Search

Due to the intensification of the exploitation of the oceans, it becomes crucial to better and at a larger scale monitor marine life before impacting operations and to monitor the evolution of the ambient noise in known habitats. This is usually done with hydrophones deployed underwater at the vicinity of the site of interest. However, the processing of the collected data is a laborious task where an expert must listen to and look at each time-frame to ensure the quality of the results. The automation of the sound identification has boomed since the progress of machine learning and in particular computer vision. Indeed, models based on Convolutional Neural Network (CNN) can now identify marine mammals or other ambient sounds in a record time. However, the question of the generalization to multiple sites of such models is barely studied. This paper trains a simple CNN with the ShipEars and DOSITS datasets to identify large vessels, small vessels, dolphins, and background noise. The model is then validated with a sample of the REP(MUS) dataset. Three types of optimizations are proposed to improve the model’s performance on REP(MUS): the data variability, the time discretization, and a Bayesian optimization of the hyperparameters. [ABSTRACT FROM AUTHOR]

Effective selection of muon candidates is the cornerstone of the LHC physics programme. The ATLAS experiment uses a two-level trigger system for real-time selection of interesting collision events. The first-level hardware trigger system uses the Resistive Plate Chamber detector (RPC) for selecting muon candidates in the central (barrel) region of the detector. With the planned upgrades, the entirely new FPGA-based muon trigger system will be installed in 2025-2026. In this paper, neural network regression models are studied for potential applications in the new RPC trigger system. A simple simulation model of the current detector is developed for training and testing neural network regression models. Effects from additional cluster hits and noise hits are evaluated. Efficiency of selecting muon candidates is estimated as a function of the transverse muon momentum. Several models are evaluated and their performance is compared to that of the current detector, showing promising potential to improve on current algorithms for the ATLAS Phase-II barrel muon trigger upgrade. [ABSTRACT FROM AUTHOR]

Classification is an important operation in Automatic Target Recognition in sonar images. However, when a sonar image snippet contains an object belonging to a new class that was not seen during the training phase, the classifier cannot assign the correct class to it. Meta-learning is a solution to this problem. In one form of meta-learning, the original system is trained to emit a similarity measure between a support image and a query image. Then in inference, a given query image is classified into the class of the support image that has the highest similarity with the query image. Inference in such a system can be viewed as "inverse image search". In this work, we consider the problem of inverse image search in the context of an target recognition operator support module for synthetic aperture sonar images. The core challenge in building a system for image retrieval based on inverse image search is to design an algorithm that can output a relevant similarity measure for two input images. In this paper, we use a convolutional neural network (CNN) for similarity computation. Our experiments show that these measures are able to capture the similarity as perceived by humans and independent of class. [ABSTRACT FROM AUTHOR]

We are studying the use of deep neural networks (DNNs) to identify and locate primary vertices (PVs) in proton-proton collisions at the LHC. Earlier work focused on finding primary vertices in simulated LHCb data using a hybrid approach that started with kernel density estimators (KDEs) derived heuristically from the ensemble of charged track parameters and predicted "target histogram" proxies, from which the actual PV positions are extracted. We have recently demonstrated that using a UNet architecture performs indistinguishably from a "flat" convolutional neural network model. We have developed an "end-to-end" tracks-to-hist DNN that predicts target histograms directly from track parameters using simulated LHCb data that provides better performance (a lower false positive rate for the same high efficiency) than the best KDE-tohists model studied. This DNN also provides better efficiency than the default heuristic algorithm for the same low false positive rate. "Quantization" of this model, using FP16 rather than FP32 arithmetic, degrades its performance minimally. Reducing the number of UNet channels degrades performance more substantially. We have demonstrated that the KDE-to-hists algorithm developed for LHCb data can be adapted to ATLAS and ACTS data using two variations of the UNet architecture. Within ATLAS/ACTS, these algorithms have been validated against the standard vertex finder algorithm. Both variations produce PVfinding efficiencies similar to that of the standard algorithm and vertex-vertex separation resolutions that are significantly better. [ABSTRACT FROM AUTHOR]

The Time Projection Chamber (TPC) of the ALICE experiment at the CERN LHC was upgraded for Run 3 and Run 4. Readout chambers based on Gas Electron Multiplier (GEM) technology and a new readout scheme allow continuous data taking at the highest interaction rates expected in Pb-Pb collisions. Due to the absence of a gating grid system, a significant amount of ions created in the multiplication region is expected to enter the TPC drift volume and distort the uniform electric field that guides the electrons to the readout pads. Analytical calculations were considered to correct for space-charge distortion fluctuations but they proved to be too slow for the calibration and reconstruction workflow in Run 3. In this paper, we discuss a novel strategy developed by the ALICE Collaboration to perform distortion-fluctuation corrections with machine learning and convolutional neural network techniques. The results of preliminary studies are shown and the prospects for further development and optimization are also discussed. [ABSTRACT FROM AUTHOR]

Ultrasound imaging is safe, relatively affordable, and capable of real-time performance. This technology has been used for real-time visualization and analyzing the functionality of human organs such as human tongue shape and motion during a real-time speech to study healthy or impaired speech production. Due to the noisy nature of ultrasound images with low-contrast characteristic, it might require expertise for non-expert users to recognize tongue shapes. To alleviate this difficulty and for quantitative analysis of tongue shape and motion, tongue surface can be extracted, tracked, and visualized instead of the whole tongue region. The progress of deep convolutional neural networks has been successfully exploited in various computer vision tasks. This paper presents two novel deep neural network models named BowNet and wBowNet benefits from the ability of global prediction of decoding-encoding models from integrated multi -scale-contextual information and the capability of full-resolution (local) extraction of dilated convolutions. Experimental results using ultrasound tongue image datasets from different ultrasound devices revealed that the combination of both localization and globalization searching could improve prediction result significantly. Assessment of BowNet models in real-time using both qualitatively and quantitatively studies showed their outstanding achievements in terms of accuracy and robustness in comparison with similar techniques. [ABSTRACT FROM AUTHOR]

Identifying changes in the Earth's phenomena is vital for understanding and mitigating the impacts of environmental issues. Monitoring the Earth's surface phenomena can be carried out effectively using satellite images acquired at different times. In addition to spectral features, spatial features play a significant role in detecting precise changes. However, classical change detection (CD) methods rarely consider spatial information and fail to account for scale variations within images. The present study introduces a novel deep learning-based CD method that hierarchically extracts spatial–spectral features at various scales to address these issues. The proposed deep neural network generates a binary change map by employing a multi-scale approach that integrates the information of patches of varied sizes at the decision level. We conducted experiments using Landsat-8 images from Sahand City, East Azarbaijan, Iran, because of their remarkable capacity to represent the Earth's surface details. Tabriz's population growth has led to rapid development in Sahand City to accommodate citizens. Studying these changes can offer valuable insights into urban planning. The performance of the proposed deep model is evaluated in comparison to two classical methods, the change vector analysis (CVA) method and a random forest (RF) algorithm. Based on the change detection results, the proposed deep learning network demonstrates a significant improvement in the kappa coefficient (KC) compared to the RF and CVA methods, with increases of approximately 11.86% and 29.36%, respectively. Furthermore, in terms of overall accuracy (O.A.), the proposed network outperforms both the RF and CVA methods by approximately 17.08% and 29.16%, respectively. The proposed multi-scale deep network performs better at detecting changes across all metrics. As a result, the CVA method fails to identify changes with sufficient accuracy. [ABSTRACT FROM AUTHOR]

Lamb-wave testing for Structural Health Monitoring (SHM) is complicated by the dispersive nature of wave modes, which deteriorates the wave spatial resolution and makes the experimental data hard to interpret. Mathematical operators such as the Warped Frequency Transform (WFT) are particularly suited for the analysis of Guided Waves (GWs). Indeed, WFT-based analysis methods are capable to achieve sparse representations of GW signals. These methods naturally lead to super-resolved and artifact-free representations, even in noisy environments, and are particularly effective to extract the information on the wave distance of propagation. The concept of sparse representations is also the basis of the so-called compressive sensing (CS) theory. CS proves that a signal which is sparse in a given representation can be compressed directly at the sampling stage. In this paper, a CS framework for Lamb wave field acquisitions with air-coupled probes or laser-Doppler vibrometers will be reviewed. Moreover, it will be shown how the capability to restore high-resolution details from CS input images can be improved dramatically by recent breakthroughs in deep learning and convolutional neural networks. [ABSTRACT FROM AUTHOR]

In the following work, a benchmark of different non-intrusive model reduction approaches is performed on an explicit dynamic contact 3D-problem. The main purpose of this work is to evaluate the stability of the reduced model with respect to time along with the precision of these approaches with respect to the true solutions of interest. These solutions are the prediction of displacement and velocity fields. The precision of these approaches is also evaluated with respect to the evolution of some materials parameters. Six parameters vary in this study and we would like to predict the whole transient fast dynamic impact response with respect to each parameters. To this end, several models are trained: Proper Orthogonal Decomposition (POD) and Deep convolutional Neural Network (DcNN), in addition, a vectorized version of Interpolation in Grassman Manifolds is proposed. The benchmark performed illustrate that using DcNN's allows to achieve the best precision and stability in predicting physical fields. [ABSTRACT FROM AUTHOR]

The use of convolutional neuronal networks (CNN) for the treatment of interferometric fringes has been introduced in recent years. In this paper, we optimize and build a CNN model, based U-NET architecture, to maximize its performance processing electronic speckle interferometry fringes (ESPI). [ABSTRACT FROM AUTHOR]

Nowadays, all-sky imagers (ASI) provide valuable information regarding the sky's state, and they have been extensively used in cloud detection, segmentation, and solar forecasting studies. In this study, global and diffuse horizontal irradiances (GHI and DHI) are modeled using a Convolutional Neural Network (CNN) and Red-Green-Blue (RGB) information retrieved through ASI images. The predicted GHI and DHI underestimated observations with systematic biases of -1.8 W m-2 and -0.5 W m-2, while the dispersion errors were 82.7 W m-2 and 39.8 W m-2, respectively. The correlation coefficient was high, approaching 0.95 and 0.85 for GHI and DHI. [ABSTRACT FROM AUTHOR]

The introduction of relevant physical information into neural network architectures has become a widely used and successful strategy for improving their performance. In lattice gauge theories, such information can be identified with gauge symmetries, which are incorporated into the network layers of our recently proposed Lattice Gauge Equivariant Convolutional Neural Networks (L-CNNs). L-CNNs can generalize better to differently sized lattices than traditional neural networks and are by construction equivariant under lattice gauge transformations. In these proceedings, we present our progress on possible applications of L-CNNs to Wilson flow or continuous normalizing flow. Our methods are based on neural ordinary differential equations which allow us to modify link configurations in a gauge equivariant manner. For simplicity, we focus on simple toy models to test these ideas in practice. [ABSTRACT FROM AUTHOR]

State-of-the art deep learning models have shown remarkable performance on computer vision tasks like object classification or detection. These networks are typically trained on large-scale datasets of natural RGB images. However, sidescan sonar images are gray-scaled images representing acoustic intensities. The fundamental differences between camera and sonar as well as the images itself makes it necessary to investigate the transfer of results achieved on RGB images to the sonar imagery domain. Therefore, we compare the deep learning detection algorithm YOLOv2 with its updated version YOLOv3, both adopted for object detection in sidescan sonar images. In addition to this, a small convolutional neural network (CNN) is trained from scratch and used for detection. The experiments answer two questions: First, whether, as for general computer vision problems, transfer learning of large deep learning models is preferable over training of custom networks when dealing with limited sonar data. Secondly, whether improvements in the YOLO architecture, developed based on RGB images, lead to significant improvements on sonar data as well. Our results show that YOLOv3 indeed performs better than YOLOv2. Furthermore, YOLOv3 achieves a true positive rate of up to 98.2% and outperforms the small CNN. [ABSTRACT FROM AUTHOR]

Efficient three-dimensional (3-d) convolutional neural networks (CNNs) are designed and trained for an underwater target classification task with volumetric synthetic aperture sonar (SAS) imagery. (The third dimension of the data represents depth into the sediment, thereby enabling the consideration of buried underwater objects.) The use of tiny networks containing relatively few parameters makes training with enormous input data volumes feasible even with modest computational power and limited computer memory. The promise of the approach is demonstrated for both buried and proud man-made objects present in real, measured SAS data cubes collected at aquatic sites by an experimental volumetric sonar system, called the Sediment Volume Search Sonar (SVSS). The classification performance of each 3-d CNN exhibits marked improvement over a prescreening detection algorithm alone, and the utility of an ensemble approach is also quantified. An analysis of the effective functionality of the learned networks is provided, with this also accompanied by figures showing example trained filters as well as intermediate representations of a data volume containing unexploded ordnance (UXO). The predictions of the 3-d CNN classifiers can provide valuable guidance for the efficient allocation of resources during real-world UXO remediation operations. [ABSTRACT FROM AUTHOR]

Deep neural networks (DNNs) have been applied to the fields of computer vision and natural language processing with great success in recent years. The success of these applications has hinged on the development of specialized DNN architectures that take advantage of specific characteristics of the problem to be solved, namely convolutional neural networks for computer vision and recurrent neural networks for natural language processing. This research explores whether a neural network architecture specific to the task of identifying t → Wb decays in particle collision data yields better performance than a generic, fully-connected DNN. Although applied here to resolved top quark decays, this approach is inspired by an DNN technique for tagging boosted top quarks, which consists of defining custom neural network layers known as the combination and Lorentz layers. These layers encode knowledge of relativistic kinematics applied to combinations of particles, and the output of these specialized layers can then be fed into a fully connected neural network to learn tasks such as classification. This research compares the performance of these physics inspired networks to that of a generic, fully-connected DNN, to see if there is any advantage in terms of classification performance, size of the network, or ease of training. [ABSTRACT FROM AUTHOR]

The success of Convolutional Neural Networks (CNNs) in image classification has prompted efforts to study their use for classifying image data obtained in Particle Physics experiments. Here, we discuss our efforts to apply CNNs to 2D and 3D image data from particle physics experiments to classify signal from background. In this work we present an extensive convolutional neural architecture search, achieving high accuracy for signal/background discrimination for a HEP classification use-case based on simulated data from the Ice Cube neutrino observatory and an ATLAS-like detector. We demonstrate among other things that we can achieve the same accuracy as complex ResNet architectures with CNNs with less parameters, and present comparisons of computational requirements, training and inference times. [ABSTRACT FROM AUTHOR]

The analysis of 2D scattering maps generated in scatterometry experiments for detection and classification of nanoparticle on surfaces is a cumbersome and slow process. Recently, deep learning techniques have been adopted to avoid manual feature extraction and classification in many research and application areas, including optics. In the present work, we collected experimental dataset of nanoparticles deposited on wafers for four different classes of polystyrene particles (with diameters of 40, 50, 60, 80 nm) plus background (no particles) class. We trained a convolutional neural network, including its architecture optimization, and achieved 95% accurate results. We compared the performance of this network to a existing method based on line-by-line search and thresholding, demonstrating up to a twofold enhanced performance in particle classification. The network is extended by a supervisor layer that can reject up to 80% of the fooling images at the cost of only rejecting 10% of original data. [ABSTRACT FROM AUTHOR]