Real-world Sequences Research Articles

Weighted t-way Sequences

We define the notion of weighted t-way sequences, which is built upon sequence covering arrays. The integration of a weight-based modelling formalism together with partitions of positive integers increases the expressiveness of the generated sequences considerably, and makes them applicable as abstract test sequences for real-world sequence testing problems. Applicability of this concept to real-world testing scenarios is investigated.

Linear programming based time lag identification in event sequences

Many technical systems like manufacturing plants or software applications generate large event sequences. Knowing the temporal relationship between events is important for gaining insights into the status and behavior of the system. This paper proposes a novel approach for identifying the time lag between different event types. This identification task is formulated as a binary integer optimization problem that can be solved efficiently and close to optimality by means of a linear programming approximation. The performance of the proposed approach is demonstrated on synthetic and real-world event sequences.

Using Kalman filter in the frequency domain for multi-frame scalable super resolution

Abstract Kalman filter (KF) as a linear estimator which is used in super-resolution (SR) problems, suffers from high computational costs and storage requirements. To gain appreciable success in the elimination of these two challenges, this paper advances a SR framework employing KF in the frequency domain, while no resort is made to any approximations or extra assumptions in the dynamic system modeling and statistical matrices. Generally, previous KF-based SR methods organized the system with huge-sized matrices in the spatial domain, following which they tried to reduce the system dimension using approximation and/or limitation on point spread function (PSF). In this study, first, several small-dimension dynamic systems are separately made in the frequency domain supporting space-invariant PSFs of an arbitrary form and size. Then, the acquired small-dimension KF estimators are applied rather than the traditional huge-dimension one. These will greatly reduce computational complexity, decrease storage requirements allowing parallel implementation as well. Furthermore, our proposed SR framework can be used to produce high resolution image of an expedient size, that is, a scalable SR. Experimental results with both simulated and real world sequences indicate that our proposed framework works more effectively than the other compared methods, especially in fine details restoration.

Two-stage pruning method for gram-based categorical sequence clustering

Gram-based vector space model has been extensively applied to categorical sequence clustering. However, there is a general lack of an efficient method to determine the length of grams and to identify redundant and non-significant grams involved in the model. In this paper, a variable-length gram model is proposed, different from previous studies mainly focused on the fixed-length grams of sequences. The variable-length grams are obtained using a two-stage pruning method aimed at selecting the irredundant and significant subsequences from the prefix trees, created from the fixed-length grams with an initially large length. A robust partitioning algorithm is then defined for categorical sequence clustering on the normalized representation model using variable-length grams collected from the pruned trees. Experimental results on real-world sequence sets from various domains are given to demonstrate the performance of the proposed methods.

Treat ER+ight: Canadian prospective observational study in post-menopausal HR+HER2- advanced breast cancer women—First interim analysis (IA).

1067 Background: Treat ER+ight is the first Canadian real-world study enrolling patients (pts) previously exposed to NSAI therapy and currently receiving endocrine therapy (ET) alone or in combination with targeted therapy (ET+TT). Methods: This first planned IA describes baseline parameters and adverse event (AE) prevention strategies adopted by 16 centers since Feb 2016 upon enrolling pts initiating ET or ET+TT. Results: See Table.Conclusions: This IA suggests that pts initiating ET+TT are younger, have a better ECOG status, mostly visceral disease and receive more prophylactic/proactive AE prevention therapies. Insights into real-world dosing and sequence will be presented. Clinical trial information: NCT02753686. [Table: see text]

BetaBit: A fast generator of autocorrelated binary processes for geophysical research

We introduce a fast and efficient non-iterative algorithm, called BetaBit, to simulate autocorrelated binary processes describing the occurrence of natural hazards, system failures, and other physical and geophysical phenomena characterized by persistence, temporal clustering, and low rate of occurrence. BetaBit overcomes the simulation constraints posed by the discrete nature of the marginal distributions of binary processes by using the link existing between the correlation coefficients of this process and those of the standard Gaussian processes. The performance of BetaBit is tested on binary signals with power-law and exponentially decaying autocorrelation functions (ACFs) corresponding to Hurst-Kolmogorov and Markov processes, respectively. An application to real-world sequences describing rainfall intermittency and the occurrence of strong positive phases of the North Atlantic Oscillation (NAO) index shows that BetaBit can also simulate surrogate data preserving the empirical ACF as well as signals with autoregressive moving average (ARMA) dependence structures. Extensions to cyclo-stationary processes accounting for seasonal fluctuations are also discussed.

Continuous Online Sequence Learning with an Unsupervised Neural Network Model.

The ability to recognize and predict temporal sequences of sensory inputs is vital for survival in natural environments. Based on many known properties of cortical neurons, hierarchical temporal memory (HTM) sequence memory recently has been proposed as a theoretical framework for sequence learning in the cortex. In this letter, we analyze properties of HTM sequence memory and apply it to sequence learning and prediction problems with streaming data. We show the model is able to continuously learn a large number of variable order temporal sequences using an unsupervised Hebbian-like learning rule. The sparse temporal codes formed by the model can robustly handle branching temporal sequences by maintaining multiple predictions until there is sufficient disambiguating evidence. We compare the HTM sequence memory with other sequence learning algorithms, including statistical methods-autoregressive integrated moving average; feedforward neural networks-time delay neural network and online sequential extreme learning machine; and recurrent neural networks-long short-term memory and echo-state networks on sequence prediction problems with both artificial and real-world data. The HTM model achieves comparable accuracy to other state-of-the-art algorithms. The model also exhibits properties that are critical for sequence learning, including continuous online learning, the ability to handle multiple predictions and branching sequences with high-order statistics, robustness to sensor noise and fault tolerance, and good performance without task-specific hyperparameter tuning. Therefore, the HTM sequence memory not only advances our understanding of how the brain may solve the sequence learning problem but is also applicable to real-world sequence learning problems from continuous data streams.

Blur robust optical flow using motion channel

It is hard to estimate optical flow given a realworld video sequence with camera shake and other motion blur. In this paper, we first investigate the blur parameterisation for video footage using near linear motion elements. We then combine a commercial 3D pose sensor with an RGB camera, in order to film video footage of interest together with the camera motion. We illustrate that this additional camera motion/trajectory channel can be embedded into a hybrid framework by interleaving an iterative blind deconvolution and warping based optical flow scheme. Our method yields improved accuracy within three other state-of-the-art baselines given our proposed ground truth blurry sequences; and several other realworld sequences filmed by our imaging system.

Cluster Validation Method for Determining the Number of Clusters in Categorical Sequences.

Cluster validation, which is the process of evaluating the quality of clustering results, plays an important role for practical machine learning systems. Categorical sequences, such as biological sequences in computational biology, have become common in real-world applications. Different from previous studies, which mainly focused on attribute-value data, in this paper, we work on the cluster validation problem for categorical sequences. The evaluation of sequences clustering is currently difficult due to the lack of an internal validation criterion defined with regard to the structural features hidden in sequences. To solve this problem, in this paper, a novel cluster validity index (CVI) is proposed as a function of clustering, with the intracluster structural compactness and intercluster structural separation linearly combined to measure the quality of sequence clusters. A partition-based algorithm for robust clustering of categorical sequences is also proposed, which provides the new measure with high-quality clustering results by the deterministic initialization and the elimination of noise clusters using an information theoretic method. The new clustering algorithm and the CVI are then assembled within the common model selection procedure to determine the number of clusters in categorical sequence sets. A case study on commonly used protein sequences and the experimental results on some real-world sequence sets from different domains are given to demonstrate the performance of the proposed method.

Multi-view and multi-plane data fusion for effective pedestrian detection in intelligent visual surveillance

For the robust detection of pedestrians in intelligent video surveillance, an approach to multi-view and multi-plane data fusion is proposed. Through the estimated homography, foreground regions are projected from multiple camera views to a reference view. To identify false-positive detections caused by foreground intersections of non-corresponding objects, the homographic transformations for a set of parallel planes, which are from the head plane to the ground, are applied. Multiple features including occupancy information and colour cues are extracted from such planes for joint decision-making. Experimental results on real world sequences have demonstrated the good performance of the proposed approach in pedestrian detection for intelligent visual surveillance.

Digital image stabilization using similarity transformation over constrained Differential-Radon warping vectors

In this paper an image-stabilizer based on novel Differential-Radon warping for combined camera motions, i.e., translation, rotation and zoom estimation has been presented. Motion estimation using intensity integrals in classical Radon transform (RT) does not consider projection shape for its correlation and leads to erroneous results under local intensity variations. As a solution, Differential-Radon (DRadon) matrix utilizing inherent curve variance as matching feature is proposed. The scheme provides peaked correlation curves, which gives better motion accuracy in comparison to the classical RT. DRadon projections of the test and target images computed at an intermediate angular-slice resolution of 0.1 degree are correlated to find relative image tilt, and the best matched projections are warped using Sakoe–Chiba constrained dynamic time warping. The estimated warping vectors are then processed using a new local-optima based interest vectors selection scheme to remove various one-to-many-matched warping singularities. A linear-motion model is applied over the extracted interest warping vectors and their least-square solution is used as inter-frame translation and zoom motions. Finally, a composite affine motion stabilizer framework is presented for shaky videos taken under different scene-capture conditions. Comparative stability performance evaluation using peak signal-to-noise ratio and mean structural similarity index is presented for various real-world sequences.

Noise robust image matching using adjacent evaluation census transform and wavelet edge joint bilateral filter in stereo vision

Automation application systems based on stereo vision require robust image matching methods to achieve available depth image information. This paper presents a novel noise robust stereo matching using adjacent evaluation census transform and wavelet edge joint bilateral filter. The adjacent evaluation census is firstly proposed to improve the robustness against noise of the census transform. Meanwhile, two different and complementary types of metrics are extracted (the adjacent evaluation census mean and the adjacent evaluation census weighted difference). Moreover, the weighted template is composed of four different directions. Then, to improve the robustness of cost aggregation and disparity optimization, the random walk is integrated into the proposed stereo matching method. Additionally, a disparity map post-processing method named wavelet edge joint bilateral filter is employed to eliminate error regions. An obtained wavelet-based edge image is considered as an important weighted coefficient to guide the post-processing. Experimental results demonstrate that the proposed method presents the best performance of the robustness against noise on the Middlebury dataset. Even in the toughest situation with additive Gaussian noise, our method can still achieve the moderate disparity map. In addition, the wider applicability of the proposed method is demonstrated on the KITTI (i.e., Karlsruhe Institute of Technology (KIT) and Toyota Technological Institute at Chicago (TTI-C)) dataset and some typical real-world sequences.

Driver head pose estimation using efficient descriptor fusion

A great interest is focused on driver assistance systems using the head pose as an indicator of the visual focus of attention and the mental state. In fact, the head pose estimation is a technique allowing to deduce head orientation relatively to a view of camera and could be performed by model-based or appearance-based approaches. Model-based approaches use a face geometrical model usually obtained from facial features, whereas appearance-based techniques use the whole face image characterized by a descriptor and generally consider the pose estimation as a classification problem. Appearance-based methods are faster and more adapted to discrete pose estimation. However, their performance depends strongly on the head descriptor, which should be well chosen in order to reduce the information about identity and lighting contained in the face appearance. In this paper, we propose an appearance-based discrete head pose estimation aiming to determine the driver attention level from monocular visible spectrum images, even if the facial features are not visible. Explicitly, we first propose a novel descriptor resulting from the fusion of four most relevant orientation-based head descriptors, namely the steerable filters, the histogram of oriented gradients (HOG), the Haar features, and an adapted version of speeded up robust feature (SURF) descriptor. Second, in order to derive a compact, relevant, and consistent subset of descriptor’s features, a comparative study is conducted on some well-known feature selection algorithms. Finally, the obtained subset is subject to the classification process, performed by the support vector machine (SVM), to learn head pose variations. As we show in experiments with the public database (Pointing’04) as well as with our real-world sequence, our approach describes the head with a high accuracy and provides robust estimation of the head pose, compared to state-of-the-art methods.

Algebraic and spectral analysis of local magnetic field intensity

One of the kind in Europe magnetometer is situated in central Lithuania and registers local magnetic field intensity. The sensitivity of the device is of pT (pico teslas) order. Magnetic field intensity is a real world sequence and as a result does not have a rank. Nevertheless, H-rank based techniques enable to analyse the data as algebraic sequences. The work also pays a considerable amount of attention to the spectral distribution of the signal. A specialized software is being developed due to the amount of the data and computational time costs for problems investigated.

A cost-effectiveness analysis of real-world treatment for elderly patients with multiple myeloma using a full disease model.

To study the impact of novel treatments for elderly (≥66 yr) patients with multiple myeloma (MM) in daily practice by comparing real-world effects [overall survival (OS) and quality-adjusted life years (QALYs)] and costs over time. Also, we calculate cost-effectiveness of treatment sequences commonly prescribed to predict effects and costs if patients had received a different treatment sequence. Real-world data including patient and disease characteristics, treatment information and resource use were collected from 1054 elderly patients with MM. Patients received first-line treatment during 2004-2007 (cohort 1) and 2008-2013 (cohort 2). The two cohorts were compared using a patient-level simulation (PLS) model comprising regression models which used patient and disease characteristics to estimate time to next treatment and death. Effects and costs from cohort 2 were compared to 4 commonly prescribed real-world sequences. Utilisation of novel agents was higher for cohort 2 compared to cohort 1. Modelled average OS for cohort 1 was 38 months (median 25) and total costs €44,200. OS for cohort 2 was 42 months (median 28) and total costs €69,017. The model identified potential OS gains if all patients were to be treated using combinations containing thalidomide, lenalidomide and bortezomib in that particular order. This sequence had, compared to real-world treatment, the most favourable incremental cost-effectiveness ratio, €24,618 per life year gained and €34,875 per QALY. Our patient-level model enabled to study the effects and costs of entire treatment sequences and to compare real-world treatment patterns over time. Increased utilisation of novel agents improved survival and increased costs for real-world patients with MM in the Netherlands.

XCS-SL: a rule-based genetic learning system for sequence labeling

Sequence labeling is an interesting classification domain where, like normal classification, every input has a class label, but unlike normal classification, prediction of an input’s label may depend on the values of other inputs or their classes, and so a learner may need to refer to inputs and classes at different time stamps to classify the current input. This is more difficult because a learner does not know where and how many inputs are needed to classify the current input. Our interest is in learning general rules for sequence labeling. The XCS algorithm is a rule-based knowledge discovery system powered by a genetic algorithm which has often been used for classification. Here we present XCS-SL, an extension of XCS classifier system which can be applicable to sequence labeling. Towards an application of Learning Classifier System (LCS) to sequence labeling, we propose a new classifier condition with memory (called a variable-length condition) and a rule-discovery system for the new classifier condition, which enables XCS to apply it to sequence labeling. In XCS-SL, classification rules (called “classifiers” here) can include extra conditions on previous inputs, which act as memories. In sequence labeling, the number of conditions/memories needed may be different for each input, hence, using a fixed number of conditions (i.e., fixed-length condition) for all classifiers is not a good solution. Instead, XCS-SL classifiers have a variable-length condition to provide more or less memory. The genetic algorithm can grow and shrink conditions to find a suitable memory size. On two synthetic benchmark problems XCS-SL learns optimal classifiers, and on a real-world sequence labeling task it derives high classification accuracy and discovers interesting knowledge that shows dependencies between inputs at different times. The comprehensively described system is the first application of a LCS to sequence labeling and we consider it a promising direction for future work.

Robust stereo matching using adaptive random walk with restart algorithm

In this paper, we propose a robust dense stereo reconstruction algorithm using a random walk with restart. The pixel-wise matching costs are aggregated into superpixels and the modified random walk with restart algorithm updates the matching cost for all possible disparities between the superpixels. In comparison to the majority of existing stereo methods using the graph cut, belief propagation, or semi-global matching, our proposed method computes the final reconstruction through the determination of the best disparity at each pixel in the matching cost update. In addition, our method also considers occlusion and depth discontinuities through the visibility and fidelity terms. These terms assist in the cost update procedure in the calculation of the standard smoothness constraint. The method results in minimal computational costs while achieving high accuracy in the reconstruction. We test our method on standard benchmark datasets and challenging real-world sequences. We also show that the processing time increases linearly in relation to an increase in the disparity search range.

Real-Time Pose Detection and Tracking of Hundreds of Objects

We propose a novel model-based method for tracking the 6-DOF pose of a very large number of rigid objects in real time. By combining dense motion and depth cues with sparse keypoint correspondences, and by feeding back information from the modeled scene to the cue extraction process, the method is both highly accurate and robust to noise and occlusions. A tight integration of the graphical and computational capability of graphics processing units allows the method to simultaneously track hundreds of objects in real time. We achieve pose updates at framerates around 40 Hz when using 500 000 data samples to track 150 objects using images of resolution $640\times 480$ . We introduce a synthetic benchmark data set with varying objects, background motion, noise, and occlusions that enables the evaluation of stereo-vision-based pose estimators in complex scenarios. Using this data set and a novel evaluation methodology, we show that the proposed method greatly outperforms state-of-the-art methods. Finally, we demonstrate excellent performance on challenging real-world sequences involving multiple objects being manipulated.

Gene Tree Affects Inference of Sites Under Selection by the Branch-Site Test of Positive Selection.

The branch-site test of positive selection is a standard approach to detect past episodic positive selection in a priori-specified branches of a gene phylogeny. Here, we ask if differences in the topology of the gene tree have any influence on the ability to infer positively selected sites. Using simulated sequences, we compare the results obtained for true and rearranged topologies. We find a strong relationship between “conflicting branch length,” which occurs when the set of sequences that experiences selection for a given topology and foreground is changed, and the ability to predict positively selected sites. Moreover, by reanalyzing a previously published data set, we show that the choice of a gene tree also affects the results obtained for real-world sequences. This is the first study to demonstrate that tree topology has a clear effect on the inference of positive selection. We conclude that the choice of a gene tree is an important factor for the branch-site analysis of positive selection.

Local Stereo Matching with Improved Matching Cost and Disparity Refinement

Recent local stereo matching methods have achieved comparable performance with global methods. However, the final disparity map still contains significant outliers. In this article, the authors propose a local stereo matching method that employs a new combined cost approach and a secondary disparity refinement mechanism. They formulate combined cost using a modified color census transform and truncated absolute differences of color and gradients. They also use symmetric guided filter for cost aggregation. Unlike in traditional stereo matching, they propose a novel secondary disparity refinement to further remove the remaining outliers. Experimental results on the Middlebury benchmark show that their method ranks fifth out of 153 submitted methods, and it's the best cost-volume filtering-based local method. Experiments on real-world sequences and depth-based applications also validate the proposed method's effectiveness.