Cross likelihood ratio based speaker clustering using eigenvoice models

This paper proposes the use of eigenvoice modeling techniques with the Cross Likelihood Ratio (CLR) as a criterion for speaker clustering within a speaker diarization system. The CLR has previously been shown to be a robust decision criterion for speaker clustering using Gaussian Mixture Models. Recently, eigenvoice modeling techniques have become increasingly popular, due to its ability to adequately represent a speaker based on sparse training data, as well as an improved capture of differences in speaker characteristics. This paper hence proposes that it would be beneficial to capitalize on the advantages of eigenvoice modeling in a CLR framework. Results obtained on the 2002 Rich Transcription (RT-02) Evaluation dataset show an improved clustering performance, resulting in a 35.1% relative improvement in the overall Diarization Error Rate (DER) compared to the baseline system.

Sparse temporal representations for facial expression recognition

In automatic facial expression recognition, an increasing number of techniques had been proposed for in the literature that exploits the temporal nature of facial expressions. As all facial expressions are known to evolve over time, it is crucially important for a classifier to be capable of modelling their dynamics. We establish that the method of sparse representation (SR) classifiers proves to be a suitable candidate for this purpose, and subsequently propose a framework for expression dynamics to be efficiently incorporated into its current formulation. We additionally show that for the SR method to be applied effectively, then a certain threshold on image dimensionality must be enforced (unlike in facial recognition problems). Thirdly, we determined that recognition rates may be significantly influenced by the size of the projection matrix \Phi. To demonstrate these, a battery of experiments had been conducted on the CK+ dataset for the recognition of the seven prototypic expressions - anger, contempt, disgust, fear, happiness, sadness and surprise - and comparisons have been made between the proposed temporal-SR against the static-SR framework and state-of-the-art support vector machine.

Activity modelling in crowded environments : a soft decision approach

Probabilistic topic models have recently been used for activity analysis in video processing, due to their strong capacity to model both local activities and interactions in crowded scenes. In those applications, a video sequence is divided into a collection of uniform non-overlaping video clips, and the high dimensional continuous inputs are quantized into a bag of discrete visual words. The hard division of video clips, and hard assignment of visual words leads to problems when an activity is split over multiple clips, or the most appropriate visual word for quantization is unclear. In this paper, we propose a novel algorithm, which makes use of a soft histogram technique to compensate for the loss of information in the quantization process; and a soft cut technique in the temporal domain to overcome problems caused by separating an activity into two video clips. In the detection process, we also apply a soft decision strategy to detect unusual events.We show that the proposed soft decision approach outperforms its hard decision counterpart in both local and global activity modelling.

Unusual Event Detection in Crowded Scenes Using Bag of LBPs in spatio-temporal patches

Modelling events in densely crowded environments remains challenging, due to the diversity of events and the noise in the scene. We propose a novel approach for anomalous event detection in crowded scenes using dynamic textures described by the Local Binary Patterns from Three Orthogonal Planes (LBP-TOP) descriptor. The scene is divided into spatio-temporal patches where LBP-TOP based dynamic textures are extracted. We apply hierarchical Bayesian models to detect the patches containing unusual events. Our method is an unsupervised approach, and it does not rely on object tracking or background subtraction. We show that our approach outperforms existing state of the art algorithms for anomalous event detection in UCSD dataset.

i-vector based speaker recognition on short utterances

Robust speaker verification on short utterances remains a key consideration when deploying automatic speaker recognition, as many real world applications often have access to only limited duration speech data. This paper explores how the recent technologies focused around total variability modeling behave when training and testing utterance lengths are reduced. Results are presented which provide a comparison of Joint Factor Analysis (JFA) and i-vector based systems including various compensation techniques; Within-Class Covariance Normalization (WCCN), LDA, Scatter Difference Nuisance Attribute Projection (SDNAP) and Gaussian Probabilistic Linear Discriminant Analysis (GPLDA). Speaker verification performance for utterances with as little as 2 sec of data taken from the NIST Speaker Recognition Evaluations are presented to provide a clearer picture of the current performance characteristics of these techniques in short utterance conditions.

Compressive sensing for gait recognition

Compressive Sensing (CS) is a popular signal processing technique, that can exactly reconstruct a signal given a small number of random projections of the original signal, provided that the signal is sufficiently sparse. We demonstrate the applicability of CS in the field of gait recognition as a very effective dimensionality reduction technique, using the gait energy image (GEI) as the feature extraction process. We compare the CS based approach to the principal component analysis (PCA) and show that the proposed method outperforms this baseline, particularly under situations where there are appearance changes in the subject. Applying CS to the gait features also avoids the need to train the models, by using a generalised random projection.

Visual voice activity detection using frontal versus profile views

Visual activity detection of lip movements can be used to overcome the poor performance of voice activity detection based solely in the audio domain, particularly in noisy acoustic conditions. However, most of the research conducted in visual voice activity detection (VVAD) has neglected addressing variabilities in the visual domain such as viewpoint variation. In this paper we investigate the effectiveness of the visual information from the speaker’s frontal and profile views (i.e left and right side views) for the task of VVAD. As far as we are aware, our work constitutes the first real attempt to study this problem. We describe our visual front end approach and the Gaussian mixture model (GMM) based VVAD framework, and report the experimental results using the freely available CUAVE database. The experimental results show that VVAD is indeed possible from profile views and we give a quantitative comparison of VVAD based on frontal and profile views The results presented are useful in the development of multi-modal Human Machine Interaction (HMI) using a single camera, where the speaker’s face may not always be frontal.

Developing a digital image watermarking model

This paper presents a key based generic model for digital image watermarking. The model aims at addressing an identified gap in the literature by providing a basis for assessing different watermarking requirements in various digital image applications. We start with a formulation of a basic watermarking system, and define system inputs and outputs. We then proceed to incorporate the use of keys in the design of various system components. Using the model, we also define a few fundamental design and evaluation parameters. To demonstrate the significance of the proposed model, we provide an example of how it can be applied to formally define common attacks.

Classification of packet contents for malware detection

Many existing schemes for malware detection are signature-based. Although they can effectively detect known malwares, they cannot detect variants of known malwares or new ones. Most network servers do not expect executable code in their in-bound network traffic, such as on-line shopping malls, Picasa, Youtube, Blogger, etc. Therefore, such network applications can be protected from malware infection by monitoring their ports to see if incoming packets contain any executable contents. This paper proposes a content-classification scheme that identifies executable content in incoming packets. The proposed scheme analyzes the packet payload in two steps. It first analyzes the packet payload to see if it contains multimedia-type data (such as . If not, then it classifies the payload either as text-type (such as or executable. Although in our experiments the proposed scheme shows a low rate of false negatives and positives (4.69% and 2.53%, respectively), the presence of inaccuracies still requires further inspection to efficiently detect the occurrence of malware. In this paper, we also propose simple statistical and combinatorial analysis to deal with false positives and negatives.

Practical improvements to simultaneous computation of multi-view geometry and radial lens distortion

This paper discusses practical issues related to the use of the division model for lens distortion in multi-view geometry computation. A data normalisation strategy is presented, which has been absent from previous discussions on the topic. The convergence properties of the Rectangular Quadric Eigenvalue Problem solution for computing division model distortion are examined. It is shown that the existing method can require more than 1000 iterations when dealing with severe distortion. A method is presented for accelerating convergence to less than 10 iterations for any amount of distortion. The new method is shown to produce equivalent or better results than the existing method with up to two orders of magnitude reduction in iterations. Through detailed simulation it is found that the number of data points used to compute geometry and lens distortion has a strong influence on convergence speed and solution accuracy. It is recommended that more than the minimal number of data points be used when computing geometry using a robust estimator such as RANSAC. Adding two to four extra samples improves the convergence rate and accuracy sufficiently to compensate for the increased number of samples required by the RANSAC process.

Negative determinant of Hessian features

Local image feature extractors that select local maxima of the determinant of Hessian function have been shown to perform well and are widely used. This paper introduces the negative local minima of the determinant of Hessian function for local feature extraction. The properties and scale-space behaviour of these features are examined and found to be desirable for feature extraction. It is shown how this new feature type can be implemented along with the existing local maxima approach at negligible extra processing cost. Applications to affine covariant feature extraction and sub-pixel precise corner extraction are demonstrated. Experimental results indicate that the new corner detector is more robust to image blur and noise than existing methods. It is also accurate for a broader range of corner geometries. An affine covariant feature extractor is implemented by combining the minima of the determinant of Hessian with existing scale and shape adaptation methods. This extractor can be implemented along side the existing Hessian maxima extractor simply by finding both minima and maxima during the initial extraction stage. The minima features increase the number of correspondences by two to four fold. The additional minima features are very distinct from the maxima features in descriptor space and do not make the matching process more ambiguous.

An architecture for enhanced assurance in e-health systems

Notwithstanding the obvious potential advantages of information and communications technology (ICT) in the enhanced provision of healthcare services, there are some concerns associated with integration of and access to electronic health records. A security violation in health records, such as an unauthorised disclosure or unauthorised alteration of an individual's health information, can significantly undermine both healthcare providers' and consumers' confidence and trust in e-health systems. A crisis in confidence in any national level e-health system could seriously degrade the realisation of the system's potential benefits. In response to the privacy and security requirements for the protection of health information, this research project investigated national and international e-health development activities to identify the necessary requirements for the creation of a trusted health information system architecture consistent with legislative and regulatory requirements and relevant health informatics standards. The research examined the appropriateness and sustainability of the current approaches for the protection of health information. It then proposed an architecture to facilitate the viable and sustainable enforcement of privacy and security in health information systems under the project title "Open and Trusted Health Information Systems (OTHIS)". OTHIS addresses necessary security controls to protect sensitive health information when such data is at rest, during processing and in transit with three separate and achievable security function-based concepts and modules: a) Health Informatics Application Security (HIAS); b) Health Informatics Access Control (HIAC); and c) Health Informatics Network Security (HINS). The outcome of this research is a roadmap for a viable and sustainable architecture for providing robust protection and security of health information including elucidations of three achievable security control subsystem requirements within the proposed architecture. The successful completion of two proof-of-concept prototypes demonstrated the comprehensibility, feasibility and practicality of the HIAC and HIAS models for the development and assessment of trusted health systems. Meanwhile, the OTHIS architecture has provided guidance for technical and security design appropriate to the development and implementation of trusted health information systems whilst simultaneously offering guidance for ongoing research projects. The socio-economic implications of this research can be summarised in the fact that this research embraces the need for low cost security strategies against economic realities by using open-source technologies for overall test implementation. This allows the proposed architecture to be publicly accessible, providing a platform for interoperability to meet real-world application security demands. On the whole, the OTHIS architecture sets a high level of security standard for the establishment and maintenance of both current and future health information systems. This thereby increases healthcare providers‘ and consumers‘ trust in the adoption of electronic health records to realise the associated benefits.

Scene invariant crowd counting and crowd occupancy analysis

In public places, crowd size may be an indicator of congestion, delay, instability, or of abnormal events, such as a fight, riot or emergency. Crowd related information can also provide important business intelligence such as the distribution of people throughout spaces, throughput rates, and local densities. A major drawback of many crowd counting approaches is their reliance on large numbers of holistic features, training data requirements of hundreds or thousands of frames per camera, and that each camera must be trained separately. This makes deployment in large multi-camera environments such as shopping centres very costly and difficult. In this chapter, we present a novel scene-invariant crowd counting algorithm that uses local features to monitor crowd size. The use of local features allows the proposed algorithm to calculate local occupancy statistics, scale to conditions which are unseen in the training data, and be trained on significantly less data. Scene invariance is achieved through the use of camera calibration, allowing the system to be trained on one or more viewpoints and then deployed on any number of new cameras for testing without further training. A pre-trained system could then be used as a ‘turn-key’ solution for crowd counting across a wide range of environments, eliminating many of the costly barriers to deployment which currently exist.

Data flow analysis of embedded program expressions

Data flow analysis techniques can be used to help assess threats to data confidentiality and integrity in security critical program code. However, a fundamental weakness of static analysis techniques is that they overestimate the ways in which data may propagate at run time. Discounting large numbers of these false-positive data flow paths wastes an information security evaluator's time and effort. Here we show how to automatically eliminate some false-positive data flow paths by precisely modelling how classified data is blocked by certain expressions in embedded C code. We present a library of detailed data flow models of individual expression elements and an algorithm for introducing these components into conventional data flow graphs. The resulting models can be used to accurately trace byte-level or even bit-level data flow through expressions that are normally treated as atomic. This allows us to identify expressions that safely downgrade their classified inputs and thereby eliminate false-positive data flow paths from the security evaluation process. To validate the approach we have implemented and tested it in an existing data flow analysis toolkit.

A taint marking approach to confidentiality violation detection

This article presents a novel approach to confidentiality violation detection based on taint marking. Information flows are dynamically tracked between applications and objects of the operating system such as files, processes and sockets. A confidentiality policy is defined by labelling sensitive information and defining which information may leave the local system through network exchanges. Furthermore, per application profiles can be defined to restrict the sets of information each application may access and/or send through the network. In previous works, we focused on the use of mandatory access control mechanisms for information flow tracking. In this current work, we have extended the previous information flow model to track network exchanges, and we are able to define a policy attached to network sockets. We show an example application of this extension in the context of a compromised web browser: our implementation detects a confidentiality violation when the browser attempts to leak private information to a remote host over the network.