Suchergebnisse

Synthetic simulated environments are gaining popularity in the Deep Learning Era, as they can alleviate the effort and cost of two critical tasks to build multi-camera systems for surveillance applications: setting up the camera system to cover the use cases and generating the labeled dataset to train the required Deep Neural Networks (DNNs). However, there are no simulated environments ready to solve them for all kind of scenarios and use cases. Typically, 'ad hoc' environments are built, which cannot be easily applied to other contexts. In this work we present a methodology to build synthetic simulated environments with sufficient generality to be usable in different contexts, with little effort. Our methodology tackles the challenges of the appropriate parameterization of scene configurations, the strategies to generate randomly a wide and balanced range of situations of interest for training DNNs with synthetic data, and the quick image capturing from virtual cameras considering the rendering bottlenecks. We show a practical implementation example for the detection of incorrectly placed luggage in aircraft cabins, including the qualitative and quantitative analysis of the data generation process and its influence in a DNN training, and the required modifications to adapt it to other surveillance contexts. ; This work has received funding from the Clean Sky 2 Joint Undertaking under the European Union's Horizon 2020 research and innovation program under grant agreement No. 865162, SmaCS (https://www.smacs.eu/)

Technology advancement has facilitated digital content, such as images, being acquired in large volumes. However, requirement from the privacy or legislation perspective still demands the need for intellectual content protection. In this paper, we propose a deep neural network (DNN) based watermarking method to achieve this goal. Instead of training a neural network for protecting a specific image, we train the network on an image dataset and generalize the trained model to protect distinct test images in a bulk manner. Respective evaluations from both the subjective and objective aspects confirm the generality and practicality of our proposed method. To demonstrate the robustness of this general neural watermarking approach, commonly used attacks are applied to the watermarked images to examine the corresponding extracted watermarks, which still retain sufficient recognizable traits for some occasions. Testing on distinctive dataset shows the satisfying generalization of our proposed method, and practice such as loss function adjustment can cater to the capacity requirement of complicated watermark. We also discuss some traits of the trained model, which incur the vulnerability to JPEG compression attack. However, remedy seeking for this can potentially open a window to understand the underlying working principle of DNN in future work. Considering its performance and economy, it is concluded that subsequent studies that generalize our work on utilizing DNN for intellectual content protection might be a promising research trend.

Conventional single-channel speech separation has two long-standing issues. The first issue, over-smoothing,
is addressed, and estimated signals are used to expand the training data set. Second, DNN generates prior knowledge to address the problem of incomplete separation and mitigate speech distortion. To overcome all current issues, we suggest employing an efficient optimal reconstruction-based speech separation (ERSS) to overcome those problems using a hybrid deep learning technique. First, we propose an integral fox ride optimization (IFRO) algorithm for spectral structure reconstruction with the help of multiple spectrum features: time dynamic information, binaural and mono features. Second, we introduce a hybrid retrieval-based deep neural network (RDNN) to reconstruct the spectrograms size of speech and noise directly. The input signals are sent to Short Term Fourier Transform (STFT).
STFT converts a clean input signal into spectrograms then uses a feature extraction technique called IFRO to extract features from spectrograms. After extracting the features, using the RDNN classification algorithm, the classified features are converted to softmax. ISTFT then applies to softmax and correctly separates speech signals. Experiments show that our proposed method achieves the highest gains in SDR, SIR, SAR STIO, and PESQ outcomes of 10.9, 15.3, 10.8, 0.08, and 0.58, respectively. The Joint-DNN-SNMF obtains 9.6, 13.4, 10.4, 0.07, and 0.50, comparable to the Joint-DNN-SNMF. The proposed result is compared to a different method and some previous work. In comparison to previous research, our proposed methodology yields better results.

The current literature on international joint ventures pays considerable attention to why joint ventures are established or why they are dissolved, but we lack studies that give insight into their dynamic development. The aim of this article is to investigate the evolution of an international joint venture over time. We confront some of the theoretical insights developed during the last decades with the dramatic history of the aluminum producer Det Norske Nitridaktieselskap (DNN). The company was established shortly beforeWorld War I and was finally disbanded over seventy years later. For most of this time, DNN was an international joint venture with shifting ownership configurations. This gives us the possibility not only to discuss the motivation for why the company was established or why it was dissolved, but also to study the mid-life of the company. What was DNN's role within the general corporate strategies of the owners? Did this role change over time?

Abstract. Given trends in more frequent and severe natural disaster events, developing effective risk mitigation strategies is crucial to reduce negative economic impacts, due to the limited budget for rehabilitation. To address this need, this study aims to develop a strategic framework for natural disaster risk mitigation, highlighting two different strategic implementation processes (SIPs). SIP-1 is intended to improve the predictability of natural disaster-triggered financial losses using deep learning. To demonstrate SIP-1, SIP-1 explores deep neural networks (DNNs) that learn storm and flood insurance loss ratios associated with selected major indicators and then develops an optimal DNN model. SIP-2 underlines the risk mitigation strategy at the project level, by adopting a cost–benefit analysis method that quantifies the cost effectiveness of disaster prevention projects. In SIP-2, a case study of disaster risk reservoir projects in South Korea was adopted. The validated result of SIP-1 confirmed that the predictability of the developed DNN is more accurate and reliable than a traditional parametric model, while SIP-2 revealed that maintenance projects are economically more beneficial in the long term as the loss amount becomes smaller after 8 years, coupled with the investment in the projects. The proposed framework is unique as it provides a combinational approach to mitigating economic damages caused by natural disasters at both financial loss and project levels. This study is its first kind and will help practitioners quantify the loss from natural disasters, while allowing them to evaluate the cost effectiveness of risk reduction projects through a holistic approach.

Objectives: Suicide risk is elevated in lesbian, gay, bisexual, and transgender (LGBT) individuals. Limited data on LGBT status in healthcare systems hinder our understanding of this risk. This study used natural language processing to extract LGBT status and a deep neural network (DNN) to examine suicidal death risk factors among US Veterans.Methods: Data on 8.8 million veterans with visits between 2010 and 2017 was used. A case-control study was performed, and suicide death risk was analyzed by a DNN. Feature impacts and interactions on the outcome were evaluated.Results: The crude suicide mortality rate was higher in LGBT patients. However, after adjusting for over 200 risk and protective factors, known LGBT status was associated with reduced risk compared to LGBT-Unknown status. Among LGBT patients, black, female, married, and older Veterans have a higher risk, while Veterans of various religions have a lower risk.Conclusion: Our results suggest that disclosed LGBT status is not directly associated with an increase suicide death risk, however, other factors (e.g., depression and anxiety caused by stigma) are associated with suicide death risks.

Benefiting from the advancement of algorithms in massive data and powerful computing resources, deep learning has been explored in a wide variety of fields and produced unparalleled performance results. It plays a vital role in daily applications and is also subtly changing the rules, habits, and behaviors of society. However, inevitably, data-based learning strategies are bound to cause potential security and privacy threats, and arouse public as well as government concerns about its promotion to the real world. In this article, we mainly focus on data security issues in deep learning. We first investigate the potential threats of deep learning in this area, and then present the latest countermeasures based on various underlying technologies, where the challenges and research opportunities on offense and defense are also discussed. Then, we propose SecureNet, the first verifiable and privacy-preserving prediction protocol to protect model integrity and user privacy in DNNs. It can significantly resist various security and privacy threats during the prediction process. We simulate SecureNet under a real dataset, and the experimental results show the superior performance of SecureNet for detecting various integrity attacks against DNN models.

BACKGROUND: Death by suicide is a preventable public health concern worldwide. The aim of this study is to investigate the probability of suicide death using baseline characteristics and simple medical facility visit history data using Cox regression, support vector machines (SVMs), and deep neural networks (DNNs). METHOD: This study included 819,951 subjects in the National Health Insurance Service (NHIS)-Cohort Sample Database from 2004 to 2013. The dataset was divided randomly into two independent training and validation groups. To improve the performance of predicting suicide death, we applied SVM and DNN to the same training set as the Cox regression model. RESULTS: Among the study population, 2546 people died by intentional self-harm during the follow-up time. Sex, age, type of insurance, household income, disability, and medical records of eight ICD-10 codes (including mental and behavioural disorders) were selected by a Cox regression model with backward stepwise elimination. The area of under the curve (AUC) of Cox regression (0.688), SVM (0.687), and DNN (0.683) were approximately the same. The group with top .5% of predicted probability had hazard ratio of 26.21 compared to that with the lowest 10% of predicted probability. LIMITATIONS: This study is limited by the lack of information on suicidal ideation and attempts, other potential covariates such as information of medication and subcategory ICD-10 codes. Moreover, predictors from the prior 12-24 months of the date of death could be expected to show better performances than predictors from up to 10 years ago. CONCLUSIONS: We suggest a 10-year probability prediction model for suicide death using general characteristics and simple insurance data, which are annually conducted by the Korean government. Suicide death prevention might be enhanced by our prediction model. ; restriction

The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-319-13623-3_5 ; Online multimedia repositories are growing rapidly. However, language barriers are often difficult to overcome for many of the current and potential users. In this paper we describe a TTS Spanish sys- tem and we apply it to the synthesis of transcribed and translated video lectures. A statistical parametric speech synthesis system, in which the acoustic mapping is performed with either HMM-based or DNN-based acoustic models, has been developed. To the best of our knowledge, this is the first time that a DNN-based TTS system has been implemented for the synthesis of Spanish. A comparative objective evaluation between both models has been carried out. Our results show that DNN-based systems can reconstruct speech waveforms more accurately. ; The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement no 287755 (transLectures) and ICT Policy Support Programme (ICT PSP/2007-2013) as part of the Competitiveness and Innovation Framework Programme (CIP) under grant agreement no 621030 (EMMA), and the Spanish MINECO Active2Trans (TIN2012-31723) research project. ; Piqueras Gozalbes, SR.; Del Agua Teba, MA.; Giménez Pastor, A.; Civera Saiz, J.; Juan Císcar, A. (2014). Statistical text-to-speech synthesis of Spanish subtitles. En Advances in Speech and Language Technologies for Iberian Languages: Second International Conference, IberSPEECH 2014, Las Palmas de Gran Canaria, Spain, November 19-21, 2014. Proceedings. Springer International Publishing. 40-48. https://doi.org/10.1007/978-3-319-13623-3_5 ; S ; 40 ; 48 ; Ahocoder, http://aholab.ehu.es/ahocoder ; Coursera, http://www.coursera.org ; HMM-Based Speech Synthesis System (HTS), http://hts.sp.nitech.ac.jp ; Khan Academy, http://www.khanacademy.org ; Axelrod, A., He, X., Gao, J.: Domain adaptation via pseudo in-domain data selection. In: Proc. of EMNLP, pp. 355–362 (2011) ; Bottou, L.: ...

هدف الدراسة: واحدة من أكثر الطرق المفيدة لتحديد قدرات الأراضي التي يجب حجزها لأفضل الأشكال وأكثرها إنتاجية هي الأقلمة الزراعية المناخية أو تقسيم المناطق (ACZ). يقسم إطار عمل ACZ العالم إلى مناطق بناءً على التضاريس والتربة واستخدام الأراضي والمناخ. يقدم هذا البحث أقلمة زراعية مناخية ACZ لولاية أندرا براديش (AP) باستخدام خوارزمية تصنيف IHW-DNN. تقسم خوارزمية التصنيف المناطق الزراعية المناخية لولاية أندرا براديش AP إلى سبع مناطق بناءً على بعض خصائص المناطق الثلاث لـ AP. في البداية، تجمع بيانات الإدخال من المواقع المتاحة علناً لإجراء التصنيف. تستخرج الميزات الخاصة لولاية AP؛ مثل الجغرافيا الطبيعية، وخصائص التربة، والمعلومات الجيولوجية، وطول فترة النمو، والأنواع المناخية، والنتح التبخري المحتمل، وأنماط المحاصيل. تصنف ولاية AP إلى سبع مناطق زراعية مناخية بناءً على هذه الميزات. ثم تقدّم جدولة الري للمناطق المصنفة باستخدام خوارزمية التحسين الأمثل، وهي خوارزمية سرب الخنفساء التي تستخدم قيمة تناسب جديدة لإعطاء مجموعة مثالية من جداول الري للمحاصيل في ولاية AP. تجرى التجارب لتحديد مدى فعالية الطرق المقترحة، وتشير النتائج إلى أن الطرق المقترحة فعالة.
الكلمات المفتاحية: الأقلمة الزراعية المناخية، تحديد المناطق المناخية الزراعية، ولاية أندرا براديش، جدولة الري، تحسين سرب الخنفساء الأمثل (BSO)، الشبكة العصبية العميقة (DNN)

Carbon emissions caused by the massive consumption of energy have brought enormous pressure on the Chinese government. Accurately and rapidly characterizing the spatiotemporal characteristics of Chinese city-level carbon emissions is crucial for policy decision making. Based on multi-dimensional data, including nighttime light (NTL) data, land use (LU) data, land surface temperature (LST) data, and added-value secondary industry (AVSI) data, a deep neural network ensemble (DNNE) model was built to analyze the nonlinear relationship between multi-dimensional data and province-level carbon emission statistics (CES) data. The city-level carbon emissions data were estimated, and the spatiotemporal characteristics were analyzed. As compared to the energy statistics released by partial cities, the results showed that the DNNE model based on multi-dimensional data could well estimate city-level carbon emissions data. In addition, according to a linear trend analysis and standard deviational ellipse (SDE) analysis of China from 2001 to 2019, we concluded that the spatiotemporal changes in carbon emissions at the city level were in accordance with the development of China's economy. Furthermore, the results can provide a useful reference for the scientific formulation, implementation, and evaluation of carbon emissions reduction policies.