Practical multi-hypothesis trackers (MHTs) often entail a number of parameters for track confirmation and extraction logic, gating and pruning, most of which are chosen heuristically to tradeoff performance and computational cost. Conceptually, these parameters are unnecessary with optimal MHT processing, as these decisions will fall out from the optimal solution, though perhaps with an increase in processing cost. We demonstrate, however, with a canonical MHT model and its attendant association assignment problem that many of these parameters can be chosen losslessly, that is, they only remove hypotheses that an optimal association solution is guaranteed to remove anyway, and thus strictly improve computational cost, with no loss in tracking performance. At the other end of the spectrum, the tools developed likewise yield a number of relations that detect when parameters are set such that practical tracking is no longer possible.

There is an inherent flexibility in typesetting a block of text. Traditionally, line breaks would be manually chosen at strategic points in such a way as to minimize the amount of whitespace in each line. Hyphenation would only be used as a last resort. Knuth and Plass automated this optimization procedure, which has been used in various typesetting systems and DTP applications ever since. However, an optimal solution for the line-breaking problem does not necessarily lead us to an optimal document layout on the whole. The flexibility of choosing line breaks enables us, in many cases, to adjust the height of a paragraph by changing the number of lines, without having to make adjustments to font size, leading, etc. In many cases, the word spacing remains within the usual tolerances and visual quality does not noticeably suffer. This paper presents a modification to the Knuth-Plass algorithm to return several results for a given column of text, each corresponding to a different height, and describes steps to quantify the amount of expected flexibility in a given paragraph. We conclude with a discussion on how such "sub-optimal" results can lead to a better overall document layout, particularly in the context of mobile layouts, where flexibility is of key importance.

Retinal images contain forests of mutually intersecting and overlapping venous and arterial vascular trees. The geometry of these trees shows adaptation to vascular diseases including diabetes, stroke and hypertension. Segmentation of the retinal vascular network is complicated by inconsistent vessel contrast, fuzzy edges, variable image quality, media opacities, complex intersections and overlaps. This paper presents a Bayesian approach to resolving the configuration of vascular junctions to correctly construct the vascular trees. A probabilistic model of vascular joints (terminals, bridges and bifurcations) and their configuration in junctions is built, and Maximum A Posteriori (MAP) estimation used to select most likely configurations. The model is built using a reference set of 3010 joints extracted from the DRIVE public domain vascular segmentation dataset, and evaluated on 3435 joints from the DRIVE test set, demonstrating an accuracy of 95.2%.

This paper presents a system for automatically classifying the resting location of a moving object in an indoor environment. The system uses an unsupervised neural network (Self Organising Feature Map) fully implemented on a low-cost, low-power automated home-based surveillance system, capable of monitoring activity level of elders living alone independently. The proposed system runs on an embedded platform with a specialised ceiling-mounted video sensor for intelligent activity monitoring. The system has the ability to learn resting locations, to measure overall activity levels and to detect specific events such as potential falls. First order motion information, including first order moving average smoothing, is generated from the 2D image coordinates (trajectories). A novel edge-based object detection algorithm capable of running at a reasonable speed on the embedded platform has been developed. The classification is dynamic and achieved in real-time. The dynamic classifier is achieved using a SOFM and a probabilistic model. Experimental results show less than 20% classification error, showing the robustness of our approach over others in literature with minimal power consumption. The head location of the subject is also estimated by a novel approach capable of running on any resource limited platform with power constraints.

This paper aims to reconstruct retinal vessel trees from the broken vessel segments in fund us images for clinical studies and early diagnosis of systemic diseases including diabetic retinopathy, atherosclerosis, and hypertension. A Naive Bayes model is proposed for correct configurations of segments at retinal junctions including bifurcations, crossovers, overlaps, and mixture of these. The Maximum A Posteriori (MAP) is established to select the most likely configuration. In addition, the feature set consists of proportional associations of vessels width, angle and orientation. These theoretically proven associations are based on the optimality principles of minimum work in the vasculature for blood flow efficiency. We modelled the system using the training set of DRIVE database, tested on the testing set of same database, and produced 93.3% overall accuracy.

We present a novel algorithm for stylizing photographs into portrait paintings comprised of curved brush strokes. Rather than drawing upon a prescribed set of heuristics to place strokes, our system learns a flexible model of artistic style by analyzing training data from a human artist. Given a training pair — a source image and painting of that image—a non-parametric model of style is learned by observing the geometry and tone of brush strokes local to image features. A Markov Random Field (MRF) enforces spatial coherence of style parameters. Style models local to facial features are learned using a semantic segmentation of the input face image, driven by a combination of an Active Shape Model and Graph-cut. We evaluate style transfer between a variety of training and test images, demonstrating a wide gamut of learned brush and shading styles.

The classification of retinal vessels into arteries and veins is an important step for the analysis of retinal vascular trees, for which the scientists have proposed several classification methods. An obvious concern regarding the strength of these methodologies is the closeness of the result of a particular method to the gold standard. Unfortunately, the research community lacks benchmarks, resulting in increased subjective error, biased opinion and an uncertain progress. This paper introduces a manually-labeled, artery/vein categorized gold standard image database, as an extension of the most widely used image set DRIVE. The labeling criterion is set after a careful analysis of the physiological facts about the retinal vascular system. In addition, the labeling process also includes several versions of original images to get certainty. A two-step validation phase consists of verification from the trained computer vision observers and a professional ophthalmologist, followed by a comparison with a gold standard set for the junction locations introduced in V4-Like filters. Our gold standard is in highly reliable form; offers research community for the result comparison and progress evaluation.

This paper introduces a tri-state logic self-organizing map (bSOM) designed and implemented on a field programmable gate array (FPGA) chip. The bSOM takes binary inputs and maintains tri-state weights. A novel training rule is presented. The bSOM is well suited to FPGA implementation, trains quicker than the original self-organizing map (SOM), and can be used in clustering and classification problems with binary input data. Two practical applications, character recognition and appearance-based object identification, are used to illustrate the performance of the implementation. The appearance-based object identification forms part of an end-to-end surveillance system implemented wholly on FPGA. In both applications, binary signatures extracted from the objects are processed by the bSOM. The system performance is compared with a traditional SOM with real-valued weights and a strictly binary weighted SOM.

Methods of segmenting objects of interest from video data typically use a background model to represent an empty, static scene. However, dynamic processes in the background, such as moving foliage and water, can act to undermine the robustness of such methods and result in false positive object detections. Techniques for reducing errors have been proposed, including Markov Random Field (MRF) based pixel classification schemes, and also the use of region-based models. The work we present here combines these two approaches, using a region-based background model to provide robust likelihoods for multi-class MRF pixel labelling. Our initial results show the effectiveness of our method, by comparing performance with an analogous per-pixel likelihood model.

A binary Self Organizing Map (SOM) has been designed and implemented on a Field Programmable Gate Array (FPGA) chip. A novel learning algorithm which takes binary inputs and maintains tri-state weights is presented. The binary SOM has the capability of recognizing binary input sequences after training. A novel tri-state rule is used in updating the network weights during the training phase. The rule implementation is highly suited to the FPGA architecture, and allows extremely rapid training. This architecture may be used in real-time for fast pattern clustering and classification of binary features.

The sparse distributed memory (SDM) proposed by Kanerva provides a simple model for human long-term memory, with a strong underlying mathematical theory. However, there are problematic features in the original SDM model that affect its efficiency and performance in real world applications and for hardware implementation. In this paper, we propose modifications to the SDM model that improve its efficiency and performance in pattern recall. First, the address matrix is built using training samples rather than random binary sequences. This improves the recall performance significantly. Second, the content matrix is modified using a simple tri-state logic rule. This reduces the storage requirements of the SDM and simplifies the implementation logic, making it suitable for hardware implementation. The modified model has been tested using pattern recall experiments. It is found that the modified model can recall clean patterns very well from noisy inputs.

The lobula giant movement detector (LGMD) is a wide-field visual neuron that is located in the lobula layer of the locust nervous system. The LGMD increases its firing rate in response to both the velocity of the approaching object and its proximity. It has been found that it can respond to looming stimuli very quickly and can trigger avoidance reactions whenever a rapidly approaching object is detected. It has been successfully applied in visual collision avoidance systems for vehicles and robots. This paper proposes a modified LGMD model that provides additional movement depth direction information. The proposed model retains the simplicity of the previous neural network model, adding only a few new cells. It has been tested on both simulated and recorded video data sets. The experimental results shows that the modified model can very efficiently provide stable information on the depth direction of movement.

This paper presents an algorithm for segmenting and measuring retinal vessels, by growing a "Ribbon of Twins" active contour model, which uses two pairs of contours to capture each vessel edge, while maintaining width consistency. The algorithm is initialized using a generalized morphological order filter to identify approximate vessels centerlines. Once the vessel segments are identified the network topology is determined using an implicit neural cost function to resolve junction configurations. The algorithm is robust, and can accurately locate vessel edges under difficult conditions, including noisy blurred edges, closely parallel vessels, light reflex phenomena, and very fine vessels. It yields precise vessel width measurements, with subpixel average width errors. We compare the algorithm with several benchmarks from the literature, demonstrating higher segmentation sensitivity and more accurate width measurement.

A new method is introduced for joining vessel segments together to form a vessel graph. Using a reference image set from the Sunderland Eye Infirmary, we analysed the retinal bifurcation geometry, to define measurements for the geometrical junction features. These distinctive measurements are employed to resolve the junctions. Self organized feature maps (SOFM) are used to ldquolearnrdquo cost functions for forming bifurcation and bridge forms. The system joins segments depending on their ldquoprojective intersectionsrdquo and the SOFM cost functions. The system includes algorithms to handle overlapping and parallel segments. Transferring the vascular network to a vascular graph provides an opportunity to extract more information and to calculate features that have been not previously calculated, by providing new measurements from graph theory.

Many existing approaches in computer vision to pose estimation make simplifications of the measurement problem, either using silhouettes or assuming knowledge of appearance or color. However, recognizing the pose of a person who is persistently under cover remains challenging. We present a real time monocular-video approach for markerless pose estimation of human body under cover without manual initialization. In order to deal with heavy occlusion, we propose a model that reinforces both feature space and model parameters by adjacent parameters and a novel search framework that aggregates detections over time to produce a more reliable hypothesis. In addition, we have introduced a novel head model, which has the combined effect of improving performance and increasing efficiency. Furthermore, we have proposed a novel representation to estimate upper leg posture using latent features. In evaluation, we demonstrate the techniques to estimate the covered body pose with various postures and obscuration levels in two environmental settings.

Existing video monitoring techniques require clinicians to analyze substantial amounts of video data in diagnosis of sleep apnea. Analysis of the covered human body from video is a challenging task as traditional computer vision methods such as correlation, template matching, background subtraction, contour models and related techniques for object tracking become ineffective because of the large degree of occlusion for long periods. In condition of persistent heavy occlusion, difficulties arise from night vision, large variances of image features according to the occlusion level, the shifting of the cover surface with movements, obscuration of the bodiespsila edges by the cover, and wrinkle noises. We propose a near real time method to robustly estimate the pose of fully/partially covered or uncovered human body. The proposed method contains a novel weak human model to accommodate large variances of image features and a strong pose recognition model derived from a stylized pose detector used for people tracking by Ramanan et al.. We improve the stylized pose detection model by modifying the cost formula and template representation to overcome weak cues and strong noise due to heavy occlusion. In evaluation, the experimental results show that the proposed model is promising to estimate the pose of a human body with fully or partially covered or without covered.

This paper introduces a real-time connected component labelling algorithm designed for field programmable gate array (FPGA) implementation. The algorithm run-length encodes the image, and performs connected component analysis on this representation. The run-length encoding, together with other parts of the algorithm, is performed in parallel; sequential operations are minimized as the number of runs are typically less than the number of pixels. The architecture is designed mainly on Block RAM (i.e. internal RAM) of the FPGA. A comparison with the multi-pass algorithm in hardware and software is presented to show the advantages of the algorithm. The algorithm runs comfortably in real-time with reasonably low resource utilization, making integration with other real-time algorithms feasible.

We present an integrated pixel segmentation and region tracking algorithm, designed for indoor environments. Visual monitoring systems often use frame differencing techniques to independently classify each image pixel as either foreground or background. Typically, this level of processing does not take account of the global image structure, resulting in frequent misclassification. We use an adaptive Gaussian mixture model in colour and space to represent background and foreground regions of the scene. This model is used to probabilistically classify observed pixel values, incorporating the global scene structure into pixel-level segmentation. We evaluate our system over 4 sequences and show that it successfully segments foreground pixels and tracks major foreground regions as they move through the scene.

Reliable and efficient optic disk localization and segmentation are important tasks in automated retinal screening. General-purpose edge detection algorithms often fail to segment the optic disk due to fuzzy boundaries, inconsistent image contrast or missing edge features. This paper presents an algorithm for the localization and segmentation of the optic nerve head boundary in low-resolution images (about 20 microns/pixel). Optic disk localization is achieved using specialized template matching, and segmentation by a deformable contour model. The latter uses a global elliptical model and a local deformable model with variable edge-strength dependent stiffness. The algorithm is evaluated against a randomly selected database of 100 images from a diabetic screening programme. Ten images were classified as unusable; the others were of variable quality. The localization algorithm succeeded on all bar one usable image; the contour estimation algorithm was qualitatively assessed by an ophthalmologist as having Excellent-Fair performance in 83% of cases, and performs well even on blurred images.

Changes in retinal vessel diameter are an important sign of diseases such as hypertension, arteriosclerosis and diabetes mellitus. Obtaining precise measurements of vascular widths is a critical and demanding process in automated retinal image analysis as the typical vessel is only a few pixels wide. This paper presents an algorithm to measure the vessel diameter to subpixel accuracy. The diameter measurement is based on a two-dimensional difference of Gaussian model, which is optimized to fit a two-dimensional intensity vessel segment. The performance of the method is evaluated against Brinchmann-Hansen's half height, Gregson's rectangular profile and Zhou's Gaussian model. Results from 100 sample profiles show that the presented algorithm is over 30% more precise than the compared techniques and is accurate to a third of a pixel.

Understanding the intricacies behind concurrency within object-oriented programming languages has always been a challenge for undergraduate students. While the lecture is a relatively passive learning experience for the student, the use of software visualisation offers the chance to examine the concepts covered in the lecture in an interactive, visual environment. Students can add further dimensions and greater depth to their understanding previously hindered by the pedagogy of this passive environment. Elucidate makes use of the JDI architecture in the Java language to create its own environment that allows students to execute any program within it. Elucidate utilises several information workspaces, each presenting a different perspective about the information, thus facilitating a students ability to employ it in a manner that best allows them to construct their own understanding. Students are able to navigate around multiple views, and through various levels of abstraction, revealing the inner workings and sequence of events in what would otherwise be a black-box program.

This paper describes the multiple object tracking component of an automated CCTV surveillance system. The system tracks objects, and alerts the operator if unusual trajectories are discovered. Objects are detected by background differencing. Low contrast levels can present problems, leading to poor object segmentation and fragmentation, particularly on older analogue surveillance networks. The model-free tracking algorithm described in this paper addresses object fragmentation, and the object merging that occurs when proximate objects segment to the same connected component. 1

Feature selection is a common and key problem in many classification and regression tasks. It can be viewed as a multiobjective optimisation problem, since, in the simplest case, it involves feature subset size minimisation and performance maximisation. This paper presents a multiobjective evolutionary approach for feature selection. A novel commonality-based crossover operator is introduced and placed in the multiobjective evolutionary setting. This specialised operator helps to preserve building blocks with promising performance. Selection bias reduction is achieved by resampling. We argue that this is a generic approach, which can be used in many modelling problems. It is applied to feature selection on different neural network architectures. Results from experiments with benchmarking data sets are given.

The paper describes a training algorithm for multilayer perceptrons. It has scalable memory requirements, which may range from O(W) to O(W/sup 2/), although in practice the useful range is limited to lower complexity levels. The algorithm is based upon a novel iterative estimation of the principal eigensubspace of the Hessian, together with a quadratic step estimation procedure. It is shown that the new algorithm has convergence time comparable to conjugate gradient descent, and may be preferable if early stopping is used as it converges more quickly during the initial phases. Results of experiments to confirm the algorithm's performance are presented.

This paper discusses the use of multicriteria genetic algorithms for feature selection in classification problems. This feature selection approach is shown to yield a diverse population of alternative feature subsets with various accuracy/complexity trade-off. The algorithm is applied to select features for performing classification with fuzzy models, and is evaluated on two real-world data sets. We discuss when multicriteria genetic algorithm feature selection is preferable to a sequential feature selection procedure, namely backwards elimination. Among the key features of the presented approach are its computational simplicity, effectiveness on real world problems and the potential it has to become a powerful tool aiding many empirical modeling and data mining processes.

Many quad encoding technique have been published previously, using a variety of approaches to the different facets of quad storage, representation and manipulation. We review the major classes and identify three parameters which distinguish them. These are then used explicitly to classify a large number of published methods, with short discussion on each. In doing this we are also able to highlight some new approaches, to identify rare examples which escape the classification, and to suggest approaches to choosing a method to suit a given application.

We describe an implementation of a networked picture browser. The system offers a pictorial interface to pictorial data, relieves the users of thinking about the underlying filing system, provides managerial tools for installing, moving and deleting pictures, offers graded access and picture sharing and supports a number of picture formats, including hierarchical encodings.

Methods of labelling quadtree nodes by extended quaternary numbers are discussed. It is shown that a fixed bit-length technique requires only 2n+1 bits to represent any node in a depth n quadtree. A technique using exactly that number of bits is described. A more useful extension, requiring 2n+2 bits, is compared with the best technique previously published, which typically requires 2n+4 bits.