Showing papers by "General Electric published in 2018"

Single-Shot Refinement Neural Network for Object Detection

Abstract: For object detection, the two-stage approach (e.g., Faster R-CNN) has been achieving the highest accuracy, whereas the one-stage approach (e.g., SSD) has the advantage of high efficiency. To inherit the merits of both while overcoming their disadvantages, in this paper, we propose a novel single-shot based detector, called RefineDet, that achieves better accuracy than two-stage methods and maintains comparable efficiency of one-stage methods. RefineDet consists of two inter-connected modules, namely, the anchor refinement module and the object detection module. Specifically, the former aims to (1) filter out negative anchors to reduce search space for the classifier, and (2) coarsely adjust the locations and sizes of anchors to provide better initialization for the subsequent regressor. The latter module takes the refined anchors as the input from the former to further improve the regression accuracy and predict multi-class label. Meanwhile, we design a transfer connection block to transfer the features in the anchor refinement module to predict locations, sizes and class labels of objects in the object detection module. The multitask loss function enables us to train the whole network in an end-to-end way. Extensive experiments on PASCAL VOC 2007, PASCAL VOC 2012, and MS COCO demonstrate that RefineDet achieves state-of-the-art detection accuracy with high efficiency. Code is available at https://github.com/sfzhang15/RefineDet.

Identifying the Best Machine Learning Algorithms for Brain Tumor Segmentation, Progression Assessment, and Overall Survival Prediction in the BRATS Challenge

Abstract: Gliomas are the most common primary brain malignancies, with different degrees of aggressiveness, variable prognosis and various heterogeneous histologic sub-regions, i.e., peritumoral edematous/invaded tissue, necrotic core, active and non-enhancing core. This intrinsic heterogeneity is also portrayed in their radio-phenotype, as their sub-regions are depicted by varying intensity profiles disseminated across multi-parametric magnetic resonance imaging (mpMRI) scans, reflecting varying biological properties. Their heterogeneous shape, extent, and location are some of the factors that make these tumors difficult to resect, and in some cases inoperable. The amount of resected tumoris a factor also considered in longitudinal scans, when evaluating the apparent tumor for potential diagnosis of progression. Furthermore, there is mounting evidence that accurate segmentation of the various tumor sub-regions can offer the basis for quantitative image analysis towards prediction of patient overall survival. This study assesses thestate-of-the-art machine learning (ML) methods used for brain tumor image analysis in mpMRI scans, during the last seven instances of the International Brain Tumor Segmentation (BraTS) challenge, i.e., 2012-2018. Specifically, we focus on i) evaluating segmentations of the various glioma sub-regions in pre-operative mpMRI scans, ii) assessing potential tumor progression by virtue of longitudinal growth of tumor sub-regions, beyond use of the RECIST/RANO criteria, and iii) predicting the overall survival from pre-operative mpMRI scans of patients that underwent gross tota lresection. Finally, we investigate the challenge of identifying the best ML algorithms for each of these tasks, considering that apart from being diverse on each instance of the challenge, the multi-institutional mpMRI BraTS dataset has also been a continuously evolving/growing dataset.

Mn2+ and Mn4+ red phosphors: synthesis, luminescence and applications in WLEDs. A review

Abstract: Transition-metal activated phosphors are an important family of luminescent materials that can produce white light with an outstanding color rendering index and correlated color temperature for use in light-emitting diodes. In recent years, work in this quite “hot” research field has focused on the development of Mn2+ and Mn4+ activated red phosphors. In this review article, we provide an overview of recent studies on Mn2+ and Mn4+ doped phosphors, including detailed synthesis routes (solid-state reaction and wet-chemical synthesis) and description of luminescence mechanisms and phosphors’ behaviors; discuss their promising applications in white light-emitting diodes; and present an extensive list of references to representative works in this field.

Learning from Synthetic Data: Addressing Domain Shift for Semantic Segmentation

Abstract: Visual Domain Adaptation is a problem of immense importance in computer vision. Previous approaches showcase the inability of even deep neural networks to learn informative representations across domain shift. This problem is more severe for tasks where acquiring hand labeled data is extremely hard and tedious. In this work, we focus on adapting the representations learned by segmentation networks across synthetic and real domains. Contrary to previous approaches that use a simple adversarial objective or superpixel information to aid the process, we propose an approach based on Generative Adversarial Networks (GANs) that brings the embeddings closer in the learned feature space. To showcase the generality and scalability of our approach, we show that we can achieve state of the art results on two challenging scenarios of synthetic to real domain adaptation. Additional exploratory experiments show that our approach: (1) generalizes to unseen domains and (2) results in improved alignment of source and target distributions.

Review of the operational flexibility and emissions of gas- and coal-fired power plants in a future with growing renewables

Abstract: This paper reviews operational flexibility and emissions of gas- and coal-fired power plants today and in the future with higher renewables. Six study cases were considered: heavy duty gas turbines in simple and combined cycle, aero-derivative gas turbines, large-scale supercritical coal power plants and small- and mid-scale sub-critical coal power plants. The most critical operational processes and pollutants associated with these plants were identified. Then, data was collected mainly from manufacturers, but also from academic research and grey literature. The data was compared and analyzed. Detailed comparisons of the power plant characteristics as well as the current and future flexibility and emissions are provided. Furthermore, a method to quantify the ability of conventional power plants to back-up renewables and the expected benefits from improved flexibility is proposed and evaluated. Results show that gas-fired power plants are not only more efficient, but also faster and generally less polluting than coal-fired power plants. However, at their respective minimum complaint load, gas plants are less flexible and produced more NOx and CO emissions than coal-fired power plants. Results also show that on average, an improvement of approximately 50% to 100% on power ramp rates, minimum power load, number of major power cycles and emissions for these plants is sought in the future to complement renewables.

Occlusion-aware R-CNN: Detecting Pedestrians in a Crowd

Abstract: Pedestrian detection in crowded scenes is a challenging problem since the pedestrians often gather together and occlude each other In this paper, we propose a new occlusion-aware R-CNN (OR-CNN) to improve the detection accuracy in the crowd Specifically, we design a new aggregation loss to enforce proposals to be close and locate compactly to the corresponding objects Meanwhile, we use a new part occlusion-aware region of interest (PORoI) pooling unit to replace the RoI pooling layer in order to integrate the prior structure information of human body with visibility prediction into the network to handle occlusion Our detector is trained in an end-to-end fashion, which achieves state-of-the-art results on three pedestrian detection datasets, ie, CityPersons, ETH, and INRIA, and performs on-pair with the state-of-the-arts on Caltech

DCAN: Dual Channel-Wise Alignment Networks for Unsupervised Scene Adaptation

Abstract: Harvesting dense pixel-level annotations to train deep neural networks for semantic segmentation is extremely expensive and unwieldy at scale. While learning from synthetic data where labels are readily available sounds promising, performance degrades significantly when testing on novel realistic data due to domain discrepancies. We present Dual Channel-wise Alignment Networks (DCAN), a simple yet effective approach to reduce domain shift at both pixel-level and feature-level. Exploring statistics in each channel of CNN feature maps, our framework performs channel-wise feature alignment, which preserves spatial structures and semantic information, in both an image generator and a segmentation network. In particular, given an image from the source domain and unlabeled samples from the target domain, the generator synthesizes new images on-the-fly to resemble samples from the target domain in appearance and the segmentation network further refines high-level features before predicting semantic maps, both of which leverage feature statistics of sampled images from the target domain. Unlike much recent and concurrent work relying on adversarial training, our framework is lightweight and easy to train. Extensive experiments on adapting models trained on synthetic segmentation benchmarks to real urban scenes demonstrate the effectiveness of the proposed framework.

Zero-Echo-Time and Dixon Deep Pseudo-CT (ZeDD CT): Direct Generation of Pseudo-CT Images for Pelvic PET/MRI Attenuation Correction Using Deep Convolutional Neural Networks with Multiparametric MRI

Abstract: Accurate quantification of uptake on PET images depends on accurate attenuation correction in reconstruction. Current MR-based attenuation correction methods for body PET use a fat and water map derived from a 2-echo Dixon MRI sequence in which bone is neglected. Ultrashort-echo-time or zero-echo-time (ZTE) pulse sequences can capture bone information. We propose the use of patient-specific multiparametric MRI consisting of Dixon MRI and proton-density–weighted ZTE MRI to directly synthesize pseudo-CT images with a deep learning model: we call this method ZTE and Dixon deep pseudo-CT (ZeDD CT). Methods: Twenty-six patients were scanned using an integrated 3-T time-of-flight PET/MRI system. Helical CT images of the patients were acquired separately. A deep convolutional neural network was trained to transform ZTE and Dixon MR images into pseudo-CT images. Ten patients were used for model training, and 16 patients were used for evaluation. Bone and soft-tissue lesions were identified, and the SUVmax was measured. The root-mean-squared error (RMSE) was used to compare the MR-based attenuation correction with the ground-truth CT attenuation correction. Results: In total, 30 bone lesions and 60 soft-tissue lesions were evaluated. The RMSE in PET quantification was reduced by a factor of 4 for bone lesions (10.24% for Dixon PET and 2.68% for ZeDD PET) and by a factor of 1.5 for soft-tissue lesions (6.24% for Dixon PET and 4.07% for ZeDD PET). Conclusion: ZeDD CT produces natural-looking and quantitatively accurate pseudo-CT images and reduces error in pelvic PET/MRI attenuation correction compared with standard methods.

A deep learning-based algorithm for 2-D cell segmentation in microscopy images

Abstract: Automatic and reliable characterization of cells in cell cultures is key to several applications such as cancer research and drug discovery. Given the recent advances in light microscopy and the need for accurate and high-throughput analysis of cells, automated algorithms have been developed for segmenting and analyzing the cells in microscopy images. Nevertheless, accurate, generic and robust whole-cell segmentation is still a persisting need to precisely quantify its morphological properties, phenotypes and sub-cellular dynamics. We present a single-channel whole cell segmentation algorithm. We use markers that stain the whole cell, but with less staining in the nucleus, and without using a separate nuclear stain. We show the utility of our approach in microscopy images of cell cultures in a wide variety of conditions. Our algorithm uses a deep learning approach to learn and predict locations of the cells and their nuclei, and combines that with thresholding and watershed-based segmentation. We trained and validated our approach using different sets of images, containing cells stained with various markers and imaged at different magnifications. Our approach achieved a 86% similarity to ground truth segmentation when identifying and separating cells. The proposed algorithm is able to automatically segment cells from single channel images using a variety of markers and magnifications.

VisDrone-DET2018: The Vision Meets Drone Object Detection in Image Challenge Results

Abstract: Object detection is a hot topic with various applications in computer vision, e.g., image understanding, autonomous driving, and video surveillance. Much of the progresses have been driven by the availability of object detection benchmark datasets, including PASCAL VOC, ImageNet, and MS COCO. However, object detection on the drone platform is still a challenging task, due to various factors such as view point change, occlusion, and scales. To narrow the gap between current object detection performance and the real-world requirements, we organized the Vision Meets Drone (VisDrone2018) Object Detection in Image challenge in conjunction with the 15th European Conference on Computer Vision (ECCV 2018). Specifically, we release a large-scale drone-based dataset, including 8, 599 images (6, 471 for training, 548 for validation, and 1, 580 for testing) with rich annotations, including object bounding boxes, object categories, occlusion, truncation ratios, etc. Featuring a diverse real-world scenarios, the dataset was collected using various drone models, in different scenarios (across 14 different cities spanned over thousands of kilometres), and under various weather and lighting conditions. We mainly focus on ten object categories in object detection, i.e., pedestrian, person, car, van, bus, truck, motor, bicycle, awning-tricycle, and tricycle. Some rarely occurring special vehicles (e.g., machineshop truck, forklift truck, and tanker) are ignored in evaluation. The dataset is extremely challenging due to various factors, including large scale and pose variations, occlusion, and clutter background. We present the evaluation protocol of the VisDrone-DET2018 challenge and the comparison results of 38 detectors on the released dataset, which are publicly available on the challenge website: http://www.aiskyeye.com/. We expect the challenge to largely boost the research and development in object detection in images on drone platforms.

Identification of cytokine-specific sensory neural signals by decoding murine vagus nerve activity.

Abstract: The nervous system maintains physiological homeostasis through reflex pathways that modulate organ function. This process begins when changes in the internal milieu (e.g., blood pressure, temperature, or pH) activate visceral sensory neurons that transmit action potentials along the vagus nerve to the brainstem. IL-1β and TNF, inflammatory cytokines produced by immune cells during infection and injury, and other inflammatory mediators have been implicated in activating sensory action potentials in the vagus nerve. However, it remains unclear whether neural responses encode cytokine-specific information. Here we develop methods to isolate and decode specific neural signals to discriminate between two different cytokines. Nerve impulses recorded from the vagus nerve of mice exposed to IL-1β and TNF were sorted into groups based on their shape and amplitude, and their respective firing rates were computed. This revealed sensory neural groups responding specifically to TNF and IL-1β in a dose-dependent manner. These cytokine-mediated responses were subsequently decoded using a Naive Bayes algorithm that discriminated between no exposure and exposures to IL-1β and TNF (mean successful identification rate 82.9 ± 17.8%, chance level 33%). Recordings obtained in IL-1 receptor-KO mice were devoid of IL-1β-related signals but retained their responses to TNF. Genetic ablation of TRPV1 neurons attenuated the vagus neural signals mediated by IL-1β, and distal lidocaine nerve block attenuated all vagus neural signals recorded. The results obtained in this study using the methodological framework suggest that cytokine-specific information is present in sensory neural signals within the vagus nerve.

Occlusion-aware R-CNN: Detecting Pedestrians in a Crowd

Abstract: Pedestrian detection in crowded scenes is a challenging problem since the pedestrians often gather together and occlude each other. In this paper, we propose a new occlusion-aware R-CNN (OR-CNN) to improve the detection accuracy in the crowd. Specifically, we design a new aggregation loss to enforce proposals to be close and locate compactly to the corresponding objects. Meanwhile, we use a new part occlusion-aware region of interest (PORoI) pooling unit to replace the RoI pooling layer in order to integrate the prior structure information of human body with visibility prediction into the network to handle occlusion. Our detector is trained in an end-to-end fashion, which achieves state-of-the-art results on three pedestrian detection datasets, i.e., CityPersons, ETH, and INRIA, and performs on-pair with the state-of-the-arts on Caltech.

A wearable patch for continuous monitoring of sweat electrolytes during exertion

Abstract: Implementation of wearable sweat sensors for continuous measurement of fluid based biomarkers (including electrolytes, metabolites and proteins) is an attractive alternative to common, yet intrusive and invasive, practices such as urine or blood analysis. Recent years have witnessed several key demonstrations of sweat based electrochemical sensing in wearable formats, however, there are still significant challenges and opportunities in this space for clinical acceptance, and thus mass implementation of these devices. For instance, there are inherent challenges in establishing direct correlations between sweat-based and gold-standard plasma-based biomarker concentrations for clinical decision-making. In addition, the wearable sweat monitoring devices themselves may exacerbate these challenges, as they can significantly alter sweat physiology (example, sweat rate and composition). Reported here is the demonstration of a fully integrated, wireless, wearable and flexible sweat sensing device for non-obtrusive and continuous monitoring of electrolytes during moderate to intense exertion as a metric for hydration status. The focus of this work is twofold: 1- design of a conformable fluidics systems to suit conditions of operation for sweat collection (to minimize sensor lag) with rapid removal of sweat from the sensing site (to minimize effects on sweat physiology). 2- integration of Na+ and K+ ion-selective electrodes (ISEs) with flexible microfluidics and low noise small footprint electronics components to enable wireless, wearable sweat monitoring. While this device is specific to electrolyte analysis during intense perspiration, the lessons in microfluidics and overall system design are likely applicable across a broad range of analytes.

Variable-Density Single-Shot Fast Spin-Echo MRI with Deep Learning Reconstruction by Using Variational Networks.

Abstract: Purpose To develop a deep learning reconstruction approach to improve the reconstruction speed and quality of highly undersampled variable-density single-shot fast spin-echo imaging by using a variational network (VN), and to clinically evaluate the feasibility of this approach. Materials and Methods Imaging was performed with a 3.0-T imager with a coronal variable-density single-shot fast spin-echo sequence at 3.25 times acceleration in 157 patients referred for abdominal imaging (mean age, 11 years; range, 1-34 years; 72 males [mean age, 10 years; range, 1-26 years] and 85 females [mean age, 12 years; range, 1-34 years]) between March 2016 and April 2017. A VN was trained based on the parallel imaging and compressed sensing (PICS) reconstruction of 130 patients. The remaining 27 patients were used for evaluation. Image quality was evaluated in an independent blinded fashion by three radiologists in terms of overall image quality, perceived signal-to-noise ratio, image contrast, sharpness, and residual artifacts with scores ranging from 1 (nondiagnostic) to 5 (excellent). Wilcoxon tests were performed to test the hypothesis that there was no significant difference between VN and PICS. Results VN achieved improved perceived signal-to-noise ratio (P = .01) and improved sharpness (P < .001), with no difference in image contrast (P = .24) and residual artifacts (P = .07). In terms of overall image quality, VN performed better than did PICS (P = .02). Average reconstruction time ± standard deviation was 5.60 seconds ± 1.30 per section for PICS and 0.19 second ± 0.04 per section for VN. Conclusion Compared with the conventional parallel imaging and compressed sensing reconstruction (PICS), the variational network (VN) approach accelerates the reconstruction of variable-density single-shot fast spin-echo sequences and achieves improved overall image quality with higher perceived signal-to-noise ratio and sharpness. © RSNA, 2018 Online supplemental material is available for this article.

Electricity Price Forecasting for Operational Scheduling of Behind-the-Meter Storage Systems

Abstract: Electricity price forecast plays a key role in strategic behavior of participants in competitive electricity markets. With the growth of behind-the-meter energy storage, price forecasting becomes important in energy management and control of such small-scale storage systems. In this paper, a forecasting strategy is proposed for real-time electricity markets using publicly available market data. The proposed strategy uses high-resolution data along with hourly data as inputs of two separate forecasting models with different forecast horizons. Moreover, an intra-hour rolling horizon framework is proposed to provide accurate updates on price predictions. The proposed forecasting strategy has the capability to detect price spikes and capture severe price variations. The real data from Ontario’s electricity market is used to evaluate the performance of the proposed forecasting strategy from the statistical point of view. The generated price forecasts are also applied to an optimization platform for operation scheduling of a battery energy storage system within a grid-connected micro-grid in Ontario to show the value of the proposed strategy from an economic perspective.

Control of Luminescence by Tuning of Crystal Symmetry and Local Structure in Mn4+-Activated Narrow Band Fluoride Phosphors

Abstract: Mn4+ -doped fluoride phosphors have been widely used in wide-gamut backlighting devices because of their extremely narrow emission band. Solid solutions of Na2 (Six Ge1-x )F6 :Mn4+ and Na2 (Gey Ti1-y )F6 :Mn4+ were successfully synthesized to elucidate the behavior of the zero-phonon line (ZPL) in different structures. The ratio between ZPL and the highest emission intensity υ6 phonon sideband exhibits a strong relationship with luminescent decay rate. First-principles calculations are conducted to model the variation in the structural and electronic properties of the prepared solid solutions as a function of the composition. To compensate for the limitations of the Rietveld refinement, electron paramagnetic resonance and high-resolution steady-state emission spectra are used to confirm the diverse local environment for Mn4+ in the structure. Finally, the spectral luminous efficacy of radiation (LER) is used to reveal the important role of ZPL in practical applications.

Zero TE-based pseudo-CT image conversion in the head and its application in PET/MR attenuation correction and MR-guided radiation therapy planning.

Abstract: Purpose: To describe a method for converting Zero TE (ZTE) MR images into Xray attenuation information in the form of pseudo-CT images and demonstrate its performance for (1) attenuation correction ...

A Megawatt-Scale Medium-Voltage High Efficiency High Power Density “SiC+Si” Hybrid Three-Level ANPC Inverter for Aircraft Hybrid-Electric Propulsion Systems

Abstract: Hybrid-electric propulsion system is an enabling technology to make the aircrafts more fuel-saving, quieter, and lower carbide emission. In this paper, a megawatt-scale power inverter based on a three-level active neutral-point-clamped (3L-ANPC) topology will be developed. To achieve high efficiency, the switching devices operating at carrier frequency in the power converter are configured by the emerging Silicon Carbide (SiC) Metal-Oxide Semiconductor Field-Effect Transistors (MOSFETs), while the conventional Silicon (Si) Insulated-Gate Bipolar Transistors (IGBTs) are selected for switches operating at the fundamental output frequency. To reduce system cable weight, the dc-bus voltage is increased to 2.4 kV. Unlike the conventional 400 Hz aircraft electric systems, the rated fundamental output frequency here is boosted to 1.4 kHz to drive the high-speed motor, which can also reduce system weight. Main hardware development and control modulation strategies are presented. Experimental results are presented to verify the performance of this MW-scale medium-voltage “SiC+Si” hybrid three-level ANPC inverter. It is shown that the 1-MW 3L-ANPC inverter can achieve a high efficiency of 99% and high power density of 12 kVA/kg.

Separable and Localized System-Level Synthesis for Large-Scale Systems

Abstract: A major challenge faced in the design of large-scale cyber-physical systems, such as power systems, the Internet of Things or intelligent transportation systems, is that traditional distributed optimal control methods do not scale gracefully, neither in controller synthesis nor in controller implementation, to systems composed of a large number (e.g., on the order of billions) of interacting subsystems. This paper shows that this challenge can now be addressed by leveraging the recently introduced system-level approach (SLA) to controller synthesis. In particular, in the context of the SLA, we define suitable notions of separability for control objective functions and system constraints such that the global optimization problem (or iterate update problems of a distributed optimization algorithm) can be decomposed into parallel subproblems. We then further show that if additional locality (i.e., sparsity) constraints are imposed, then these subproblems can be solved using local models and local decision variables. The SLA is essential to maintain the convexity of the aforementioned problems under locality constraints. As a consequence, the resulting synthesis methods have $O(1)$ complexity relative to the size of the global system. We further show that many optimal control problems of interest, such as (localized) LQR and LQG, $\mathcal {H}_2$ optimal control with joint actuator and sensor regularization, and (localized) mixed $\mathcal {H}_2/\mathcal {L}_1$ optimal control problems, satisfy these notions of separability, and use these problems to explore tradeoffs in performance, actuator, and sensing density, and average versus worst-case performance for a large-scale power inspired system.

Peripheral nerve diffusion tensor imaging: Overview, pitfalls, and future directions

Abstract: Diffusion tensor imaging (DTI) is a noninvasive magnetic resonance imaging (MRI) technique that measures the extent of restricted water diffusion and anisotropy in biological tissue. Although DTI has been widely applied in the brain, more recently researchers have used it to characterize nerve pathology in the setting of entrapment neuropathy, traumatic injury, and tumor. DTI artifacts are exacerbated when imaging off isocenter in the body. Anecdotally, the most significant artifacts in peripheral nerve DTI include magnetic field inhomogeneity, motion, incomplete fat suppression, aliasing, and distortion. High spatial resolution is also required to reliably evaluate smaller peripheral nerves. This article provides an overview of such technical issues, particularly when trying to apply DTI in the clinical setting, and offers potential solutions. LEVEL OF EVIDENCE 5 Technical Efficacy: Stage 3 J. Magn. Reson. Imaging 2018;47:1171-1189.

Spectroscopy of Mn4+ in Double Perovskites, La2LiSbO6 and La2MgTiO6: Deep Red Photon Generators for Agriculture LEDs

Abstract: aGE Global Research, One Research Circle, Niskayuna, New York 12309, USA bCollege of Sciences, Chongqing University of Posts and Telecommunications, Chongqing 400065, People’s Republic of China cInstitute of Physics, University of Tartu, Tartu 50411, Estonia dInstitute of Physics, Jan Dlugosz University, Armii Krajowej 13/15, PL-42200 Czestochowa, Poland eCurrent & Lighting, Current Powered by GE, Cleveland, Ohio 44112, USA fLighting Enabled Systems & Applications ERC, Rensselaer Polytechnic Institute, Troy, New York 12180, USA

Critical Review—A Review of the Electronic Structure and Optical Properties of Ions with d3 Electron Configuration (V2+, Cr3+, Mn4+, Fe5+) and Main Related Misconceptions

Abstract: Tetravalent manganese ions with the 3d3 electronic configuration have recently began to play a major role as a red photon generator in the LED based lighting and display devices. The goal of this article is to tutorize the fundamental optical properties of Mn4+ and to clear up some common misconceptions that we have encountered in the archival literature that pertains to the spectroscopic properties of this ion and other ions with the same electron configuration. The methods to properly interpret the optical spectra of these systems are also discussed. This is accomplished by presenting a discussion on the electronic properties of the free d3 ions and the changes which occur when the ion is introduced in a crystalline lattice. It is hoped that such systematic presentation of spectroscopic properties of the d3 ions and their variation from the free state to the crystalline solids will be useful for many researchers – mainly experimentalists – actively working in the field and will help them avoiding many mistakes when presenting their experimental results.

VisDrone-VDT2018: The Vision Meets Drone Video Detection and Tracking Challenge Results

Abstract: Drones equipped with cameras have been fast deployed to a wide range of applications, such as agriculture, aerial photography, fast delivery, and surveillance. As the core steps in those applications, video object detection and tracking attracts much research effort in recent years. However, the current video object detection and tracking algorithms are not usually optimal for dealing with video sequences captured by drones, due to various challenges, such as viewpoint change and scales. To promote and track the development of the detection and tracking algorithms with drones, we organized the Vision Meets Drone Video Detection and Tracking (VisDrone-VDT2018) challenge, which is a subtrack of the Vision Meets Drone 2018 challenge workshop in conjunctiohe 15th European Conference on Computer Vision (ECCV 2018). Specifically, this workshop challenge consists of two tasks, (1) video object detection, and (2) multi-object tracking. We present a large-scale video object detection and tracking dataset, which consists of 79 video clips with about 1.5 million annotated bounding boxes in 33, 366 frames. We also provide rich annotations, including object categories, occlusion, and truncation ratios for better data usage. Being the largest such dataset ever published, the challenge enables extensive evaluation, investigation and tracking the progress of object detection and tracking algorithms on the drone platform. We present the evaluation protocol of the VisDrone-VDT2018 challenge and the results of the algorithms on the benchmark dataset, which are publicly available on the challenge website: http://www.aiskyeye.com/. We hope the challenge largely boost the research and development in related fields.

Impact of GPS Signal Loss and Its Mitigation in Power System Synchronized Measurement Devices

Abstract: With the aid of global positioning system (GPS), synchronized measurement devices (SMDs) are increasingly deployed across power systems to monitor the status of electric grids by providing accurate measurement data along with unified time stamps. Unfortunately, GPS receivers tend to lose signal lock when certain uncontrollable and unpredictable factors arise. In order to investigate the presence of GPS signal loss (GSL) issues on measurement devices, analysis is performed on historical data from both phasor data concentrators and FNET/GridEye servers. Meanwhile, the impact of GSL on field measurement accuracy has not been previously explored in depth. Through analysis and experimental tests, this paper discovers angle drift caused by GSL, which consequently leads to the total vector error exceeding the IEEE standard C37.118.1-2011. Furthermore, a compensation method is proposed to rectify the angle drift and laboratory experiments demonstrate that the proposed method does effectively reduce angle drift and mitigate the impact of GSL in SMDs.

A data-level fusion approach for degradation modeling and prognostic analysis under multiple failure modes

Abstract: Operating units, in practice, often suffer from multiple modes of failure, and each failure mode has a distinct influence on the service life cycle path of a unit. The rapid development of sensor and communication technologies has enabled multiple sensors to simultaneously monitor and track the health status of a unit in real time. However, one challenging question that remains to be resolved is how to leverage data from multiple sensors for better degradation modeling and prognostic analysis, especially when there are multiple failure modes. Currently, many of the existing approaches in prognostics either (a) fail to capture the dependency between sensors and instead focus on analyzing each sensor independently or (b) fail to incorporate the failure-mode diagnosis for better degradation modeling and prognostics during condition monitoring. To address the limitations in the existing literature, we propose a data-level fusion methodology to construct a composite failure-mode index, named FM-INDEX, ...

Enabling Large Superalloy Parts Using Compact Coprecipitation of γ′ and γ′′

Abstract: Next-generation gas turbines will require disk materials capable of operating at 923 K (650 °C) and above to achieve efficiencies well beyond today’s 62 pct benchmark. This temperature requirement marks a critical turning point in materials selection. Current turbine disk alloys, such as 706 and 718, are limited by the stability of their major strengthening phase, γ′′, which coarsens rapidly beyond 923 K (650 °C) resulting in significant degradation in properties. More capable γ′ strengthened superalloys, such as those used in jet engine disks, are also limited due to the sheer size of gas turbine hardware; the γ′ phase overages during the slow cooling rates inherent in processing thick-section parts. In the present work, we address this fundamental gap in available superalloy materials. Through careful control of Al, Ti, and Nb levels, we show that fine (<100 nm) γ′ and compact γ′/γ′′ coprecipitate structures can be formed even under extremely slow cooling rates from high temperature. The presence of Ti is shown to have a dominant effect on phase formation, dictating whether γ′, γ′/γ′′ coprecipitates, or other less desirable acicular phases form on cooling. Sensitivity to cooling rate and aging heat treatment is also explored. A custom phase field model along with commercial precipitation kinetics software is used to better understand the phase evolution and stability of compact coprecipitates. The alloying strategies discussed here enable a new class of superalloys suitable for applications requiring large parts operating at high temperature.

SiC Trench MOSFET With Integrated Self-Assembled Three-Level Protection Schottky Barrier Diode

Abstract: A silicon carbide (SiC) trench MOSFET (TMOS) with integrated three-level protection (TLP) Schottky barrier diode (SBD), named ITS-TMOS, is proposed and investigated by simulation. The device features the integrated TLP-SBD that remarkably improves body diode characteristics while guarantees excellent fundamental performance of TMOS. In the blocking state, the P-base region, the trench gate, and the P+ shield at the trench bottom serve as the TLP of the Schottky contact. Each protection assists in depleting the drift region beneath Schottky contact. Benefiting from the self-assembled TLP, the leakage current of the integrated body diode of the ITS-TMOS is significantly reduced. Moreover, the reverse turn-on voltage ( ${V} _{ \mathrm{\scriptscriptstyle ON}}$ ) and the gate charge ( ${Q} _{g}$ ) of the ITS-TMOS are 65% and 18% lower than those of the conventional TMOS, respectively. The improved overall performances make the SiC ITS-TMOS a competitive candidate for high-efficiency and high power density applications.

VisDrone-SOT2018: The Vision Meets Drone Single-Object Tracking Challenge Results

Abstract: Single-object tracking, also known as visual tracking, on the drone platform attracts much attention recently with various applications in computer vision, such as filming and surveillance. However, the lack of commonly accepted annotated datasets and standard evaluation platform prevent the developments of algorithms. To address this issue, the Vision Meets Drone Single-Object Tracking (VisDrone-SOT2018) Challenge workshop was organized in conjunction with the 15th European Conference on Computer Vision (ECCV 2018) to track and advance the technologies in such field. Specifically, we collect a dataset, including 132 video sequences divided into three non-overlapping sets, i.e., training (86 sequences with 69, 941 frames), validation (11 sequences with 7, 046 frames), and testing (35 sequences with 29, 367 frames) sets. We provide fully annotated bounding boxes of the targets as well as several useful attributes, e.g., occlusion, background clutter, and camera motion. The tracking targets in these sequences include pedestrians, cars, buses, and animals. The dataset is extremely challenging due to various factors, such as occlusion, large scale, pose variation, and fast motion. We present the evaluation protocol of the VisDrone-SOT2018 challenge and the results of a comparison of 22 trackers on the benchmark dataset, which are publicly available on the challenge website: http://www.aiskyeye.com/. We hope this challenge largely boosts the research and development in single object tracking on drone platforms.