Cohort normalization based sparse representation for undersampled face recognition

Abstract. In recent years, sparse representation based classification(SRC) has received much attention in face recognition with multipletraining samples of each subject. However, it cannot be easily applied toa recognition task with insufficient training samples under uncontrolledenvironments. On the other hand, cohort normalization, as a way of mea-suring the degradation effect under challenging environments in relationto a pool of cohort samples, has been widely used in the area of biometricauthentication. In this paper, for the first time, we introduce cohort nor-malization to SRC-based face recognition with insufficient training sam-ples. Specifically, a user-specific cohort set is selected to normalize theraw residual, which is obtained from comparing the test sample with itssparse representations corresponding to the gallery subject, using poly-nomial regression. Experimental results on AR and FERET databases show that cohort normalization can bring SRC much robustness against various forms of degradation factors for undersampled face recognition.

Improving short utterance based I-vector speaker recognition using source and utterance-duration normalization techniques

A significant amount of speech is typically required for speaker verification system development and evaluation, especially in the presence of large intersession variability. This paper introduces a source and utterance duration normalized linear discriminant analysis (SUN-LDA) approaches to compensate session variability in short-utterance i-vector speaker verification systems. Two variations of SUN-LDA are proposed where normalization techniques are used to capture source variation from both short and full-length development i-vectors, one based upon pooling (SUN-LDA-pooled) and the other on concatenation (SUN-LDA-concat) across the duration and source-dependent session variation. Both the SUN-LDA-pooled and SUN-LDA-concat techniques are shown to provide improvement over traditional LDA on NIST 08 truncated 10sec-10sec evaluation conditions, with the highest improvement obtained with the SUN-LDA-concat technique achieving a relative improvement of 8% in EER for mis-matched conditions and over 3% for matched conditions over traditional LDA approaches.

Dataset-invariant covariance normalization for out-domain PLDA speaker verification

In this paper we introduce a novel domain-invariant covariance normalization (DICN) technique to relocate both in-domain and out-domain i-vectors into a third dataset-invariant space, providing an improvement for out-domain PLDA speaker verification with a very small number of unlabelled in-domain adaptation i-vectors. By capturing the dataset variance from a global mean using both development out-domain i-vectors and limited unlabelled in-domain i-vectors, we could obtain domain- invariant representations of PLDA training data. The DICN- compensated out-domain PLDA system is shown to perform as well as in-domain PLDA training with as few as 500 unlabelled in-domain i-vectors for NIST-2010 SRE and 2000 unlabelled in-domain i-vectors for NIST-2008 SRE, and considerable relative improvement over both out-domain and in-domain PLDA development if more are available.

A biomechanical approach to iris normalization

The richness of the iris texture and its variability across individuals make it a useful biometric trait for personal authentication. One of the key stages in classical iris recognition is the normalization process, where the annular iris region is mapped to a dimensionless pseudo-polar coordinate system. This process results in a rectangular structure that can be used to compensate for differences in scale and variations in pupil size. Most iris recognition methods in the literature adopt linear sampling in the radial and angular directions when performing iris normalization. In this paper, a biomechanical model of the iris is used to define a novel nonlinear normalization scheme that improves iris recognition accuracy under different degrees of pupil dilation. The proposed biomechanical model is used to predict the radial displacement of any point in the iris at a given dilation level, and this information is incorporated in the normalization process. Experimental results on the WVU pupil light reflex database (WVU-PLR) indicate the efficacy of the proposed technique, especially when matching iris images with large differences in pupil size.

The hunt for disability: The new eugenics and the normalization of school children

Examines issues of sameness, difference, equality, and democracy in present public school systems, focusing on the question of (dis)ability and implications of rethinking (dis)ability as an ontological issue before its inscription as an educational one concerning the politics of inclusion. The paper analyzes old and new discourses of eugenics as quality control of national populations. (Contains references.) (SM)

Design of automatic vision-based inspection system for solder joint segmentation

Purpose: Computer vision has been widely used in the inspection of electronic components. This paper proposes a computer vision system for the automatic detection, localisation, and segmentation of solder joints on Printed Circuit Boards (PCBs) under different illumination conditions. Design/methodology/approach: An illumination normalization approach is applied to an image, which can effectively and efficiently eliminate the effect of uneven illumination while keeping the properties of the processed image the same as in the corresponding image under normal lighting conditions. Consequently special lighting and instrumental setup can be reduced in order to detect solder joints. These normalised images are insensitive to illumination variations and are used for the subsequent solder joint detection stages. In the segmentation approach, the PCB image is transformed from an RGB color space to a YIQ color space for the effective detection of solder joints from the background. Findings: The segmentation results show that the proposed approach improves the performance significantly for images under varying illumination conditions. Research limitations/implications: This paper proposes a front-end system for the automatic detection, localisation, and segmentation of solder joint defects. Further research is required to complete the full system including the classification of solder joint defects. Practical implications: The methodology presented in this paper can be an effective method to reduce cost and improve quality in production of PCBs in the manufacturing industry. Originality/value: This research proposes the automatic location, identification and segmentation of solder joints under different illumination conditions.

Hierarchical models for 2D presence/absence data having ambiguous zeroes: With a biogeographical case study on dingo behaviour

This dissertation is primarily an applied statistical modelling investigation, motivated by a case study comprising real data and real questions. Theoretical questions on modelling and computation of normalization constants arose from pursuit of these data analytic questions. The essence of the thesis can be described as follows. Consider binary data observed on a two-dimensional lattice. A common problem with such data is the ambiguity of zeroes recorded. These may represent zero response given some threshold (presence) or that the threshold has not been triggered (absence). Suppose that the researcher wishes to estimate the effects of covariates on the binary responses, whilst taking into account underlying spatial variation, which is itself of some interest. This situation arises in many contexts and the dingo, cypress and toad case studies described in the motivation chapter are examples of this. Two main approaches to modelling and inference are investigated in this thesis. The first is frequentist and based on generalized linear models, with spatial variation modelled by using a block structure or by smoothing the residuals spatially. The EM algorithm can be used to obtain point estimates, coupled with bootstrapping or asymptotic MLE estimates for standard errors. The second approach is Bayesian and based on a three- or four-tier hierarchical model, comprising a logistic regression with covariates for the data layer, a binary Markov Random field (MRF) for the underlying spatial process, and suitable priors for parameters in these main models. The three-parameter autologistic model is a particular MRF of interest. Markov chain Monte Carlo (MCMC) methods comprising hybrid Metropolis/Gibbs samplers is suitable for computation in this situation. Model performance can be gauged by MCMC diagnostics. Model choice can be assessed by incorporating another tier in the modelling hierarchy. This requires evaluation of a normalization constant, a notoriously difficult problem. Difficulty with estimating the normalization constant for the MRF can be overcome by using a path integral approach, although this is a highly computationally intensive method. Different methods of estimating ratios of normalization constants (N Cs) are investigated, including importance sampling Monte Carlo (ISMC), dependent Monte Carlo based on MCMC simulations (MCMC), and reverse logistic regression (RLR). I develop an idea present though not fully developed in the literature, and propose the Integrated mean canonical statistic (IMCS) method for estimating log NC ratios for binary MRFs. The IMCS method falls within the framework of the newly identified path sampling methods of Gelman & Meng (1998) and outperforms ISMC, MCMC and RLR. It also does not rely on simplifying assumptions, such as ignoring spatio-temporal dependence in the process. A thorough investigation is made of the application of IMCS to the three-parameter Autologistic model. This work introduces background computations required for the full implementation of the four-tier model in Chapter 7. Two different extensions of the three-tier model to a four-tier version are investigated. The first extension incorporates temporal dependence in the underlying spatio-temporal process. The second extensions allows the successes and failures in the data layer to depend on time. The MCMC computational method is extended to incorporate the extra layer. A major contribution of the thesis is the development of a fully Bayesian approach to inference for these hierarchical models for the first time. Note: The author of this thesis has agreed to make it open access but invites people downloading the thesis to send her an email via the 'Contact Author' function.

Automatic spoken language identification utilizing acoustic and phonetic speech information

Automatic spoken Language Identi¯cation (LID) is the process of identifying the language spoken within an utterance. The challenge that this task presents is that no prior information is available indicating the content of the utterance or the identity of the speaker. The trend of globalization and the pervasive popularity of the Internet will amplify the need for the capabilities spoken language identi¯ca- tion systems provide. A prominent application arises in call centers dealing with speakers speaking di®erent languages. Another important application is to index or search huge speech data archives and corpora that contain multiple languages. The aim of this research is to develop techniques targeted at producing a fast and more accurate automatic spoken LID system compared to the previous National Institute of Standards and Technology (NIST) Language Recognition Evaluation. Acoustic and phonetic speech information are targeted as the most suitable fea- tures for representing the characteristics of a language. To model the acoustic speech features a Gaussian Mixture Model based approach is employed. Pho- netic speech information is extracted using existing speech recognition technol- ogy. Various techniques to improve LID accuracy are also studied. One approach examined is the employment of Vocal Tract Length Normalization to reduce the speech variation caused by di®erent speakers. A linear data fusion technique is adopted to combine the various aspects of information extracted from speech. As a result of this research, a LID system was implemented and presented for evaluation in the 2003 Language Recognition Evaluation conducted by the NIST.

Digit recognition using trispectral features

Features derived from the trispectra of DFT magnitude slices are used for multi-font digit recognition. These features are insensitive to translation, rotation, or scaling of the input. They are also robust to noise. Classification accuracy tests were conducted on a common data base of 256× 256 pixel bilevel images of digits in 9 fonts. Randomly rotated and translated noisy versions were used for training and testing. The results indicate that the trispectral features are better than moment invariants and affine moment invariants. They achieve a classification accuracy of 95% compared to about 81% for Hu's (1962) moment invariants and 39% for the Flusser and Suk (1994) affine moment invariants on the same data in the presence of 1% impulse noise using a 1-NN classifier. For comparison, a multilayer perceptron with no normalization for rotations and translations yields 34% accuracy on 16× 16 pixel low-pass filtered and decimated versions of the same data.

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.

PLDA based speaker recognition on short utterances

This paper investigates the effects of limited speech data in the context of speaker verification using a probabilistic linear discriminant analysis (PLDA) approach. Being able to reduce the length of required speech data is important to the development of automatic speaker verification system in real world applications. When sufficient speech is available, previous research has shown that heavy-tailed PLDA (HTPLDA) modeling of speakers in the i-vector space provides state-of-the-art performance, however, the robustness of HTPLDA to the limited speech resources in development, enrolment and verification is an important issue that has not yet been investigated. In this paper, we analyze the speaker verification performance with regards to the duration of utterances used for both speaker evaluation (enrolment and verification) and score normalization and PLDA modeling during development. Two different approaches to total-variability representation are analyzed within the PLDA approach to show improved performance in short-utterance mismatched evaluation conditions and conditions for which insufficient speech resources are available for adequate system development. The results presented within this paper using the NIST 2008 Speaker Recognition Evaluation dataset suggest that the HTPLDA system can continue to achieve better performance than Gaussian PLDA (GPLDA) as evaluation utterance lengths are decreased. We also highlight the importance of matching durations for score normalization and PLDA modeling to the expected evaluation conditions. Finally, we found that a pooled total-variability approach to PLDA modeling can achieve better performance than the traditional concatenated total-variability approach for short utterances in mismatched evaluation conditions and conditions for which insufficient speech resources are available for adequate system development.

Investigating homework as a social practice : a qualitative approach

This thesis examines the social practice of homework. It explores how homework is shaped by the discourses, policies and guidelines in circulation in a society at any given time with particular reference to one school district in the province of Newfoundland and Labrador, Canada. This study investigates how contemporary homework reconstitutes the home as a pedagogical site where the power of the institution of schooling circulates regularly from school to home. It examines how the educational system shapes the organization of family life and how family experiences with homework may be different in different sites depending on the accessibility of various forms of cultural capital. This study employs a qualitative approach, incorporating multiple case studies, and is complemented by insights from institutional ethnography and critical discourse analysis. It draws on the theoretical concepts of Foucault including power and power relations, and governmentality and surveillance, as well as Bourdieu’s concepts of economic, social and cultural capital for analysis. It employs concepts from Bourdieu’s work as they have been expanded on by researchers including Reay (1998), Lareau (2000), and Griffith and Smith (2005). The studies of these researchers allowed for an examination of homework as it related to families and mothers’ work. Smith’s (1987; 1999) concepts of ruling relations, mothers’ unpaid labour, and the engine of inequality were also employed in the analysis. Family interviews with ten volunteer families, teacher focus group sessions with 15 teachers from six schools, homework artefacts, school newsletters, homework brochures, and publicly available assessment and evaluation policy documents from one school district were analyzed. From this analysis key themes emerged and the findings are documented throughout five data analysis chapters. This study shows a change in education in response to a system shaped by standards, accountability and testing. It documents an increased transference of educational responsibility from one educational stakeholder to another. This transference of responsibility shifts downward until it eventually reaches the family in the form of homework and educational activities. Texts in the form of brochures and newsletters, sent home from school, make available to parents specific subject positions that act as instruments of normalization. These subject positions promote a particular ‘ideal’ family that has access to certain types of cultural capital needed to meet the school’s expectations. However, the study shows that these resources are not equally available to all and some families struggle to obtain what is necessary to complete educational activities in the home. The increase in transference of educational work from the school to the home results in greater work for parents, particularly mothers. As well, consideration is given to mother’s role in homework and how, in turn, classroom instructional practices are sometimes dependent on the work completed at home with differential effects for children. This study confirms previous findings that it is mothers who assume the greatest role in the educational trajectory of their children. An important finding in this research is that it is not only middle-class mothers who dedicate extensive time working hard to ensure their children’s educational success; working-class mothers also make substantial contributions of time and resources to their children’s education. The assignments and educational activities distributed as homework require parents’ knowledge of technical school pedagogy to help their children. Much of the homework being sent home from schools is in the area of literacy, particularly reading, but requires parents to do more than read with children. A key finding is that the practices of parents are changing and being reconfigured by the expectations of schools in regard to reading. Parents are now being required to monitor and supervise children’s reading, as well as help children complete reading logs, written reading responses, and follow up questions. The reality of family life as discussed by the participants in this study does not match the ‘ideal’ as portrayed in the educational documents. Homework sessions often create frustrations and tensions between parents and children. Some of the greatest struggles for families were created by mathematical homework, homework for those enrolled in the French Immersion program, and the work required to complete Literature, Heritage and Science Fair projects. Even when institutionalized and objectified capital was readily available, many families still encountered struggles when trying to carry out the assigned educational tasks. This thesis argues that homework and education-related activities play out differently in different homes. Consideration of this significance may assist educators to better understand and appreciate the vast difference in families and the ways in which each family can contribute to their children’s educational trajectory.

Long exposure localization in darkness using consumer cameras

In this paper we demonstrate passive vision-based localization in environments more than two orders of magnitude darker than the current benchmark using a 100 webcam and a 500 camera. Our approach uses the camera’s maximum exposure duration and sensor gain to achieve appropriately exposed images even in unlit night-time environments, albeit with extreme levels of motion blur. Using the SeqSLAM algorithm, we first evaluate the effect of variable motion blur caused by simulated exposures of 132 ms to 10000 ms duration on localization performance. We then use actual long exposure camera datasets to demonstrate day-night localization in two different environments. Finally we perform a statistical analysis that compares the baseline performance of matching unprocessed greyscale images to using patch normalization and local neighbourhood normalization – the two key SeqSLAM components. Our results and analysis show for the first time why the SeqSLAM algorithm is effective, and demonstrate the potential for cheap camera-based localization systems that function across extreme perceptual change.

Rural self-reliance: the impact on health experiences of people living with type II diabetes in rural Queensland, Australia

Objective: The objective of the study was to explore whether and how rural culture influences type II diabetes management and to better understand the social processes that rural people construct in coping with diabetes and its complications. In particular, the study aimed to analyse the interface and interactions between rural people with type II diabetes and the Australian health care system, and to develop a theoretical understanding that reflects constructs that may be more broadly applicable. Methods: The study applied constructivist grounded theory methods within an interpretive interactionist framework. Data from 39 semi-structured interviews with rural and urban type II diabetes patients and a mix of rural health care providers were analysed to develop a theoretical understanding of the social processes that define diabetes management in that context. Results: The analysis suggests that although type II diabetes imposes limitations that require adjustment and adaptation, these processes are actively negotiated by rural people within the environmental context to fit the salient social understandings of autonomy and self-reliance. Thus, people normalized self-reliant diabetes management behaviours because this was congruent with the rural culture. Factors that informed the actions of normalization were relationships between participants and health care professionals, support, and access to individual resources. Conclusions: The findings point to ways in which rural self-reliance is conceived as the primary strategy of diabetes management. People face the paradox of engaging with a health care system that at the same time maximizes individual responsibility for health and minimizes the social support by which individuals manage the condition. The emphasis on self-reliance gives some legitimacy to a lack of prevention and chronic care services. Success of diabetes management behaviours is, however, contingent on relative resources. Where there is good primary care, there develops a number of downstream effects including a sense of empowerment to manage difficult rural environmental circumstances. This has particular bearing on health outcomes for people with fewer resources.

Addressing accuracy and precision issues in iTRAQ quantitation

iTRAQ (isobaric tags for relative or absolute quantitation) is a mass spectrometry technology that allows quantitative comparison of protein abundance by measuring peak intensities of reporter ions released from iTRAQ-tagged peptides by fragmentation during MS/MS. However, current data analysis techniques for iTRAQ struggle to report reliable relative protein abundance estimates and suffer with problems of precision and accuracy. The precision of the data is affected by variance heterogeneity: low signal data have higher relative variability; however, low abundance peptides dominate data sets. Accuracy is compromised as ratios are compressed toward 1, leading to underestimation of the ratio. This study investigated both issues and proposed a methodology that combines the peptide measurements to give a robust protein estimate even when the data for the protein are sparse or at low intensity. Our data indicated that ratio compression arises from contamination during precursor ion selection, which occurs at a consistent proportion within an experiment and thus results in a linear relationship between expected and observed ratios. We proposed that a correction factor can be calculated from spiked proteins at known ratios. Then we demonstrated that variance heterogeneity is present in iTRAQ data sets irrespective of the analytical packages, LC-MS/MS instrumentation, and iTRAQ labeling kit (4-plex or 8-plex) used. We proposed using an additive-multiplicative error model for peak intensities in MS/MS quantitation and demonstrated that a variance-stabilizing normalization is able to address the error structure and stabilize the variance across the entire intensity range. The resulting uniform variance structure simplifies the downstream analysis. Heterogeneity of variance consistent with an additive-multiplicative model has been reported in other MS-based quantitation including fields outside of proteomics; consequently the variance-stabilizing normalization methodology has the potential to increase the capabilities of MS in quantitation across diverse areas of biology and chemistry.