Marine mammal species detection and classification

Thesis (Ph.D.)--University of Washington, 2016-04

GANN: Genetic algorithm neural networks for the detection of conserved combinations of features in DNA

Background: The multitude of motif detection algorithms developed to date have largely focused on the detection of patterns in primary sequence. Since sequence-dependent DNA structure and flexibility may also play a role in protein-DNA interactions, the simultaneous exploration of sequence-and structure-based hypotheses about the composition of binding sites and the ordering of features in a regulatory region should be considered as well. The consideration of structural features requires the development of new detection tools that can deal with data types other than primary sequence. Results: GANN ( available at http://bioinformatics.org.au/gann) is a machine learning tool for the detection of conserved features in DNA. The software suite contains programs to extract different regions of genomic DNA from flat files and convert these sequences to indices that reflect sequence and structural composition or the presence of specific protein binding sites. The machine learning component allows the classification of different types of sequences based on subsamples of these indices, and can identify the best combinations of indices and machine learning architecture for sequence discrimination. Another key feature of GANN is the replicated splitting of data into training and test sets, and the implementation of negative controls. In validation experiments, GANN successfully merged important sequence and structural features to yield good predictive models for synthetic and real regulatory regions. Conclusion: GANN is a flexible tool that can search through large sets of sequence and structural feature combinations to identify those that best characterize a set of sequences.

Multisensorial obstacles detection and classification for ADAS and autonomous driving

A reliable perception of the real world is a key-feature for an autonomous vehicle and the Advanced Driver Assistance Systems (ADAS). Obstacles detection (OD) is one of the main components for the correct reconstruction of the dynamic world. Historical approaches based on stereo vision and other 3D perception technologies (e.g. LIDAR) have been adapted to the ADAS first and autonomous ground vehicles, after, providing excellent results. The obstacles detection is a very broad field and this domain counts a lot of works in the last years. In academic research, it has been clearly established the essential role of these systems to realize active safety systems for accident prevention, reflecting also the innovative systems introduced by industry. These systems need to accurately assess situational criticalities and simultaneously assess awareness of these criticalities by the driver; it requires that the obstacles detection algorithms must be reliable and accurate, providing: a real-time output, a stable and robust representation of the environment and an estimation independent from lighting and weather conditions. Initial systems relied on only one exteroceptive sensor (e.g. radar or laser for ACC and camera for LDW) in addition to proprioceptive sensors such as wheel speed and yaw rate sensors. But, current systems, such as ACC operating at the entire speed range or autonomous braking for collision avoidance, require the use of multiple sensors since individually they can not meet these requirements. It has led the community to move towards the use of a combination of them in order to exploit the benefits of each one. Pedestrians and vehicles detection are ones of the major thrusts in situational criticalities assessment, still remaining an active area of research. ADASs are the most prominent use case of pedestrians and vehicles detection. Vehicles should be equipped with sensing capabilities able to detect and act on objects in dangerous situations, where the driver would not be able to avoid a collision. A full ADAS or autonomous vehicle, with regard to pedestrians and vehicles, would not only include detection but also tracking, orientation, intent analysis, and collision prediction. The system detects obstacles using a probabilistic occupancy grid built from a multi-resolution disparity map. Obstacles classification is based on an AdaBoost SoftCascade trained on Aggregate Channel Features. A final stage of tracking and fusion guarantees stability and robustness to the result.

Anomaly Detection in Noisy Images

Finding rare events in multidimensional data is an important detection problem that has applications in many fields, such as risk estimation in insurance industry, finance, flood prediction, medical diagnosis, quality assurance, security, or safety in transportation. The occurrence of such anomalies is so infrequent that there is usually not enough training data to learn an accurate statistical model of the anomaly class. In some cases, such events may have never been observed, so the only information that is available is a set of normal samples and an assumed pairwise similarity function. Such metric may only be known up to a certain number of unspecified parameters, which would either need to be learned from training data, or fixed by a domain expert. Sometimes, the anomalous condition may be formulated algebraically, such as a measure exceeding a predefined threshold, but nuisance variables may complicate the estimation of such a measure. Change detection methods used in time series analysis are not easily extendable to the multidimensional case, where discontinuities are not localized to a single point. On the other hand, in higher dimensions, data exhibits more complex interdependencies, and there is redundancy that could be exploited to adaptively model the normal data. In the first part of this dissertation, we review the theoretical framework for anomaly detection in images and previous anomaly detection work done in the context of crack detection and detection of anomalous components in railway tracks. In the second part, we propose new anomaly detection algorithms. The fact that curvilinear discontinuities in images are sparse with respect to the frame of shearlets, allows us to pose this anomaly detection problem as basis pursuit optimization. Therefore, we pose the problem of detecting curvilinear anomalies in noisy textured images as a blind source separation problem under sparsity constraints, and propose an iterative shrinkage algorithm to solve it. Taking advantage of the parallel nature of this algorithm, we describe how this method can be accelerated using graphical processing units (GPU). Then, we propose a new method for finding defective components on railway tracks using cameras mounted on a train. We describe how to extract features and use a combination of classifiers to solve this problem. Then, we scale anomaly detection to bigger datasets with complex interdependencies. We show that the anomaly detection problem naturally fits in the multitask learning framework. The first task consists of learning a compact representation of the good samples, while the second task consists of learning the anomaly detector. Using deep convolutional neural networks, we show that it is possible to train a deep model with a limited number of anomalous examples. In sequential detection problems, the presence of time-variant nuisance parameters affect the detection performance. In the last part of this dissertation, we present a method for adaptively estimating the threshold of sequential detectors using Extreme Value Theory on a Bayesian framework. Finally, conclusions on the results obtained are provided, followed by a discussion of possible future work.

Precoder Detection for Cooperative Decode-and-Forward Relaying in OFDMA Systems

We consider an LTE network where a secondary user acts as a relay, transmitting data to the primary user using a decode-and-forward mechanism, transparent to the base-station (eNodeB). Clearly, the relay can decode symbols more reliably if the employed precoder matrix indicators (PMIs) are known. However, for closed loop spatial multiplexing (CLSM) transmit mode, this information is not always embedded in the downlink signal, leading to a need for effective methods to determine the PMI. In this thesis, we consider 2x2 MIMO and 4x4 MIMO downlink channels corresponding to CLSM and formulate two techniques to estimate the PMI at the relay using a hypothesis testing framework. We evaluate their performance via simulations for various ITU channel models over a range of SNR and for different channel quality indicators (CQIs). We compare them to the case when the true PMI is known at the relay and show that the performance of the proposed schemes are within 2 dB at 10% block error rate (BLER) in almost all scenarios. Furthermore, the techniques add minimal computational overhead over existent receiver structure. Finally, we also identify scenarios when using the proposed precoder detection algorithms in conjunction with the cooperative decode-and-forward relaying mechanism benefits the PUE and improves the BLER performance for the PUE. Therefore, we conclude from this that the proposed algorithms as well as the cooperative relaying mechanism at the CMR can be gainfully employed in a variety of real-life scenarios in LTE networks.

Proposição e avaliação de um método baseado em projeção de luz para reconstrução e análise da superfície do tronco durante a respiração

Universidade Estadual de Campinas . Faculdade de Educação Física

S/MIMO MC-CDMA Heuristic Multiuser Detectors Based on Single-Objective Optimization

This paper analyzes the complexity-performance trade-off of several heuristic near-optimum multiuser detection (MuD) approaches applied to the uplink of synchronous single/multiple-input multiple-output multicarrier code division multiple access (S/MIMO MC-CDMA) systems. Genetic algorithm (GA), short term tabu search (STTS) and reactive tabu search (RTS), simulated annealing (SA), particle swarm optimization (PSO), and 1-opt local search (1-LS) heuristic multiuser detection algorithms (Heur-MuDs) are analyzed in details, using a single-objective antenna-diversity-aided optimization approach. Monte- Carlo simulations show that, after convergence, the performances reached by all near-optimum Heur-MuDs are similar. However, the computational complexities may differ substantially, depending on the system operation conditions. Their complexities are carefully analyzed in order to obtain a general complexity-performance framework comparison and to show that unitary Hamming distance search MuD (uH-ds) approaches (1-LS, SA, RTS and STTS) reach the best convergence rates, and among them, the 1-LS-MuD provides the best trade-off between implementation complexity and bit error rate (BER) performance.

Guitar solo tracker

A motivação para este trabalho vem da necessidade que o autor tem em poder registar as notas tocadas na guitarra durante o processo de improviso. Quando o músico está a improvisar na guitarra, muitas vezes não se recorda das notas tocadas no momento, este trabalho trata o desenvolvimento de uma aplicação para guitarristas, que permita registar as notas tocadas na guitarra eléctrica ou clássica. O sinal é adquirido a partir da guitarra e processado com requisitos de tempo real na captura do sinal. As notas produzidas pela guitarra eléctrica, ligada ao computador, são representadas no formato de tablatura e/ou partitura. Para este efeito a aplicação capta o sinal proveniente da guitarra eléctrica a partir da placa de som do computador e utiliza algoritmos de detecção de frequência e algoritmos de estimação de duração de cada sinal para construir o registo das notas tocadas. A aplicação é desenvolvida numa perspectiva multi-plataforma, podendo ser executada em diferentes sistemas operativos Windows e Linux, usando ferramentas e bibliotecas de domínio público. Os resultados obtidos mostram a possibilidade de afinar a guitarra com valores de erro na ordem de 2 Hz em relação às frequências de afinação standard. A escrita da tablatura apresenta resultados satisfatórios, mas que podem ser melhorados. Para tal será necessário melhorar a implementação de técnicas de processamento do sinal bem como a comunicação entre processos para resolver os problemas encontrados nos testes efectuados.

Groundtruth system for underwater benchmarking

Oceans - San Diego, 2013

Deteção remota na avaliação das alterações da cobertura do solo no Parque Nacional da Peneda-Gerês

Dissertação de mestrado integrado em Engenharia Civil

Towards Detecting Changes in Underwater Image Sequences

Detecting changes between images of the same scene taken at different times is of great interest for monitoring and understanding the environment. It is widely used for on-land application but suffers from different constraints. Unfortunately, Change detection algorithms require highly accurate geometric and photometric registration. This requirement has precluded their use in underwater imagery in the past. In this paper, the change detection techniques available nowadays for on-land application were analyzed and a method to automatically detect the changes in sequences of underwater images is proposed. Target application scenarios are habitat restoration sites, or area monitoring after sudden impacts from hurricanes or ship groundings. The method is based on the creation of a 3D terrain model from one image sequence over an area of interest. This model allows for synthesizing textured views that correspond to the same viewpoints of a second image sequence. The generated views are photometrically matched and corrected against the corresponding frames from the second sequence. Standard change detection techniques are then applied to find areas of difference. Additionally, the paper shows that it is possible to detect false positives, resulting from non-rigid objects, by applying the same change detection method to the first sequence exclusively. The developed method was able to correctly find the changes between two challenging sequences of images from a coral reef taken one year apart and acquired with two different cameras

A copula-based method for synthetic microarray data generation

In this work, we propose a copula-based method to generate synthetic gene expression data that account for marginal and joint probability distributions features captured from real data. Our method allows us to implant significant genes in the synthetic dataset in a controlled manner, giving the possibility of testing new detection algorithms under more realistic environments.

Towards Detecting Changes in Underwater Image Sequences

Detecting changes between images of the same scene taken at different times is of great interest for monitoring and understanding the environment. It is widely used for on-land application but suffers from different constraints. Unfortunately, Change detection algorithms require highly accurate geometric and photometric registration. This requirement has precluded their use in underwater imagery in the past. In this paper, the change detection techniques available nowadays for on-land application were analyzed and a method to automatically detect the changes in sequences of underwater images is proposed. Target application scenarios are habitat restoration sites, or area monitoring after sudden impacts from hurricanes or ship groundings. The method is based on the creation of a 3D terrain model from one image sequence over an area of interest. This model allows for synthesizing textured views that correspond to the same viewpoints of a second image sequence. The generated views are photometrically matched and corrected against the corresponding frames from the second sequence. Standard change detection techniques are then applied to find areas of difference. Additionally, the paper shows that it is possible to detect false positives, resulting from non-rigid objects, by applying the same change detection method to the first sequence exclusively. The developed method was able to correctly find the changes between two challenging sequences of images from a coral reef taken one year apart and acquired with two different cameras

Cloud clearing techniques over land for land surface temperature retrieval from the Advanced Along Track Scanning Radiometer

We present five new cloud detection algorithms over land based on dynamic threshold or Bayesian techniques, applicable to the Advanced Along Track Scanning Radiometer (AATSR) instrument and compare these with the standard threshold based SADIST cloud detection scheme. We use a manually classified dataset as a reference to assess algorithm performance and quantify the impact of each cloud detection scheme on land surface temperature (LST) retrieval. The use of probabilistic Bayesian cloud detection methods improves algorithm true skill scores by 8-9 % over SADIST (maximum score of 77.93 % compared to 69.27 %). We present an assessment of the impact of imperfect cloud masking, in relation to the reference cloud mask, on the retrieved AATSR LST imposing a 2 K tolerance over a 3x3 pixel domain. We find an increase of 5-7 % in the observations falling within this tolerance when using Bayesian methods (maximum of 92.02 % compared to 85.69 %). We also demonstrate that the use of dynamic thresholds in the tests employed by SADIST can significantly improve performance, applicable to cloud-test data to provided by the Sea and Land Surface Temperature Radiometer (SLSTR) due to be launched on the Sentinel 3 mission (estimated 2014).

THE VORONOI TESSELLATION CLUSTER FINDER IN 2+1 DIMENSIONS

We present a detailed description of the Voronoi Tessellation (VT) cluster finder algorithm in 2+1 dimensions, which improves on past implementations of this technique. The need for cluster finder algorithms able to produce reliable cluster catalogs up to redshift 1 or beyond and down to 10(13.5) solar masses is paramount especially in light of upcoming surveys aiming at cosmological constraints from galaxy cluster number counts. We build the VT in photometric redshift shells and use the two-point correlation function of the galaxies in the field to both determine the density threshold for detection of cluster candidates and to establish their significance. This allows us to detect clusters in a self-consistent way without any assumptions about their astrophysical properties. We apply the VT to mock catalogs which extend to redshift 1.4 reproducing the ACDM cosmology and the clustering properties observed in the Sloan Digital Sky Survey data. An objective estimate of the cluster selection function in terms of the completeness and purity as a function of mass and redshift is as important as having a reliable cluster finder. We measure these quantities by matching the VT cluster catalog with the mock truth table. We show that the VT can produce a cluster catalog with completeness and purity > 80% for the redshift range up to similar to 1 and mass range down to similar to 10(13.5) solar masses.