Towards autonomous habitat classification using Gaussian Mixture Models

In this work we present and apply a new Hyperspectral image classification method based on Dirichlet Process Gaussian mixture models (also known as infinite Gaussian mixture models - IGMMs). Since this approach is a non-parametric Bayesian method, we circumvent the problem of model selection which is unavoidable and often difficult when employing traditional parametric Gaussian mixture models (GMM). In order to infer model parameters from observations, we adopt a Gibbs Sampling method to sample the posterior distributions of these parameters. In the preprocessing step, we use Local Fisher's Discriminant Analysis (LFDA) for dimension reduction since we expect it to preserve the multi-modal, non-Gaussian structure of the hyperspectral data. We compared our proposed IGMM based classification method to the existing state-of-the-art classification methods using popular hyperspectral imagery datasets. To simulate challenging real-world sensing environment, we mix every pixel with background and study the performance of these classification approaches. Our experiments show that the proposed LFDA-IGMM method and GMM method have almost the same performance for most cases, given “pure” data - with mixed pixels, LFDA-IGMM outperformed GMM for sub-pixel classification. Both methods outperformed the other commonly used classification approaches when the training sample size is relatively large. Also, the IGMM based method is less sensitive to pixel mixing than other methods.

Air quality has always been a hot issue of concern to the people, the environmental protection department and the government. Among the massive air quality data, abnormal data can interfere with subsequent experiments and analysis. Therefore, it is necessary to detect abnormal data to improve the accuracy of the data. However, traditional air outlier detection methods require at least one year's data to make inferences about air quality. This paper firstly analyzes the characteristics of air quality big data, and then proposes a framework based on Bayesian non-parametric clustering, namely Dirichlet Process (DP) clustering framework, to realize the outlier detection of air quality. The framework optimizes Gaussian mixture model into infinite Gaussian mixture model according to the results of data analysis, and uses neural network to cluster the data processed by infinite Gaussian mixture model, which effectively improves the clustering accuracy and avoids the need of collecting a large number of training data.

In this paper, the impact of channel state estimation errors in a system employing AMC and multicodes is studied. The channel is modelled as a finite-state Markov chain (FSMC), and a hidden Markov model (HMM) formulation for the above problem is presented. An HMM filter is used and shown to yield an improved estimate of the channel state. Since the transition probabilities of the FSMC are required by the HMM filter, the problem of estimating these probabilities is also addressed. The performance of the proposed scheme is evaluated using computer simulation. The results show that the HMM filter provides a significant throughput improvement over the unfiltered case, especially when the channel state information (CSI) is quite noisy or the normalized Doppler rate (defined as the product of the Doppler rate of the channel and the transmission period) is small

In this chapter, the receding horizon Kalman filter is applied to underwater navigation systems. The ocean covers about two-thirds of the earth and has a great effect on human beings. However, the ocean is overlooked while we focus our attention on land and atmospheric issues; we have not been able to explore the full depths of the ocean, its abundant livings and non-living resources. For example, only recently we have discovered, by using manned submersibles, that a large amount of methane and carbon dioxide comes from the seafloor and extraordinary groups of organisms live in hydrothermal vent areas. However, a number of complex issues due to the unstructured and hazardous undersea environment make it difficult to survey in the ocean even though today’s technologies have allowed humans to land on the moon and robots to travel to Mars. Unmanned underwater vehicles (UUVs) can help us better understand marine and other environmental issues, protect the ocean resources of the earth from pollution, and efficiently utilize them for human welfare. The UUV is a platform for a variety of sensors: acoustic, magnetic, gravimetric and chemical ones. Most commercial UUVs are tethered and remotely operated, referred to as remotely operated vehicles (ROVs). Extensive use of manned submersibles and ROVs are currently limited to a few applications because of very high operational costs, operator fatigue and safety issues. The demand for advanced underwater vehicle technologies is growing and will eventually lead to fully autonomous and reliable underwater vehicles. Autonomous underwater vehicles (AUVs) were initially developed to perform missions that were not easy for ROVs and manned underwater vehicles. Since the autonomy allows AUVs to be used for risky missions such as a mine countermeasure (MCM) or under-ice operations, AUVs are replacing ROVs towed vehicles as well as manned underwater vehicles (Whitcomb, 2000). For detailed ocean surveys, an AUV acts as a more stable platform for precision sensors than ROVs or towed vehicles because an AUV is not subject to physical disturbances transmitted along the cable to the surface vessel. This absence of physical attachment also allows AUVs to measure ocean characteristics at specific depths and perform bottom-following missions as owing to its autonomy. In short, An AUV provides marine researchers with a new form of access to deeper ocean. For an AUV to successfully complete a typical survey mission, it must follow a path specified by the operator as closely as possible and arrive at a precise location for collecting data. When an AUV is not able to follow the path accurately during the mission, critical Source: Kalman Filter, Book edited by: Vedran Kordic, ISBN 978-953-307-094-0, pp. 390, May 2010, INTECH, Croatia, downloaded from SCIYO.COM

This paper presents a novel statistical method that can classify given audio events into known classes or recognize them as an unknown class. We propose a nested infinite Gaussian mixture model (iGMM) to represent varied audio events in real environment. One of the main problems of conventional classification methods is that we need to specify a fixed number of classes in advance. Therefore, all audio events are forced to be classified into known classes. To solve the problem, the proposed method formulates a infinite Gaussian mixture model (iGMM) in which the number of classes are allowed to increase without bound. Another problem is that the complexity of each audio event is different. Then, the nested iGMM using nonparametric Bayesian approach is applied to adjust the needed dimension of each audio model. Experimental results show the effectiveness for these two problems to represent the given audio events.

A hybrid model comprised of Gaussian Mixtures Models (GMMs) and Hidden Markov Models (HMMs) is used to model generic sounds with large intra class perceptual variations. Each class has variable number of mixture components in the GMM. The number of mixture components is derived using the Minimum Description Length (MDL) criterion. The overall performance of the hybrid model was compared against models based on HMMs and GMMs with a fixed number of mixture components across all classes. We show that a hybrid model outperforms both class-based GMMs, HMMs, and GMMs based on fixed number of components. Further, our experiments revealed that the contribution of transitions between states in HMMs has no significant effect on the overall classification performance of generic sounds when large intra class perceptual variations are present among sounds in the training and test datasets. Sounds that show multi-event structure with events that tend to be similar (repetitive) indicated improved performance when modeled with HMMs that can be attributed to HMM’s state transition property. Conversely, GMMs indicate better performance when the sound samples show subtle or no repetitive behavior. These results were validated using the MuscleFish sound database.

Probabilistic load flow (PLF) calculation, as a fundamental tool to analyze transmission system behavior, has been studied for decades. Despite a variety of available methods, existing PLF approaches rarely take system control into account. However, system control, as an automatic buffer between the fluctuations in random variables and the variations in system variables (e.g., nodal voltages), has a significant impact on the final PLF result. To consider control actions' influence, this paper proposes the first analytical PLF method for the transmission grid that takes into account primary and secondary frequency controls. This method is based on a high-precision linear power flow model, whose precision is even further improved in this paper by an original correction approach. This paper also proves that if the joint probability distribution (JPD) of random variables is expressed by a Gaussian mixture model (GMM), then the JPD of system variables is an infinite GMM. By leveraging this proposition, the proposed method generates the joint PLF of the whole system at the quasi-steady and steady states. The high accuracy and satisfactory efficiency of this method are verified on multiple test systems with different percentages of renewables.

Unsupervised speech processing methods are essential for applications ranging from zero-resource speech technology to modelling child language acquisition. One challenging problem is discovering the word inventory of the language: the lexicon. Lexical clustering is the task of grouping unlabelled acoustic word tokens according to type. We propose a novel lexical clustering model: variable-length word segments are embedded in a fixed-dimensional acoustic space in which clustering is then performed. We evaluate several clustering algorithms and find that the best methods produce clusters with wide variation in sizes, as observed in natural language. The best probabilistic approach is an infinite Gaussian mixture model (IGMM), which automatically chooses the number of clusters. Performance is comparable to that of non-probabilistic Chinese Whispers and average-linkage hierarchical clustering. We conclude that IGMM clustering of fixed-dimensional embeddings holds promise as the lexical clustering component in unsupervised speech processing systems.

We present a new framework to detect various types of variable objects within massive astronomical time-series data. Assuming that the dominant population of objects is non-variable, we find outliers from this population by using a non-parametric Bayesian clustering algorithm based on an infinite GaussianMixtureModel (GMM) and the Dirichlet Process. The algorithm extracts information from a given dataset, which is described by six variability indices. The GMM uses those variability indices to recover clusters that are described by six-dimensional multivariate Gaussian distributions, allowing our approach to consider the sampling pattern of time-series data, systematic biases, the number of data points for each light curve, and photometric quality. Using the Northern Sky Variability Survey data, we test our approach and prove that the infinite GMM is useful at detecting variable objects, while providing statistical inference estimation that suppresses false detection. The proposed approach will be effective in the exploration of future surveys such as GAIA, Pan-Starrs, and LSST, which will produce massive time-series data.

The underwater environment severely constrains robotic navigation and communications, making the use of traditional multi-robot control and coordination schemes very difficult. These challenges are exacerbated on a new generation of low-cost autonomous underwater vehicles (AUVs) that lack a Doppler velocity log, acoustic modem, or high-end inertial sensors typically used for underwater robotic navigation and communications.This letter demonstrates multi-robot operations for low-cost AUVs via a novel and user-friendly operating paradigm that allows intuitive command and control of an AUV group. Each vehicle is equipped with a low-power and inexpensive acoustic system for navigation and receipt of operator commands. This system consists of a passive array of hydrophones and a timed acquisition and data processing stack that allows each AUV to self-localize relative to a time-synchronized acoustic beacon. Switching between different operational “modes” on the beacon causes it to broadcast different acoustic signals which, when received by the AUVs, result in the vehicles switching between different autonomous behaviors. These behaviors are defined in a beacon-centric coordinate system using pre-defined parameters unique to each vehicle; as a result, the movement of the beacon itself allows the operator to control the group-wide movement of all vehicles concurrently. This letter presents field experiments with three SandShark AUVs in which operational mode and beacon movement were controlled by an operator. However, by installing the beacon on an autonomous vehicle, this paradigm would provide a method for remote command and control of an arbitrarily large number of miniature, low-cost AUVs, without the need for sophisticated navigational sensors or acoustic modems.

Human-driven Vehicles (HVs) and Autonomous Vehicles (AVs) will inevitably coexist in the future, resulting in mixed traffic conditions. Modeling car-following behavior in such mixed traffic is crucial for better understanding the differences in driving behavior between AVs and HVs, thereby facilitating the development of AV driving strategies to enhance traffic safety. In such mixed traffic, the car-following modes typically include AVs following HVs, HVs following HVs and HVs following AVs. Previous studies have compared the driving behavior of HVs following HVs and HVs following AVs, but few have considered the driving behavior of AVs following HVs in the comparison. To bridge this gap, this paper first decomposes car-following events in mixed traffic extracted from a Lyft level-5 dataset into driving primitives using the Hidden Markov Model with Gaussian Mixture Model (GMM-HMM). Then, the Latent Dirichlet Allocation with GMM (GMM-LDA) is introduced to cluster the driving primitives into different car-following behavior patterns. Based on the identified patterns, the differences across the three car-following modes are further explored. The results show that the driving behavior under three car-following modes differs from each other. Specifically, compared to the HVs following HVs mode, in the AVs following HVs mode, AVs tend to maintain car-following behavior patterns with a longer distance gap and adopt a lower speed to ensure safety. This study offers a comprehensive understanding of car-following behavior in mixed traffic, which will facilitate the development of human-like driving strategies for AVs and enhance the safety of mixed traffic.

In this dissertation, a novel approach of integrating template matching with statistical modeling is proposed to improve continuous speech recognition. Hidden Markov Modeling (HMMs) has been the dominant approach in statistical speech recognition since it provides a principled way of jointly modeling speech spectral variations and time dynamics. However, HMMs have the shortcoming of assuming the observations being independent within each state, which makes it ineffective in modeling the details of speech temporal evolutions that are important for characterizing nonstationary speech sounds. Template-based methods make comparisons between a test pattern and the templates derived from training data, and therefore they are able to capture speech dynamics and time correlation of speech frames better than HMM based methods. However, template matching requires large memory space and computational time since feature vectors of training data need to be stored in computer memory for access at the recognition stage, which is difficult in large vocabulary continuous speech recognition (LVCSR). Our proposed approach takes advantages of both statistical modeling and template matching, which overcomes the weakness of conventional template-based method and is feasible for LVCSR. We use multiple Gaussian Mixture Model (GMM) indices to represent each frame of speech templates, and define the template unit to be context-dependent phone segments (triphone context). We also use phonetic decision trees borrowed from those commonly used in HMMs to tie triphone templates and predict triphones unseen in training data. Two local distances, log likelihood ratio (LLR) and Kullback-Leibler (KL) divergence, are proposed for dynamic time warping (DTW) based template matching. In order to reduce computational complexity and storage space, we propose methods of minimum distance template selection (MDTS) and maximum log-likelihood template selection (MLTS), and investigate a template compression method on top of template selection to further improve recognition performance. The template based methods were used to rescore lattices generated by baseline HMMs on the tasks of TIMIT continuous phone recognition and teleheath LVCSR and experimental results demonstrated that the proposed approach of integrating template matching with statistical modeling significantly improved recognition performances over the HMM baselines. The template selection methods also provided significant recognition accuracy improvements over the HMM baseline while largely reducing the computation and storage complexities. When all templates or MDTS were used, using the LLR local distance obtained better recognition performance than the KL divergence local distance. For MLTS and template compression, KL divergence local distance provided better performance than the LLR local distance, and the template compression method made further improvements over KL based MLTS. Since the templates were constructed based on the GMM indices extracted from HMM baselines, we also validated the effectiveness of the proposed template methods based on enhanced HMM baselines. Experimental results showed that LLR based all template method was able to consistently improve TIMIT phone recognition accuracies based on four enhanced HMM baselines. Prosodic features such as duration, energy, and pitch can reflect longer span information of speech than conventional single frame vectors but they have commonly been ignored by HMMs. Template based methods provide possibilities to conveniently integrate prosodic features into speech recognition, which has not been well studied in the past. In this dissertation, we investigate combining template based methods with the speech prosodic features of duration, energy and pitch to further improve speech recognition accuracy. The scores of prosodic information were computed by a GMM based method and a non-parametric method, and the prosodic scores were combined with the acoustic scores in triphone template matching. Experimental results obtained on the telehealth task showed that prosodic information had positive effects on vowel sound recognition.

SUMMARYGround‐penetrating radar (GPR) is a useful tool for performing subsurface imaging by using radar pulses. In a previous paper we proposed a method for denoising GPR signals by using the 2D Gabor wavelet transform. In this paper we present a new method for emphasizing GPR reflected waves from buried objects. We evaluate the results of time–frequency analysis of the reflected waves on the basis of the Markov Chain Monte Carlo (MCMC) and Infinite Gaussian Mixture Model (IGMM) methods. Our methods are found to be effective as preprocessing for detecting the positions of buried metal pipes. © 2013 Wiley Periodicals, Inc. Electron Comm Jpn, 96(10): 22–36, 2013; Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/ecj.11484

For reducing the channel switching frequency due to uncertain and frequent primary users' interruptions, we design a nonparametric channel allocation algorithm which can effectively manage the spectrum resources in distributed cognitive radio networks and serve for delay-sensitive data traffics. In the proposed algorithm, an infinite Gaussian mixture model is used to get the posterior decision from the feature spaces of the channel idle duration and the primary user's frequency channel, while the collapsed Gibbs sampling is applied for fast convergence. Experimental and simulation results demonstrate that the proposed channel allocation algorithm outperforms the random channel allocation method and the idle probability based channel allocation algorithm in terms of throughput and packet drops.

In recent years, precious observations by autonomous underwater vehicle (AUV) have been performed. Tracking and navigation of AUV are important for precious observations. In Japan agency for Marine-earth Science and Technology (JAMSTEC) the observations with AUV ldquoURASHIMArdquo have been implemented in deep ocean area. AUV URASHIMA is navigated by an inertial navigation system (INS) mainly. The absolute position of AUV is tracked on the support vessel with super short base-line (SSBL) method of acoustic positioning. And sometime the own position in AUV is updated by the SSBL estimation result to correct the drift error of INS. However the SSBL is not so accurate. So the position correction includes the SSBL error and the position correction is performed drastically. In this paper, a method for accurate tracking of AUV and rapid correction of the INS status is proposed. The tracking is mainly achieved by SSBL on the support vessel. Additionally the INS data is transmitted from AUV including the acoustic positioning signal. The INS output of relative position to past own position is very accurate within short time. The INS data is used for tracking. Furthermore the position result estimated on the vessel is transmitted to AUV with next interrogation signal. And the own position in AUV is rapidly updated with accurate estimation. Some digital data including INS data is modulated with phase shift keying (PSK) method and direct sequence spread spectrum (DSSS) technique for transmission with acoustic positioning signal. The proposed method was considered by simulation. And good results were obtained.