Wavelet-based dynamic time warping

Dynamic Time Warping (DTW), a pattern matching technique traditionally used for restricted vocabulary speech recognition, is based on a temporal alignment of the input signal with the template models. The principal drawback of DTW is its high computational cost as the lengths of the signals increase. This paper shows extended results over our previously published conference paper, which introduces an optimized version of the DTW I hat is based on the Discrete Wavelet Transform (DWT). (C) 2008 Elsevier B.V. All rights reserved.

Threshold dynamic time warping for spatial activity recognition

Non-invasive spatial activity recognition is a difficult task, complicated by variation in how the same activities are conducted and furthermore by noise introduced by video tracking procedures. In this paper we propose an algorithm based on dynamic time warping (DTW) as a viable method with which to quantify segmented spatial activity sequences from a video tracking system. DTW is a widely used technique for optimally aligning or warping temporal sequences through minimisation of the distance between their components. The proposed algorithm threshold DTW (TDTW) is capable of accurate spatial sequence distance quantification and is shown using a three class spatial data set to be more robust and accurate than DTW and the discrete hidden markov model (HMM). We also evaluate the application of a band dynamic programming (DP) constraint to TDTW in order to reduce extraneous warping between sequences and to reduce the computation complexity of the approach. Results show that application of a band DP constraint to TDTW improves runtime performance significantly, whilst still maintaining a high precision and recall.

Online context recognition in multisensor systems using Dynamic Time Warping

In this paper, we present our system for online context recognition of multimodal sequences acquired from multiple sensors. The system uses Dynamic Time Warping (DTW) to recognize multimodal sequences of different lengths, embedded in continuous data streams. We evaluate the performance of our system on two real world datasets: 1) accelerometer data acquired from performing two hand gestures and 2) NOKIA's benchmark dataset for context recognition. The results from both datasets demonstrate that the system can perform online context recognition efficiently and achieve high recognition accuracy.

Temporal data fusion in multisensor systems using dynamic time warping

Data acquired from multiple sensors can be fused at a variety of levels: the raw data level, the feature level, or the decision level. An additional dimension to the fusion process is temporal fusion, which is fusion of data or information acquired from multiple sensors of different types over a period of time. We propose a technique that can perform such temporal fusion. The core of the system is the fusion processor that uses Dynamic Time Warping (DTW) to perform temporal fusion. We evaluate the performance of the fusion system on two real world datasets: 1) accelerometer data acquired from performing two hand gestures and 2) NOKIA’s benchmark dataset for context recognition. The results of the first experiment show that the system can perform temporal fusion on both raw data and features derived from the raw data. The system can also recognize the same class of multisensor temporal sequences even though they have different lengths e.g. the same human gestures can be performed at different speeds. In addition, the fusion processor can infer decisions from the temporal sequences fast and accurately. The results of the second experiment show that the system can perform fusion on temporal sequences that have large dimensions and are a mix of discrete and continuous variables. The proposed fusion system achieved good classification rates efficiently in both experiments

Changes Measurement in Biological Signals Based on Dynamic Time Warping Procedure on the Example of Repolarization Part of HeartbeatCycle

Master thesis discusses the analysis of changes in biological signals on time based on dynamic time warping algorithm (DTW). Special attention is paid to problems of tiny changes analysis incomplex nonstationary biological signals. Electrocardiographic (ECG) signals are used as an example inthis study; in particular, repolarization segments of heart beat cycles. The aim of the research is studyingthe possibility of applying DTW algorithm for the analysis of small changes in the repolarization segments of heart beat cycles. The research has the following tasks:- Studying repolarization segments of heart beat cycles, andmethods of their analysis;- Studying DTW algorithm and its modifications, finding the most appropriate modification for analyzing changes in biological signals;- Development of methods for analyzing the warping path(output parameter of DTW algorithm).

Similarity Measurement of Biological Signals Using Dynamic Time Warping Algorithm

The problem of similarity measurement of biological signals is considered on this article. The dynamic time warping algorithm is used as a possible solution. A short overview of this algorithm and its modifications are given. Testing procedure for different modifications of DTW, which are based on artificial test signals, are presented.

Multi Pattern Dynamic Time Warping For Automatic Speech Recognition

We are addressing the problem of jointly using multiple noisy speech patterns for automatic speech recognition (ASR), given that they come from the same class. If the user utters a word K times, the ASR system should try to use the information content in all the K patterns of the word simultaneously and improve its speech recognition accuracy compared to that of the single pattern based speech recognition. T address this problem, recently we proposed a Multi Pattern Dynamic Time Warping (MPDTW) algorithm to align the K patterns by finding the least distortion path between them. A Constrained Multi Pattern Viterbi algorithm was used on this aligned path for isolated word recognition (IWR). In this paper, we explore the possibility of using only the MPDTW algorithm for IWR. We also study the properties of the MPDTW algorithm. We show that using only 2 noisy test patterns (10 percent burst noise at -5 dB SNR) reduces the noisy speech recognition error rate by 37.66 percent when compared to the single pattern recognition using the Dynamic Time Warping algorithm.

Autenticacão de Assinaturas Online: Estudo dos Parâmetros do Dynamic Time Warping e da Representação da Assinatura

O reconhecimento de padões é uma área da inteligência computacional que apoia a resolução de problemas utilizando ferramentas computacionais. Dentre esses problemas podem ser citados o reconhecimento de faces, a identificação de impressões digitais e a autenticação de assinaturas. A autenticação de assinaturas de forma automática tem sua relevância pois está ligada ao reconhecimento de indivíduos e suas credenciais em sistemas complexos e a questões financeiras. Neste trabalho é apresentado um estudo dos parâmetros do Dynamic Time Warping, um algoritmo utilizado para alinhar duas assinaturas e medir a similaridade existente entre elas. Variando-se os principais parâmetros desse algoritmo, sobre uma faixa ampla de valores, foram obtidas as médias dos resultados de erros na classificação, e assim, estas médias foram avaliadas. Com base nas primeiras avaliação, foi identificada a necessidade de se calcular um desses parâmetros de forma dinâmica, o gap cost, a fim de ajustá-lo no uso de uma aplicação prática. Uma proposta para a realização deste cálculo é apresentada e também avaliada. É também proposta e avaliada uma maneira alternativa de representação dos atributos da assinatura, de forma a considerar sua curvatura em cada ponto adquirido no processo de aquisição, utilizando os vetores normais como forma de representação. As avaliações realizadas durante as diversas etapas do estudo consideraram o Equal Error Rate (EER) como indicação de qualidade e as técnicas propostas foram comparadas com técnicas já estabelecidas, obtendo uma média percentual de EER de 3,47%.

A grid-based index method for time warping distance

Recently DTW (dynamic time warping) has been recognized as the most robust distance function to measure the similarity between two time series, and this fact has spawned a flurry of research on this topic. Most indexing methods proposed for DTW are based on the R-tree structure. Because of high dimensionality and loose lower bounds for time warping distance, the pruning power of these tree structures are quite weak, resulting in inefficient search. In this paper, we propose a dimensionality reduction method motivated by observations about the inherent character of each time series. A very compact index file is constructed. By scanning the index file, we can get a very small candidate set, so that the number of page access is dramatically reduced. We demonstrate the effectiveness of our approach on real and synthetic datasets.

A dynamic time warped clustering technique for discrete event simulation-based system analysis

Abstract This paper introduces a novel approach for discrete event simulation output analysis. The approach combines dynamic time warping and clustering to enable the identification of system behaviours contributing to overall system performance, by linking the clustering cases to specific causal events within the system. Simulation model event logs have been analysed to group entity flows based on the path taken and travel time through the system. The proposed approach is investigated for a discrete event simulation of an international airport baggage handling system. Results show that the method is able to automatically identify key factors that influence the overall dwell time of system entities, such as bags that fail primary screening. The novel analysis methodology provides insight into system performance, beyond that achievable through traditional analysis techniques. This technique also has potential application to agent-based modelling paradigms and also business event logs traditionally studied using process mining techniques.

Efficient subsequence alignment of time series

Zeitreihen sind allgegenwärtig. Die Erfassung und Verarbeitung kontinuierlich gemessener Daten ist in allen Bereichen der Naturwissenschaften, Medizin und Finanzwelt vertreten. Das enorme Anwachsen aufgezeichneter Datenmengen, sei es durch automatisierte Monitoring-Systeme oder integrierte Sensoren, bedarf außerordentlich schneller Algorithmen in Theorie und Praxis. Infolgedessen beschäftigt sich diese Arbeit mit der effizienten Berechnung von Teilsequenzalignments. Komplexe Algorithmen wie z.B. Anomaliedetektion, Motivfabfrage oder die unüberwachte Extraktion von prototypischen Bausteinen in Zeitreihen machen exzessiven Gebrauch von diesen Alignments. Darin begründet sich der Bedarf nach schnellen Implementierungen. Diese Arbeit untergliedert sich in drei Ansätze, die sich dieser Herausforderung widmen. Das umfasst vier Alignierungsalgorithmen und ihre Parallelisierung auf CUDA-fähiger Hardware, einen Algorithmus zur Segmentierung von Datenströmen und eine einheitliche Behandlung von Liegruppen-wertigen Zeitreihen.rnrnDer erste Beitrag ist eine vollständige CUDA-Portierung der UCR-Suite, die weltführende Implementierung von Teilsequenzalignierung. Das umfasst ein neues Berechnungsschema zur Ermittlung lokaler Alignierungsgüten unter Verwendung z-normierten euklidischen Abstands, welches auf jeder parallelen Hardware mit Unterstützung für schnelle Fouriertransformation einsetzbar ist. Des Weiteren geben wir eine SIMT-verträgliche Umsetzung der Lower-Bound-Kaskade der UCR-Suite zur effizienten Berechnung lokaler Alignierungsgüten unter Dynamic Time Warping an. Beide CUDA-Implementierungen ermöglichen eine um ein bis zwei Größenordnungen schnellere Berechnung als etablierte Methoden.rnrnAls zweites untersuchen wir zwei Linearzeit-Approximierungen für das elastische Alignment von Teilsequenzen. Auf der einen Seite behandeln wir ein SIMT-verträgliches Relaxierungschema für Greedy DTW und seine effiziente CUDA-Parallelisierung. Auf der anderen Seite führen wir ein neues lokales Abstandsmaß ein, den Gliding Elastic Match (GEM), welches mit der gleichen asymptotischen Zeitkomplexität wie Greedy DTW berechnet werden kann, jedoch eine vollständige Relaxierung der Penalty-Matrix bietet. Weitere Verbesserungen umfassen Invarianz gegen Trends auf der Messachse und uniforme Skalierung auf der Zeitachse. Des Weiteren wird eine Erweiterung von GEM zur Multi-Shape-Segmentierung diskutiert und auf Bewegungsdaten evaluiert. Beide CUDA-Parallelisierung verzeichnen Laufzeitverbesserungen um bis zu zwei Größenordnungen.rnrnDie Behandlung von Zeitreihen beschränkt sich in der Literatur in der Regel auf reellwertige Messdaten. Der dritte Beitrag umfasst eine einheitliche Methode zur Behandlung von Liegruppen-wertigen Zeitreihen. Darauf aufbauend werden Distanzmaße auf der Rotationsgruppe SO(3) und auf der euklidischen Gruppe SE(3) behandelt. Des Weiteren werden speichereffiziente Darstellungen und gruppenkompatible Erweiterungen elastischer Maße diskutiert.

Riconoscimento real-time di gesture tramite tecniche di machine learning

Il riconoscimento delle gesture è un tema di ricerca che sta acquisendo sempre più popolarità, specialmente negli ultimi anni, grazie ai progressi tecnologici dei dispositivi embedded e dei sensori. Lo scopo di questa tesi è quello di utilizzare alcune tecniche di machine learning per realizzare un sistema in grado di riconoscere e classificare in tempo reale i gesti delle mani, a partire dai segnali mioelettrici (EMG) prodotti dai muscoli. Inoltre, per consentire il riconoscimento di movimenti spaziali complessi, verranno elaborati anche segnali di tipo inerziale, provenienti da una Inertial Measurement Unit (IMU) provvista di accelerometro, giroscopio e magnetometro. La prima parte della tesi, oltre ad offrire una panoramica sui dispositivi wearable e sui sensori, si occuperà di analizzare alcune tecniche per la classificazione di sequenze temporali, evidenziandone vantaggi e svantaggi. In particolare, verranno considerati approcci basati su Dynamic Time Warping (DTW), Hidden Markov Models (HMM), e reti neurali ricorrenti (RNN) di tipo Long Short-Term Memory (LSTM), che rappresentano una delle ultime evoluzioni nel campo del deep learning. La seconda parte, invece, riguarderà il progetto vero e proprio. Verrà impiegato il dispositivo wearable Myo di Thalmic Labs come caso di studio, e saranno applicate nel dettaglio le tecniche basate su DTW e HMM per progettare e realizzare un framework in grado di eseguire il riconoscimento real-time di gesture. Il capitolo finale mostrerà i risultati ottenuti (fornendo anche un confronto tra le tecniche analizzate), sia per la classificazione di gesture isolate che per il riconoscimento in tempo reale.

Unsupervised temporal ensemble alignment for rapid annotation

This paper presents a novel framework for the unsupervised alignment of an ensemble of temporal sequences. This approach draws inspiration from the axiom that an ensemble of temporal signals stemming from the same source/class should have lower rank when "aligned" rather than "misaligned". Our approach shares similarities with recent state of the art methods for unsupervised images ensemble alignment (e.g. RASL) which breaks the problem into a set of image alignment problems (which have well known solutions i.e. the Lucas-Kanade algorithm). Similarly, we propose a strategy for decomposing the problem of temporal ensemble alignment into a similar set of independent sequence problems which we claim can be solved reliably through Dynamic Time Warping (DTW). We demonstrate the utility of our method using the Cohn-Kanade+ dataset, to align expression onset across multiple sequences, which allows us to automate the rapid discovery of event annotations.

Joint evaluation of multiple speech patterns for speech recognition and training

We are addressing the novel problem of jointly evaluating multiple speech patterns for automatic speech recognition and training. We propose solutions based on both the non-parametric dynamic time warping (DTW) algorithm, and the parametric hidden Markov model (HMM). We show that a hybrid approach is quite effective for the application of noisy speech recognition. We extend the concept to HMM training wherein some patterns may be noisy or distorted. Utilizing the concept of ``virtual pattern'' developed for joint evaluation, we propose selective iterative training of HMMs. Evaluating these algorithms for burst/transient noisy speech and isolated word recognition, significant improvement in recognition accuracy is obtained using the new algorithms over those which do not utilize the joint evaluation strategy.