence d’une concordance entre la consommation de marijuana et la détresse psychologique. Cette interrelation est toutefois complexe puisque les trajectoires de détresse psychologique élevée sont associées à un niveau de consommation de marijuana plus problématique mais l’inverse de cette association est moins probable. Notre étude met en lumière la nature asymétrique de la concordance entre la consommation de marijuana et la détresse psychologique.

Measures of fit in multiple correspondence analysis of crisp and fuzzy coded data

When continuous data are coded to categorical variables, two types of coding are possible: crisp coding in the form of indicator, or dummy, variables with values either 0 or 1; or fuzzy coding where each observation is transformed to a set of "degrees of membership" between 0 and 1, using co-called membership functions. It is well known that the correspondence analysis of crisp coded data, namely multiple correspondence analysis, yields principal inertias (eigenvalues) that considerably underestimate the quality of the solution in a low-dimensional space. Since the crisp data only code the categories to which each individual case belongs, an alternative measure of fit is simply to count how well these categories are predicted by the solution. Another approach is to consider multiple correspondence analysis equivalently as the analysis of the Burt matrix (i.e., the matrix of all two-way cross-tabulations of the categorical variables), and then perform a joint correspondence analysis to fit just the off-diagonal tables of the Burt matrix - the measure of fit is then computed as the quality of explaining these tables only. The correspondence analysis of fuzzy coded data, called "fuzzy multiple correspondence analysis", suffers from the same problem, albeit attenuated. Again, one can count how many correct predictions are made of the categories which have highest degree of membership. But here one can also defuzzify the results of the analysis to obtain estimated values of the original data, and then calculate a measure of fit in the familiar percentage form, thanks to the resultant orthogonal decomposition of variance. Furthermore, if one thinks of fuzzy multiple correspondence analysis as explaining the two-way associations between variables, a fuzzy Burt matrix can be computed and the same strategy as in the crisp case can be applied to analyse the off-diagonal part of this matrix. In this paper these alternative measures of fit are defined and applied to a data set of continuous meteorological variables, which are coded crisply and fuzzily into three categories. Measuring the fit is further discussed when the data set consists of a mixture of discrete and continuous variables.