Robust image hash functions using higher order spectra

Robust hashing is an emerging field that can be used to hash certain data types in applications unsuitable for traditional cryptographic hashing methods. Traditional hashing functions have been used extensively for data/message integrity, data/message authentication, efficient file identification and password verification. These applications are possible because the hashing process is compressive, allowing for efficient comparisons in the hash domain but non-invertible meaning hashes can be used without revealing the original data. These techniques were developed with deterministic (non-changing) inputs such as files and passwords. For such data types a 1-bit or one character change can be significant, as a result the hashing process is sensitive to any change in the input. Unfortunately, there are certain applications where input data are not perfectly deterministic and minor changes cannot be avoided. Digital images and biometric features are two types of data where such changes exist but do not alter the meaning or appearance of the input. For such data types cryptographic hash functions cannot be usefully applied. In light of this, robust hashing has been developed as an alternative to cryptographic hashing and is designed to be robust to minor changes in the input. Although similar in name, robust hashing is fundamentally different from cryptographic hashing. Current robust hashing techniques are not based on cryptographic methods, but instead on pattern recognition techniques. Modern robust hashing algorithms consist of feature extraction followed by a randomization stage that introduces non-invertibility and compression, followed by quantization and binary encoding to produce a binary hash output. In order to preserve robustness of the extracted features, most randomization methods are linear and this is detrimental to the security aspects required of hash functions. Furthermore, the quantization and encoding stages used to binarize real-valued features requires the learning of appropriate quantization thresholds. How these thresholds are learnt has an important effect on hashing accuracy and the mere presence of such thresholds are a source of information leakage that can reduce hashing security. This dissertation outlines a systematic investigation of the quantization and encoding stages of robust hash functions. While existing literature has focused on the importance of quantization scheme, this research is the first to emphasise the importance of the quantizer training on both hashing accuracy and hashing security. The quantizer training process is presented in a statistical framework which allows a theoretical analysis of the effects of quantizer training on hashing performance. This is experimentally verified using a number of baseline robust image hashing algorithms over a large database of real world images. This dissertation also proposes a new randomization method for robust image hashing based on Higher Order Spectra (HOS) and Radon projections. The method is non-linear and this is an essential requirement for non-invertibility. The method is also designed to produce features more suited for quantization and encoding. The system can operate without the need for quantizer training, is more easily encoded and displays improved hashing performance when compared to existing robust image hashing algorithms. The dissertation also shows how the HOS method can be adapted to work with biometric features obtained from 2D and 3D face images.

A pattern based two-stage text classifier

In a classification problem typically we face two challenging issues, the diverse characteristic of negative documents and sometimes a lot of negative documents that are closed to positive documents. Therefore, it is hard for a single classifier to clearly classify incoming documents into classes. This paper proposes a novel gradual problem solving to create a two-stage classifier. The first stage identifies reliable negatives (negative documents with weak positive characteristics). It concentrates on minimizing the number of false negative documents (recall-oriented). We use Rocchio, an existing recall based classifier, for this stage. The second stage is a precision-oriented “fine tuning”, concentrates on minimizing the number of false positive documents by applying pattern (a statistical phrase) mining techniques. In this stage a pattern-based scoring is followed by threshold setting (thresholding). Experiment shows that our statistical phrase based two-stage classifier is promising.

Plasma metabolomics reveals biomarkers of the atherosclerosis

Atherosclerotic cardiovascular disease remains the leading cause of morbidity and mortality in industrialized societies. The lack of metabolite biomarkers has impeded the clinical diagnosis of atherosclerosis so far. In this study, stable atherosclerosis patients (n=16) and age- and sex-matched non-atherosclerosis healthy subjects (n=28) were recruited from the local community (Harbin, P. R. China). The plasma was collected from each study subject and was subjected to metabolomics analysis by GC/MS. Pattern recognition analyses (principal components analysis, orthogonal partial least-squares discriminate analysis, and hierarchical clustering analysis) commonly demonstrated plasma metabolome, which was significantly different from atherosclerotic and non-atherosclerotic subjects. The development of atherosclerosis-induced metabolic perturbations of fatty acids, such as palmitate, stearate, and 1-monolinoleoylglycerol, was confirmed consistent with previous publication, showing that palmitate significantly contributes to atherosclerosis development via targeting apoptosis and inflammation pathways. Altogether, this study demonstrated that the development of atherosclerosis directly perturbed fatty acid metabolism, especially that of palmitate, which was confirmed as a phenotypic biomarker for clinical diagnosis of atherosclerosis.

Histogram of weighted local directions for gait recognition

In this paper, we explore the effectiveness of patch-based gradient feature extraction methods when applied to appearance-based gait recognition. Extending existing popular feature extraction methods such as HOG and LDP, we propose a novel technique which we term the Histogram of Weighted Local Directions (HWLD). These 3 methods are applied to gait recognition using the GEI feature, with classification performed using SRC. Evaluations on the CASIA and OULP datasets show significant improvements using these patch-based methods over existing implementations, with the proposed method achieving the highest recognition rate for the respective datasets. In addition, the HWLD can easily be extended to 3D, which we demonstrate using the GEV feature on the DGD dataset, observing improvements in performance.

Pre-crash traffic flow trend analysis on motorways

Crashes on motorway contribute to a significant proportion (40-50%) of non-recurrent motorway congestions. Hence reduce crashes will help address congestion issues (Meyer, 2008). Crash likelihood estimation studies commonly focus on traffic conditions in a Short time window around the time of crash while longer-term pre-crash traffic flow trends are neglected. In this paper we will show, through data mining techniques, that a relationship between pre-crash traffic flow patterns and crash occurrence on motorways exists, and that this knowledge has the potential to improve the accuracy of existing models and opens the path for new development approaches. The data for the analysis was extracted from records collected between 2007 and 2009 on the Shibuya and Shinjuku lines of the Tokyo Metropolitan Expressway in Japan. The dataset includes a total of 824 rear-end and sideswipe crashes that have been matched with traffic flow data of one hour prior to the crash using an incident detection algorithm. Traffic flow trends (traffic speed/occupancy time series) revealed that crashes could be clustered with regards of the dominant traffic flow pattern prior to the crash. Using the k-means clustering method allowed the crashes to be clustered based on their flow trends rather than their distance. Four major trends have been found in the clustering results. Based on these findings, crash likelihood estimation algorithms can be fine-tuned based on the monitored traffic flow conditions with a sliding window of 60 minutes to increase accuracy of the results and minimize false alarms.

Timed Probabilistic Automaton : a bridge between Raven and Song Scope for automatic species recognition

Raven and Song Scope are two automated sound anal-ysis tools based on machine learning technique for en-vironmental monitoring. Many research works have been conducted upon them, however, no or rare explo-ration mentions about the performance and comparison between them. This paper investigates the comparisons from six aspects: theory, software interface, ease of use, detection targets, detection accuracy, and potential application. Through deep exploration one critical gap is identified that there is a lack of approach to detect both syllables and call structures, since Raven only aims to detect syllables while Song Scope targets call structures. Therefore, a Timed Probabilistic Automata (TPA) system is proposed which separates syllables first and clusters them into complex structures after.

Traffic safety risks trends and patterns analysis on motorways

Crashes that occur on motorways contribute to a significant proportion (40-50%) of non-recurrent motorway congestion. Hence, reducing the frequency of crashes assist in addressing congestion issues (Meyer, 2008). Analysing traffic conditions and discovering risky traffic trends and patterns are essential basics in crash likelihood estimations studies and still require more attention and investigation. In this paper we will show, through data mining techniques, that there is a relationship between pre-crash traffic flow patterns and crash occurrence on motorways, compare them with normal traffic trends, and that this knowledge has the potentiality to improve the accuracy of existing crash likelihood estimation models, and opens the path for new development approaches. The data for the analysis was extracted from records collected between 2007 and 2009 on the Shibuya and Shinjuku lines of the Tokyo Metropolitan Expressway in Japan. The dataset includes a total of 824 rear-end and sideswipe crashes that have been matched with crashes corresponding traffic flow data using an incident detection algorithm. Traffic trends (traffic speed time series) revealed that crashes can be clustered with regards to the dominant traffic patterns prior to the crash occurrence. K-Means clustering algorithm applied to determine dominant pre-crash traffic patterns. In the first phase of this research, traffic regimes identified by analysing crashes and normal traffic situations using half an hour speed in upstream locations of crashes. Then, the second phase investigated the different combination of speed risk indicators to distinguish crashes from normal traffic situations more precisely. Five major trends have been found in the first phase of this paper for both high risk and normal conditions. The study discovered traffic regimes had differences in the speed trends. Moreover, the second phase explains that spatiotemporal difference of speed is a better risk indicator among different combinations of speed related risk indicators. Based on these findings, crash likelihood estimation models can be fine-tuned to increase accuracy of estimations and minimize false alarms.

Improved sub-cellular resolution via simultaneous analysis of organelle proteomics data across varied experimental conditions

Spatial organisation of proteins according to their function plays an important role in the specificity of their molecular interactions. Emerging proteomics methods seek to assign proteins to sub-cellular locations by partial separation of organelles and computational analysis of protein abundance distributions among partially separated fractions. Such methods permit simultaneous analysis of unpurified organelles and promise proteome-wide localisation in scenarios wherein perturbation may prompt dynamic re-distribution. Resolving organelles that display similar behavior during a protocol designed to provide partial enrichment represents a possible shortcoming. We employ the Localisation of Organelle Proteins by Isotope Tagging (LOPIT) organelle proteomics platform to demonstrate that combining information from distinct separations of the same material can improve organelle resolution and assignment of proteins to sub-cellular locations. Two previously published experiments, whose distinct gradients are alone unable to fully resolve six known protein-organelle groupings, are subjected to a rigorous analysis to assess protein-organelle association via a contemporary pattern recognition algorithm. Upon straightforward combination of single-gradient data, we observe significant improvement in protein-organelle association via both a non-linear support vector machine algorithm and partial least-squares discriminant analysis. The outcome yields suggestions for further improvements to present organelle proteomics platforms, and a robust analytical methodology via which to associate proteins with sub-cellular organelles.

De-Identification of health records using Anonym: Effectiveness and robustness across datasets

Objective Evaluate the effectiveness and robustness of Anonym, a tool for de-identifying free-text health records based on conditional random fields classifiers informed by linguistic and lexical features, as well as features extracted by pattern matching techniques. De-identification of personal health information in electronic health records is essential for the sharing and secondary usage of clinical data. De-identification tools that adapt to different sources of clinical data are attractive as they would require minimal intervention to guarantee high effectiveness. Methods and Materials The effectiveness and robustness of Anonym are evaluated across multiple datasets, including the widely adopted Integrating Biology and the Bedside (i2b2) dataset, used for evaluation in a de-identification challenge. The datasets used here vary in type of health records, source of data, and their quality, with one of the datasets containing optical character recognition errors. Results Anonym identifies and removes up to 96.6% of personal health identifiers (recall) with a precision of up to 98.2% on the i2b2 dataset, outperforming the best system proposed in the i2b2 challenge. The effectiveness of Anonym across datasets is found to depend on the amount of information available for training. Conclusion Findings show that Anonym compares to the best approach from the 2006 i2b2 shared task. It is easy to retrain Anonym with new datasets; if retrained, the system is robust to variations of training size, data type and quality in presence of sufficient training data.

Automated species recognition in environmental recordings

Faunal vocalisations are vital indicators for environmental change and faunal vocalisation analysis can provide information for answering ecological questions. Therefore, automated species recognition in environmental recordings has become a critical research area. This thesis presents an automated species recognition approach named Timed and Probabilistic Automata. A small lexicon for describing animal calls is defined, six algorithms for acoustic component detection are developed, and a series of species recognisers are built and evaluated.The presented automated species recognition approach yields significant improvement on the analysis performance over a real world dataset, and may be transferred to commercial software in the future.

Random projections on manifolds of symmetric positive definite matrices for image classification

Recent advances suggest that encoding images through Symmetric Positive Definite (SPD) matrices and then interpreting such matrices as points on Riemannian manifolds can lead to increased classification performance. Taking into account manifold geometry is typically done via (1) embedding the manifolds in tangent spaces, or (2) embedding into Reproducing Kernel Hilbert Spaces (RKHS). While embedding into tangent spaces allows the use of existing Euclidean-based learning algorithms, manifold shape is only approximated which can cause loss of discriminatory information. The RKHS approach retains more of the manifold structure, but may require non-trivial effort to kernelise Euclidean-based learning algorithms. In contrast to the above approaches, in this paper we offer a novel solution that allows SPD matrices to be used with unmodified Euclidean-based learning algorithms, with the true manifold shape well-preserved. Specifically, we propose to project SPD matrices using a set of random projection hyperplanes over RKHS into a random projection space, which leads to representing each matrix as a vector of projection coefficients. Experiments on face recognition, person re-identification and texture classification show that the proposed approach outperforms several recent methods, such as Tensor Sparse Coding, Histogram Plus Epitome, Riemannian Locality Preserving Projection and Relational Divergence Classification.

Matching image sets via adaptive multi convex hull

Traditional nearest points methods use all the samples in an image set to construct a single convex or affine hull model for classification. However, strong artificial features and noisy data may be generated from combinations of training samples when significant intra-class variations and/or noise occur in the image set. Existing multi-model approaches extract local models by clustering each image set individually only once, with fixed clusters used for matching with various image sets. This may not be optimal for discrimination, as undesirable environmental conditions (eg. illumination and pose variations) may result in the two closest clusters representing different characteristics of an object (eg. frontal face being compared to non-frontal face). To address the above problem, we propose a novel approach to enhance nearest points based methods by integrating affine/convex hull classification with an adapted multi-model approach. We first extract multiple local convex hulls from a query image set via maximum margin clustering to diminish the artificial variations and constrain the noise in local convex hulls. We then propose adaptive reference clustering (ARC) to constrain the clustering of each gallery image set by forcing the clusters to have resemblance to the clusters in the query image set. By applying ARC, noisy clusters in the query set can be discarded. Experiments on Honda, MoBo and ETH-80 datasets show that the proposed method outperforms single model approaches and other recent techniques, such as Sparse Approximated Nearest Points, Mutual Subspace Method and Manifold Discriminant Analysis.

Dictionary learning and sparse coding on Grassmann manifolds : an extrinsic solution

Recent advances in computer vision and machine learning suggest that a wide range of problems can be addressed more appropriately by considering non-Euclidean geometry. In this paper we explore sparse dictionary learning over the space of linear subspaces, which form Riemannian structures known as Grassmann manifolds. To this end, we propose to embed Grassmann manifolds into the space of symmetric matrices by an isometric mapping, which enables us to devise a closed-form solution for updating a Grassmann dictionary, atom by atom. Furthermore, to handle non-linearity in data, we propose a kernelised version of the dictionary learning algorithm. Experiments on several classification tasks (face recognition, action recognition, dynamic texture classification) show that the proposed approach achieves considerable improvements in discrimination accuracy, in comparison to state-of-the-art methods such as kernelised Affine Hull Method and graph-embedding Grassmann discriminant analysis.

Physical activity recognition from accelerometer data using a multi-scale ensemble method

Accurate and detailed measurement of an individual's physical activity is a key requirement for helping researchers understand the relationship between physical activity and health. Accelerometers have become the method of choice for measuring physical activity due to their small size, low cost, convenience and their ability to provide objective information about physical activity. However, interpreting accelerometer data once it has been collected can be challenging. In this work, we applied machine learning algorithms to the task of physical activity recognition from triaxial accelerometer data. We employed a simple but effective approach of dividing the accelerometer data into short non-overlapping windows, converting each window into a feature vector, and treating each feature vector as an i.i.d training instance for a supervised learning algorithm. In addition, we improved on this simple approach with a multi-scale ensemble method that did not need to commit to a single window size and was able to leverage the fact that physical activities produced time series with repetitive patterns and discriminative features for physical activity occurred at different temporal scales.

Multi-shot person re-identification via relational stein divergence

Person re-identification is particularly challenging due to significant appearance changes across separate camera views. In order to re-identify people, a representative human signature should effectively handle differences in illumination, pose and camera parameters. While general appearance-based methods are modelled in Euclidean spaces, it has been argued that some applications in image and video analysis are better modelled via non-Euclidean manifold geometry. To this end, recent approaches represent images as covariance matrices, and interpret such matrices as points on Riemannian manifolds. As direct classification on such manifolds can be difficult, in this paper we propose to represent each manifold point as a vector of similarities to class representers, via a recently introduced form of Bregman matrix divergence known as the Stein divergence. This is followed by using a discriminative mapping of similarity vectors for final classification. The use of similarity vectors is in contrast to the traditional approach of embedding manifolds into tangent spaces, which can suffer from representing the manifold structure inaccurately. Comparative evaluations on benchmark ETHZ and iLIDS datasets for the person re-identification task show that the proposed approach obtains better performance than recent techniques such as Histogram Plus Epitome, Partial Least Squares, and Symmetry-Driven Accumulation of Local Features.