Comparison of Image Transform-Based Features for Visual Speech Recognition in Clean and Corrupted Videos

We present results of a study into the performance of a variety of different image transform-based feature types for speaker-independent visual speech recognition of isolated digits. This includes the first reported use of features extracted using a discrete curvelet transform. The study will show a comparison of some methods for selecting features of each feature type and show the relative benefits of both static and dynamic visual features. The performance of the features will be tested on both clean video data and also video data corrupted in a variety of ways to assess each feature type's robustness to potential real-world conditions. One of the test conditions involves a novel form of video corruption we call jitter which simulates camera and/or head movement during recording.

Strategies for adding features to CAD models in order to optimize performance

This paper presents an approach which enables new parameters to be added to a CAD model for optimization purposes. It aims to remove a common roadblock to CAD based optimization, where the parameterization of the model does not offer the shape sufficient flexibility for a truly optimized shape to be created. A technique has been developed which uses adjoint based sensitivity maps to predict
the sensitivity of performance to the addition to a model of four different feature types, allowing the feature providing the greatest benefit to be selected. The optimum position to add the feature is also discussed. It is anticipated that the approach could be used to iteratively add features to a model, providing greater flexibility to the shape of the model, and allowing the newly-added parameters to be used as design variables in a subsequent shape optimization.

CAMP factor homologues in Propionibacterium acnes: a new protein family differentially expressed by types I and II

Analysis of the draft genome sequence of the opportunistic pathogen Propionibacterium acnes type strain NCTC 737 (=ATCC 6919) revealed five genes with sequence identity to the co-haemolytic Christie-Atkins-Munch-Peterson (CAMP) factor of Streptococcus agalactiae. The predicted molecular masses for the expressed proteins ranged from 28 to 30 kDa. The genes were present in each of the three recently identified recA-based phylogenetic groupings of P. acnes (IA, IB and 11), as assessed by PCR amplification. Conserved differences in CAMP factor gene sequences between these three groups were also consistent with their previous phylogenetic designations. All type IA, IB and 11 isolates were positive for the co-haemolytic; reaction on sheep blood agar. Immunoblotting and silver staining of SIDS-PAGE gels, however, revealed differential protein expression of CAMP factors amongst the different groups. Type IB and 11 isolates produced an abundance of CAMP factor 1, detectable by specific antibody labelling and silver staining of SDS-PAGE gels. In contrast, abundant CAMP factor production was lacking in type A isolates, although larger amounts of CAMP factor 2 were detectable by immunoblotting compared with type 11 isolates. While the potential role of the abundant CAMP factor 1 in host colonization or virulence remains to be determined, it should be noted that the type strain of P. acnes used in much of the published literature is a type A isolate and is, therefore, lacking in this attribute.

Teenagers Telling Sectarian Stories

The purpose of this article is to examine the socially constructed nature of the story telling process by drawing on an example from one locality in Northern Ireland. The research draws on focus group interviews with teenagers from polarized working-class communities in North Belfast. The overall locality is divided into Catholic and Protestant areas and a recurring feature of the data is the tendency for each group to define themselves in opposition to the other. Throughout the focus group interviews, the teenagers produced four types of stories and the article assesses the relevance of each type to producing, reproducing or challenging sectarian divisions. The first three groups of stories, First-hand stories, Second-hand stories and Collective stories reflect individual and group attitudes to distinctions between ‘us’ and ‘them’ while the fourth, Alternative stories, questions the homogeneity of the in-group and the immutability of these divisions. These stories verbalize the internal recollections of both individuals and groups and rely on real and imagined memories. The thrust of the article illustrates the ways in which sectarian identities are constructed, shaped and diluted through these narrative encounters.

Speech Recognition with unknown partial feature corruption - a review of the union model

This paper provides a summary of our studies on robust speech recognition based on a new statistical approach – the probabilistic union model. We consider speech recognition given that part of the acoustic features may be corrupted by noise. The union model is a method for basing the recognition on the clean part of the features, thereby reducing the effect of the noise on recognition. To this end, the union model is similar to the missing feature method. However, the two methods achieve this end through different routes. The missing feature method usually requires the identity of the noisy data for noise removal, while the union model combines the local features based on the union of random events, to reduce the dependence of the model on information about the noise. We previously investigated the applications of the union model to speech recognition involving unknown partial corruption in frequency band, in time duration, and in feature streams. Additionally, a combination of the union model with conventional noise-reduction techniques was studied, as a means of dealing with a mixture of known or trainable noise and unknown unexpected noise. In this paper, a unified review, in the context of dealing with unknown partial feature corruption, is provided into each of these applications, giving the appropriate theory and implementation algorithms, along with an experimental evaluation.

Simultaneous feature selection and feature weighting using Hybrid Tabu Search/K-nearest neighbor classifier

Feature selection and feature weighting are useful techniques for improving the classification accuracy of K-nearest-neighbor (K-NN) rule. The term feature selection refers to algorithms that select the best subset of the input feature set. In feature weighting, each feature is multiplied by a weight value proportional to the ability of the feature to distinguish pattern classes. In this paper, a novel hybrid approach is proposed for simultaneous feature selection and feature weighting of K-NN rule based on Tabu Search (TS) heuristic. The proposed TS heuristic in combination with K-NN classifier is compared with several classifiers on various available data sets. The results have indicated a significant improvement in the performance in classification accuracy. The proposed TS heuristic is also compared with various feature selection algorithms. Experiments performed revealed that the proposed hybrid TS heuristic is superior to both simple TS and sequential search algorithms. We also present results for the classification of prostate cancer using multispectral images, an important problem in biomedicine.

Robust Speaker Recognition in Noisy Conditions

This paper investigates the problem of speaker identi-fication and verification in noisy conditions, assuming that speechsignals are corrupted by environmental noise, but knowledgeabout the noise characteristics is not available. This research ismotivated in part by the potential application of speaker recog-nition technologies on handheld devices or the Internet. Whilethe technologies promise an additional biometric layer of securityto protect the user, the practical implementation of such systemsfaces many challenges. One of these is environmental noise. Due tothe mobile nature of such systems, the noise sources can be highlytime-varying and potentially unknown. This raises the require-ment for noise robustness in the absence of information about thenoise. This paper describes a method that combines multicondi-tion model training and missing-feature theory to model noisewith unknown temporal-spectral characteristics. Multiconditiontraining is conducted using simulated noisy data with limitednoise variation, providing a “coarse” compensation for the noise,and missing-feature theory is applied to refine the compensationby ignoring noise variation outside the given training conditions,thereby reducing the training and testing mismatch. This paperis focused on several issues relating to the implementation of thenew model for real-world applications. These include the gener-ation of multicondition training data to model noisy speech, thecombination of different training data to optimize the recognitionperformance, and the reduction of the model’s complexity. Thenew algorithm was tested using two databases with simulated andrealistic noisy speech data. The first database is a redevelopmentof the TIMIT database by rerecording the data in the presence ofvarious noise types, used to test the model for speaker identifica-tion with a focus on the varieties of noise. The second database isa handheld-device database collected in realistic noisy conditions,used to further validate the model for real-world speaker verifica-tion. The new model is compared to baseline systems and is foundto achieve lower error rates.