Real-time video event detection in crowded scenes using MPEG derived features : a multiple instance learning approach

This paper presents an investigation into event detection in crowded scenes, where the event of interest co-occurs with other activities and only binary labels at the clip level are available. The proposed approach incorporates a fast feature descriptor from the MPEG domain, and a novel multiple instance learning (MIL) algorithm using sparse approximation and random sensing. MPEG motion vectors are used to build particle trajectories that represent the motion of objects in uniform video clips, and the MPEG DCT coefficients are used to compute a foreground map to remove background particles. Trajectories are transformed into the Fourier domain, and the Fourier representations are quantized into visual words using the K-Means algorithm. The proposed MIL algorithm models the scene as a linear combination of independent events, where each event is a distribution of visual words. Experimental results show that the proposed approaches achieve promising results for event detection compared to the state-of-the-art.

An efficient hybrid evolutionary optimization algorithm based on PSO and SA for clustering

The K-means algorithm is one of the most popular techniques in clustering. Nevertheless, the performance of the K-means algorithm depends highly on initial cluster centers and converges to local minima. This paper proposes a hybrid evolutionary programming based clustering algorithm, called PSO-SA, by combining particle swarm optimization (PSO) and simulated annealing (SA). The basic idea is to search around the global solution by SA and to increase the information exchange among particles using a mutation operator to escape local optima. Three datasets, Iris, Wisconsin Breast Cancer, and Ripley’s Glass, have been considered to show the effectiveness of the proposed clustering algorithm in providing optimal clusters. The simulation results show that the PSO-SA clustering algorithm not only has a better response but also converges more quickly than the K-means, PSO, and SA algorithms.

Assessment and application of clustering techniques to atmospheric particle number size distribution for the purpose of source apportionment

Long-term measurements of particle number size distribution (PNSD) produce a very large number of observations and their analysis requires an efficient approach in order to produce results in the least possible time and with maximum accuracy. Clustering techniques are a family of sophisticated methods which have been recently employed to analyse PNSD data, however, very little information is available comparing the performance of different clustering techniques on PNSD data. This study aims to apply several clustering techniques (i.e. K-means, PAM, CLARA and SOM) to PNSD data, in order to identify and apply the optimum technique to PNSD data measured at 25 sites across Brisbane, Australia. A new method, based on the Generalised Additive Model (GAM) with a basis of penalised B-splines, was proposed to parameterise the PNSD data and the temporal weight of each cluster was also estimated using the GAM. In addition, each cluster was associated with its possible source based on the results of this parameterisation, together with the characteristics of each cluster. The performances of four clustering techniques were compared using the Dunn index and Silhouette width validation values and the K-means technique was found to have the highest performance, with five clusters being the optimum. Therefore, five clusters were found within the data using the K-means technique. The diurnal occurrence of each cluster was used together with other air quality parameters, temporal trends and the physical properties of each cluster, in order to attribute each cluster to its source and origin. The five clusters were attributed to three major sources and origins, including regional background particles, photochemically induced nucleated particles and vehicle generated particles. Overall, clustering was found to be an effective technique for attributing each particle size spectra to its source and the GAM was suitable to parameterise the PNSD data. These two techniques can help researchers immensely in analysing PNSD data for characterisation and source apportionment purposes.

From ImageNet to mining: Adapting visual object detection with minimal supervision

This paper presents visual detection and classification of light vehicles and personnel on a mine site.We capitalise on the rapid advances of ConvNet based object recognition but highlight that a naive black box approach results in a significant number of false positives. In particular, the lack of domain specific training data and the unique landscape in a mine site causes a high rate of errors. We exploit the abundance of background-only images to train a k-means classifier to complement the ConvNet. Furthermore, localisation of objects of interest and a reduction in computation is enabled through region proposals. Our system is tested on over 10km of real mine site data and we were able to detect both light vehicles and personnel. We show that the introduction of our background model can reduce the false positive rate by an order of magnitude.

Acoustic feature extraction using perceptual wavelet packet decomposition for frog call classification

Frog protection has become increasingly essential due to the rapid decline of its biodiversity. Therefore, it is valuable to develop new methods for studying this biodiversity. In this paper, a novel feature extraction method is proposed based on perceptual wavelet packet decomposition for classifying frog calls in noisy environments. Pre-processing and syllable segmentation are first applied to the frog call. Then, a spectral peak track is extracted from each syllable if possible. Track duration, dominant frequency and oscillation rate are directly extracted from the track. With k-means clustering algorithm, the calculated dominant frequency of all frog species is clustered into k parts, which produce a frequency scale for wavelet packet decomposition. Based on the adaptive frequency scale, wavelet packet decomposition is applied to the frog calls. Using the wavelet packet decomposition coefficients, a new feature set named perceptual wavelet packet decomposition sub-band cepstral coefficients is extracted. Finally, a k-nearest neighbour (k-NN) classifier is used for the classification. The experiment results show that the proposed features can achieve an average classification accuracy of 97.45% which outperforms syllable features (86.87%) and Mel-frequency cepstral coefficients (MFCCs) feature (90.80%).

Traffic and nucleation events as main sources of ultrafine particles in high-insolation developed world cities

Road traffic emissions are often considered the main source of ultrafine particles (UFP, diameter smaller than 100 nm) in urban environments. However, recent studies worldwide have shown that - in high-insolation urban regions at least - new particle formation events can also contribute to UFP. In order to quantify such events we systematically studied three cities located in predominantly sunny environments: Barcelona (Spain), Madrid (Spain) and Brisbane (Australia). Three long term datasets (1-2 years) of fine and ultrafine particle number size distributions (measured by SMPS, Scanning Mobility Particle Sizer) were analysed. Compared to total particle number concentrations, aerosol size distributions offer far more information on the type, origin and atmospheric evolution of the particles. By applying k-Means clustering analysis, we categorized the collected aerosol size distributions in three main categories: “Traffic” (prevailing 44-63% of the time), “Nucleation” (14-19%) and “Background pollution and Specific cases” (7-22%). Measurements from Rome (Italy) and Los Angeles (California) were also included to complement the study. The daily variation of the average UFP concentrations for a typical nucleation day at each site revealed a similar pattern for all cities, with three distinct particle bursts. A morning and an evening spike reflected traffic rush hours, whereas a third one at midday showed nucleation events. The photochemically nucleated particles burst lasted 1-4 hours, reaching sizes of 30-40 nm. On average, the occurrence of particle size spectra dominated by nucleation events was 16% of the time, showing the importance of this process as a source of UFP in urban environments exposed to high solar radiation. On average, nucleation events lasting for 2 hours or more occurred on 55% of the days, this extending to >4hrs in 28% of the days, demonstrating that atmospheric conditions in urban environments are not favourable to the growth of photochemically nucleated particles. In summary, although traffic remains the main source of UFP in urban areas, in developed countries with high insolation urban nucleation events are also a main source of UFP. If traffic-related particle concentrations are reduced in the future, nucleation events will likely increase in urban areas, due to the reduced urban condensation sinks.

Teologian opiskelijoiden henkilökohtaiset projektit ja opintomenestys

Motivation and personal goals play an important role in the ways in which people direct their behavior. Personal goals are closely connected with well-being but they also relate to how people perform in different achievement domains. Many studies show that evaluating study-related goals as important, easy to attain and non stressful, predict better academic achievements than evaluating them as non attainable and stressful (Salmela-Aro & Nurmi, 1997b). The aim of this study was to describe motivational factors among theology students. They form an interesting group in terms of exploring connections between motivation, spiritual goals and academic achievements. The average duration of graduation at the Faculty of Theology is among the highest at the University of Helsinki. On the other hand, it may be assumed that many theology students have spiritual goals which affect their studies. A special focus was paid on the different evaluations of study-related personal projects and how they are related to academic achievement. A methodology of personal projects (Little, 1983) was used to study what kind of personal goals theology students are engaged in during their studies. In the first part of the questionnaire the subjects (N=133) were asked to describe important personal projects. They were given four numbered lines for their written responses. In the second part the subjects were asked to rate projects concerning their studies according to 13 dimensions using a 7-point Likert-scale. Three subgroups were formed on a K-Means Cluster Analysis on the basis of evaluations of the study-related projects. The groups were named committed, self-fulfillers and non-committed according to their evaluations of their study related projects. Academic achievements among the different groups varied substantially. After two years of studying the students who were in the committed group had completed on an average twenty study credits more than those who were in the non-committed group. Self-fulfillers placed in the middle of the three groups. Committed and self-fulfiller students also reported higher levels of intrinsic reasons for striving towards study-related goals. The results indicate that goals reported at the beginning of studies predicted academic achievement later on. The results also showed that different evaluations of goals have long lasting connections to progress in studying. Implications for student well-being and how these results can be utilized for student counseling are discussed.

Insights into the growth of newly formed particles in a subtropical urban environment

The role of different chemical compounds, particularly organics, involved in the new particle formation (NPF) and its consequent growth are not fully understood. Therefore, this study was conducted to investigate the chemical composition of aerosol particles during NPF events in an urban subtropical environment. Aerosol chemical composition was measured along with particle number size distribution (PNSD) and several other air quality parameters at five sites across an urban subtropical environment. An Aerodyne compact Time-of-Flight Aerosol Mass Spectrometer (c-TOF-AMS) and a TSI Scanning Mobility Particle Sizer (SMPS) measured aerosol chemical composition (particles above 50 nm in vacuum aerodynamic diameter) and PNSD (particles within 9-414 nm in mobility diameter), respectively. Five NPF events, with growth rates in the range 3.3-4.6 nm, were detected at two of the sites. The NPF events happened on relatively warmer days with lower condensation sink (CS). Temporal percent fractions of organics increased after the particles grew enough to have a significant contribution to particles volume, while the mass fraction of ammonium and sulphate decreased. This uncovered the important role of organics in the growth of newly formed particles. Three organic markers, factors f43, f44 and f57, were calculated and the f44 vs f43 trends were compared between nucleation and non-nucleation days. K-means cluster analysis was performed on f44 vs f43 data and it was found that they follow different patterns on nucleation days compared to non-nucleation days, whereby f43 decreased for vehicle emission generated particles, while both f44 and f43 decreased for NPF generated particles. It was found for the first time that vehicle generated and newly formed particles cluster in different locations on f44 vs f43 plot and this finding can be potentially used as a tool for source apportionment of measured particles.

A hybrid clustering procedure for concentric and chain-like clusters

K-means algorithm is a well known nonhierarchical method for clustering data. The most important limitations of this algorithm are that: (1) it gives final clusters on the basis of the cluster centroids or the seed points chosen initially, and (2) it is appropriate for data sets having fairly isotropic clusters. But this algorithm has the advantage of low computation and storage requirements. On the other hand, hierarchical agglomerative clustering algorithm, which can cluster nonisotropic (chain-like and concentric) clusters, requires high storage and computation requirements. This paper suggests a new method for selecting the initial seed points, so that theK-means algorithm gives the same results for any input data order. This paper also describes a hybrid clustering algorithm, based on the concepts of multilevel theory, which is nonhierarchical at the first level and hierarchical from second level onwards, to cluster data sets having (i) chain-like clusters and (ii) concentric clusters. It is observed that this hybrid clustering algorithm gives the same results as the hierarchical clustering algorithm, with less computation and storage requirements.

Maternal and paternal support for physical activity and healthy eating in preschool children: A cross-sectional study

BACKGROUND Parental support is a key influence on children's health behaviours; however, no previous investigation has simultaneously explored the influence of mothers' and fathers' social support on eating and physical activity in preschool-aged children. This study evaluated the singular and combined effects of maternal and paternal support for physical activity (PA) and fruit and vegetable consumption (FV) on preschoolers' PA and FV. METHODS A random sample comprising 173 parent-child dyads completed validated scales assessing maternal and paternal instrumental support and child PA and FV behaviour. Pearson correlations, controlling for child age, parental age, and parental education, were used to evaluate relationships between maternal and paternal support and child PA and FV. K-means cluster analysis was used to identify families with distinct patterns of maternal and paternal support for PA and FV, and one-way ANOVA examined the impact of cluster membership on child PA and FV. RESULTS Maternal and paternal support for PA were positively associated with child PA (r = 0.37 and r = 0.36, respectively; P < 0.001). Maternal but not paternal support for FV was positively associated with child FV (r = 0.35; P < 0.001). Five clusters characterised groups of families with distinct configurations of maternal and paternal support for PA and FV: 1) above average maternal and paternal support for PA and FV, 2) below average maternal and paternal support for PA and FV, 3) above average maternal and paternal support for PA but below average maternal and paternal support for FV, 4) above average maternal and paternal support for FV but below average maternal and paternal support for PA, and 5) above average maternal support but below average paternal support for PA and FV. Children from families with above average maternal and paternal support for both health behaviours had higher PA and FV levels than children from families with above average support for just one health behaviour, or below average support for both behaviours. CONCLUSIONS The level and consistency of instrumental support from mothers and fathers for PA and FV may be an important target for obesity prevention in preschool-aged children.

Adaptive frequency scaled wavelet packet decomposition for frog call classification

Environmental changes have put great pressure on biological systems leading to the rapid decline of biodiversity. To monitor this change and protect biodiversity, animal vocalizations have been widely explored by the aid of deploying acoustic sensors in the field. Consequently, large volumes of acoustic data are collected. However, traditional manual methods that require ecologists to physically visit sites to collect biodiversity data are both costly and time consuming. Therefore it is essential to develop new semi-automated and automated methods to identify species in automated audio recordings. In this study, a novel feature extraction method based on wavelet packet decomposition is proposed for frog call classification. After syllable segmentation, the advertisement call of each frog syllable is represented by a spectral peak track, from which track duration, dominant frequency and oscillation rate are calculated. Then, a k-means clustering algorithm is applied to the dominant frequency, and the centroids of clustering results are used to generate the frequency scale for wavelet packet decomposition (WPD). Next, a new feature set named adaptive frequency scaled wavelet packet decomposition sub-band cepstral coefficients is extracted by performing WPD on the windowed frog calls. Furthermore, the statistics of all feature vectors over each windowed signal are calculated for producing the final feature set. Finally, two well-known classifiers, a k-nearest neighbour classifier and a support vector machine classifier, are used for classification. In our experiments, we use two different datasets from Queensland, Australia (18 frog species from commercial recordings and field recordings of 8 frog species from James Cook University recordings). The weighted classification accuracy with our proposed method is 99.5% and 97.4% for 18 frog species and 8 frog species respectively, which outperforms all other comparable methods.

Multiple instance dictionary learning for activity representation

This paper presents an effective feature representation method in the context of activity recognition. Efficient and effective feature representation plays a crucial role not only in activity recognition, but also in a wide range of applications such as motion analysis, tracking, 3D scene understanding etc. In the context of activity recognition, local features are increasingly popular for representing videos because of their simplicity and efficiency. While they achieve state-of-the-art performance with low computational requirements, their performance is still limited for real world applications due to a lack of contextual information and models not being tailored to specific activities. We propose a new activity representation framework to address the shortcomings of the popular, but simple bag-of-words approach. In our framework, first multiple instance SVM (mi-SVM) is used to identify positive features for each action category and the k-means algorithm is used to generate a codebook. Then locality-constrained linear coding is used to encode the features into the generated codebook, followed by spatio-temporal pyramid pooling to convey the spatio-temporal statistics. Finally, an SVM is used to classify the videos. Experiments carried out on two popular datasets with varying complexity demonstrate significant performance improvement over the base-line bag-of-feature method.

Automating Crop Damage Assessments with Unmanned Aerial Systems and Machine Learning

Agricultural pests are responsible for millions of dollars in crop losses and management costs every year. In order to implement optimal site-specific treatments and reduce control costs, new methods to accurately monitor and assess pest damage need to be investigated. In this paper we explore the combination of unmanned aerial vehicles (UAV), remote sensing and machine learning techniques as a promising methodology to address this challenge. The deployment of UAVs as a sensor platform is a rapidly growing field of study for biosecurity and precision agriculture applications. In this experiment, a data collection campaign is performed over a sorghum crop severely damaged by white grubs (Coleoptera: Scarabaeidae). The larvae of these scarab beetles feed on the roots of plants, which in turn impairs root exploration of the soil profile. In the field, crop health status could be classified according to three levels: bare soil where plants were decimated, transition zones of reduced plant density and healthy canopy areas. In this study, we describe the UAV platform deployed to collect high-resolution RGB imagery as well as the image processing pipeline implemented to create an orthoimage. An unsupervised machine learning approach is formulated in order to create a meaningful partition of the image into each of the crop levels. The aim of this approach is to simplify the image analysis step by minimizing user input requirements and avoiding the manual data labelling necessary in supervised learning approaches. The implemented algorithm is based on the K-means clustering algorithm. In order to control high-frequency components present in the feature space, a neighbourhood-oriented parameter is introduced by applying Gaussian convolution kernels prior to K-means clustering. The results show the algorithm delivers consistent decision boundaries that classify the field into three clusters, one for each crop health level as shown in Figure 1. The methodology presented in this paper represents a venue for further esearch towards automated crop damage assessments and biosecurity surveillance.

Assessment of crop insect damage using unmanned aerial systems: A machine learning approach

Agricultural pests are responsible for millions of dollars in crop losses and management costs every year. In order to implement optimal site-specific treatments and reduce control costs, new methods to accurately monitor and assess pest damage need to be investigated. In this paper we explore the combination of unmanned aerial vehicles (UAV), remote sensing and machine learning techniques as a promising technology to address this challenge. The deployment of UAVs as a sensor platform is a rapidly growing field of study for biosecurity and precision agriculture applications. In this experiment, a data collection campaign is performed over a sorghum crop severely damaged by white grubs (Coleoptera: Scarabaeidae). The larvae of these scarab beetles feed on the roots of plants, which in turn impairs root exploration of the soil profile. In the field, crop health status could be classified according to three levels: bare soil where plants were decimated, transition zones of reduced plant density and healthy canopy areas. In this study, we describe the UAV platform deployed to collect high-resolution RGB imagery as well as the image processing pipeline implemented to create an orthoimage. An unsupervised machine learning approach is formulated in order to create a meaningful partition of the image into each of the crop levels. The aim of the approach is to simplify the image analysis step by minimizing user input requirements and avoiding the manual data labeling necessary in supervised learning approaches. The implemented algorithm is based on the K-means clustering algorithm. In order to control high-frequency components present in the feature space, a neighbourhood-oriented parameter is introduced by applying Gaussian convolution kernels prior to K-means. The outcome of this approach is a soft K-means algorithm similar to the EM algorithm for Gaussian mixture models. The results show the algorithm delivers decision boundaries that consistently classify the field into three clusters, one for each crop health level. The methodology presented in this paper represents a venue for further research towards automated crop damage assessments and biosecurity surveillance.