This paper addresses the challenge of reading low contrast text on tyre sidewall images of vehicles in motion. It presents first of its kind, a full scale industrial system which can read tyre codes when installed along driveways such as at gas stations or parking lots with vehicles driving under 10 mph. Tyre circularity is first detected using a circular Hough transform with dynamic radius detection. The detected tyre arches are then unwarped into rectangular patches. A cascade of convolutional neural network (CNN) classifiers is then applied for text recognition. Firstly, a novel proposal generator for the code localization is introduced by integrating convolutional layers producing HOG-like (Histogram of Oriented Gradients) features into a CNN. The proposals are then filtered using a deep network. After the code is localized, character detection and recognition are carried out using two separate deep CNNs. The results (accuracy, repeatability and efficiency) are impressive and show promise for the intended application.

Abstract Visual analytics systems combine machine learning or other analytic techniques with interactive data visualization to promote sensemaking and analytical reasoning. It is through such techniques that people can make sense of large, complex data. While progress has been made, the tactful combination of machine learning and data visualization is still under‐explored. This state‐of‐the‐art report presents a summary of the progress that has been made by highlighting and synthesizing select research advances. Further, it presents opportunities and challenges to enhance the synergy between machine learning and visual analytics for impactful future research directions.

We propose a novel template matching approach for the discrimination of handwritten and machine-printed text. We first pre-process the scanned document images by performing denoising, circles/lines exclusion and word-block level segmentation. We then align and match characters in a flexible sized gallery with the segmented regions, using parallelised normalised cross-correlation. The experimental results over the Pattern Recognition & Image Analysis Research Lab-Natural History Museum (PRImA-NHM) dataset show remarkably high robustness of the algorithm in classifying cluttered, occluded and noisy samples, in addition to those with significant high missing data. The algorithm, which gives 84.0% classification rate with false positive rate 0.16 over the dataset, does not require training samples and generates compelling results as opposed to the training-based approaches, which have used the same benchmark.

UNLABELLED: A protein's isoelectric point or pI corresponds to the solution pH at which its net surface charge is zero. Since the early days of solution biochemistry, the pI has been recorded and reported, and thus literature reports of pI abound. The Protein Isoelectric Point database (PIP-DB) has collected and collated these data to provide an increasingly comprehensive database for comparison and benchmarking purposes. A web application has been developed to warehouse this database and provide public access to this unique resource. PIP-DB is a web-enabled SQL database with an HTML GUI front-end. PIP-DB is fully searchable across a range of properties. AVAILABILITY AND IMPLEMENTATION: The PIP-DB database and documentation are available at http://www.pip-db.org.

Visualization of high-dimensional data has always been a challenging task. Here we discuss and propose variants of non-linear data projection methods (Generative Topographic Mapping (GTM) and GTM with simultaneous feature saliency (GTM-FS)) that are adapted to be effective on very high-dimensional data. The adaptations use log space values at certain steps of the Expectation Maximization (EM) algorithm and during the visualization process. We have tested the proposed algorithms by visualizing electrostatic potential data for Major Histocompatibility Complex (MHC) class-I proteins. The experiments show that the variation in the original version of GTM and GTM-FS worked successfully with data of more than 2000 dimensions and we compare the results with other linear/nonlinear projection methods: Principal Component Analysis (PCA), Neuroscale (NSC) and Gaussian Process Latent Variable Model (GPLVM).

This paper presents a forecasting technique for forward electricity/gas prices, one day ahead. This technique combines a Kalman filter (KF) and a generalised autoregressive conditional heteroschedasticity (GARCH) model (often used in financial forecasting). The GARCH model is used to compute next value of a time series. The KF updates parameters of the GARCH model when the new observation is available. This technique is applied to real data from the UK energy markets to evaluate its performance. The results show that the forecasting accuracy is improved significantly by using this hybrid model. The methodology can be also applied to forecasting market clearing prices and electricity/gas loads.

This paper presents a forecasting technique for forward energy prices, one day ahead. This technique combines a wavelet transform and forecasting models such as multi-layer perceptron, linear regression or GARCH. These techniques are applied to real data from the UK gas markets to evaluate their performance. The results show that the forecasting accuracy is improved significantly by using the wavelet transform. The methodology can be also applied to forecasting market clearing prices and electricity/gas loads.

Data visualization algorithms and feature selection techniques are both widely used in bioinformatics but as distinct analytical approaches. Until now there has been no method of measuring feature saliency while training a data visualization model. We derive a generative topographic mapping (GTM) based data visualization approach which estimates feature saliency simultaneously with the training of the visualization model. The approach not only provides a better projection by modeling irrelevant features with a separate noise model but also gives feature saliency values which help the user to assess the significance of each feature. We compare the quality of projection obtained using the new approach with the projections from traditional GTM and self-organizing maps (SOM) algorithms. The results obtained on a synthetic and a real-life chemoinformatics dataset demonstrate that the proposed approach successfully identifies feature significance and provides coherent (compact) projections

We introduce a flexible visual data mining framework which combines advanced projection algorithms from the machine learning domain and visual techniques developed in the information visualization domain. The advantage of such an interface is that the user is directly involved in the data mining process. We integrate principled projection algorithms, such as generative topographic mapping (GTM) and hierarchical GTM (HGTM), with powerful visual techniques, such as magnification factors, directional curvatures, parallel coordinates and billboarding, to provide a visual data mining framework. Results on a real-life chemoinformatics dataset using GTM are promising and have been analytically compared with the results from the traditional projection methods. It is also shown that the HGTM algorithm provides additional value for large datasets. The computational complexity of these algorithms is discussed to demonstrate their suitability for the visual data mining framework.

MOTIVATION: Clustering techniques such as k-means and hierarchical clustering are commonly used to analyze DNA microarray derived gene expression data. However, the interactions between processes underlying the cell activity suggest that the complexity of the microarray data structure may not be fully represented with discrete clustering methods. RESULTS: A newly developed software tool called MILVA (microarray latent visualization and analysis) is presented here to investigate microarray data without separating gene expression profiles into discrete classes. The underpinning of the MILVA software is the two-dimensional topographic representation of multidimensional microarray data. On this basis, the interactive MILVA functions allow a continuous exploration of microarray data driven by the direct supervision of the biologist in detecting activity patterns of co-regulated genes. AVAILABILITY: The MILVA software is freely available. The software and the related documentation can be downloaded from http://www.ncrg.aston.ac.uk/Projects/milva. User 'surrey' as username and '3245' as password to login. The software is currently available for Windows platform only.

Recently, we have developed the hierarchical generative topographic mapping (HGTM), an interactive method for visualization of large high-dimensional real-valued data sets. We propose a more general visualization system by extending HGTM in three ways, which allows the user to visualize a wider range of data sets and better support the model development process. 1) We integrate HGTM with noise models from the exponential family of distributions. The basic building block is the latent trait model (LTM). This enables us to visualize data of inherently discrete nature, e.g., collections of documents, in a hierarchical manner. 2) We give the user a choice of initializing the child plots of the current plot in either interactive, or automatic mode. In the interactive mode, the user selects "regions of interest", whereas in the automatic mode, an unsupervised minimum message length (MML)-inspired construction of a mixture of LTMs is employed. The unsupervised construction is particularly useful when high-level plots are covered with dense clusters of highly overlapping data projections, making it difficult to use the interactive mode. Such a situation often arises when visualizing large data sets. 3) We derive general formulas for magnification factors in latent trait models. Magnification factors are a useful tool to improve our understanding of the visualization plots, since they can highlight the boundaries between data clusters. We illustrate our approach on a toy example and evaluate it on three more complex real data sets.

Radial Basis Function networks with linear outputs are often used in regression problems because they can be substantially faster to train than Multi-layer Perceptrons. For classification problems, the use of linear outputs is less appropriate as the outputs are not guaranteed to represent probabilities. We show how RBFs with logistic and softmax outputs can be trained efficiently using the Fisher scoring algorithm. This approach can be used with any model which consists of a generalised linear output function applied to a model which is linear in its parameters. We compare this approach with standard non-linear optimisation algorithms on a number of datasets.

It has been argued that a single two-dimensional visualization plot may not be sufficient to capture all of the interesting aspects of complex data sets and, therefore, a hierarchical visualization system is desirable. In this paper, we extend an existing locally linear hierarchical visualization system PhiVis in several directions: 1) We allow for nonlinear projection manifolds. The basic building block is the Generative Topographic Mapping (GTM). 2) We introduce a general formulation of hierarchical probabilistic models consisting of local probabilistic models organized in a hierarchical tree. General training equations are derived, regardless of the position of the model in the tree. 3) Using tools from differential geometry, we derive expressions for local directional curvatures of the projection manifold. Like PhiVis, our system is statistically principled and is built interactively in a top-down fashion using the EM algorithm. It enables the user to interactively highlight those data in the ancestor visualization plots which are captured by a child model. We also incorporate into our system a hierarchical, locally selective representation of magnification factors and directional curvatures of the projection manifolds. Such information is important for further refinement of the hierarchical visualization plot, as well as for controlling the amount of regularization imposed on the local models. We demonstrate the principle of the approach on a toy data set and apply our system to two more complex 12- and 18-dimensional data sets.

Mixture Density Networks (MDNs) are a well-established method for modelling the conditional probability density which is useful for complex multi-valued functions where regression methods (such as MLPs) fail. In this paper we extend earlier research of a regularisation method for a special case of MDNs to the general case using evidence based regularisation and we show how the Hessian of the MDN error function can be evaluated using R-propagation. The method is tested on two data sets and compared with early stopping.

The deficiencies of stationary models applied to financial time series are well documented. A special form of non-stationarity, where the underlying generator switches between (approximately) stationary regimes, seems particularly appropriate for financial markets. We use a dynamic switching (modelled by a hidden Markov model) combined with a linear dynamical system in a hybrid switching state space model (SSSM) and discuss the practical details of training such models with a variational EM algorithm due to Z. Ghahramani and G.E. Hinton (1998). The performance of the SSSM is evaluated on several financial data sets and it is shown to improve on a number of existing benchmark methods.

Most conventional techniques for estimating conditional probability densities are inappropriate for applications involving periodic variables. In this paper we introduce three related techniques for tackling such problems, and investigate their performance using synthetic data. We then apply these techniques to the problem of extracting the distribution of wind vector directions from radar scatterometer data gathered by a remote-sensing satellite.