دانلود و نمایش مقالات مرتبط با Deep neural network::صفحه 1
بلافاصله پس از پرداخت دانلود کنید

با سلام خدمت کاربران در صورتی که با خطای سیستم پرداخت بانکی مواجه شدید از طریق کارت به کارت (6037997535328901 بانک ملی ناصر خنجری ) مقاله خود را دریافت کنید (تا مشکل رفع گردد). 

نتیجه جستجو - Deep neural network

تعداد مقالات یافته شده: 112
ردیف عنوان نوع
1 DQRA: Deep Quantum Routing Agent for Entanglement Routing in Quantum Networks
DQRA: عامل مسیریابی کوانتومی عمیق برای مسیریابی درهم تنیده در شبکه های کوانتومی-2022
Quantum routing plays a key role in the development of the next-generation network system. In particular, an entangled routing path can be constructed with the help of quantum entanglement and swapping among particles (e.g., photons) associated with nodes in the network. From another side of computing, machine learning has achieved numerous breakthrough successes in various application domains, including networking. Despite its advantages and capabilities, machine learning is not as much utilized in quantum networking as in other areas. To bridge this gap, in this article, we propose a novel quantum routing model for quantum networks that employs machine learning architectures to construct the routing path for the maximum number of demands (source–destination pairs) within a time window. Specifically, we present a deep reinforcement routing scheme that is called Deep Quantum Routing Agent (DQRA). In short, DQRA utilizes an empirically designed deep neural network that observes the current network states to accommodate the network’s demands, which are then connected by a qubit-preserved shortest path algorithm. The training process of DQRA is guided by a reward function that aims toward maximizing the number of accommodated requests in each routing window. Our experiment study shows that, on average, DQRA is able to maintain a rate of successfully routed requests at above 80% in a qubit-limited grid network and approximately 60% in extreme conditions, i.e., each node can be repeater exactly once in a window. Furthermore, we show that the model complexity and the computational time of DQRA are polynomial in terms of the sizes of the quantum networks.
INDEX TERMS: Deep learning | deep reinforcement learning (DRL) | machine learning | next-generation network | quantum network routing | quantum networks.
مقاله انگلیسی
2 A survey on adversarial attacks in computer vision: Taxonomy, visualization and future directions
بررسی حملات خصمانه در بینایی کامپیوتر: طبقه بندی، تجسم و جهت گیری های آینده-2022
Deep learning has been widely applied in various fields such as computer vision, natural language pro- cessing, and data mining. Although deep learning has achieved significant success in solving complex problems, it has been shown that deep neural networks are vulnerable to adversarial attacks, result- ing in models that fail to perform their tasks properly, which limits the application of deep learning in security-critical areas. In this paper, we first review some of the classical and latest representative adversarial attacks based on a reasonable taxonomy of adversarial attacks. Then, we construct a knowl- edge graph based on the citation relationship relying on the software VOSviewer, visualize and analyze the subject development in this field based on the information of 5923 articles from Scopus. In the end, possible research directions for the development about adversarial attacks are proposed based on the trends deduced by keywords detection analysis. All the data used for visualization are available at: https://github.com/NanyunLengmu/Adversarial- Attack- Visualization .
keywords: یادگیری عمیق | حمله خصمانه | حمله جعبه سیاه | حمله به جعبه سفید | نیرومندی | تجزیه و تحلیل تجسم | Deep learning | Adversarial attack | Black-box attack | White-box attack | Robustness | Visualization analysis
مقاله انگلیسی
3 Computer vision for anatomical analysis of equipment in civil infrastructure projects: Theorizing the development of regression-based deep neural networks
چشم انداز کامپیوتری برای تجزیه و تحلیل آناتومیکی تجهیزات در پروژه های زیرساختی عمرانی: نظریه پردازی توسعه شبکه های عصبی عمیق مبتنی بر رگرسیون-2022
There is high demand for heavy equipment in civil infrastructure projects and their performance is a determinant of the successful delivery of site operations. Although manufacturers provide equipment performance hand- books, additional monitoring mechanisms are required to depart from measuring performance on the sole basis of unit cost for moved materials. Vision-based tracking and pose estimation can facilitate site performance monitoring. This research develops several regression-based deep neural networks (DNNs) to monitor equipment with the aim of ensuring safety, productivity, sustainability and quality of equipment operations. Annotated image libraries are used to train and test several backbone architectures. Experimental results reveal the pre- cision of DNNs with depthwise separable convolutions and computational efficiency of DNNs with channel shuffle. This research provides scientific utility by developing a method for equipment pose estimation with the ability to detect anatomical angles and critical keypoints. The practical utility of this study is the provision of potentials to influence current practice of articulated machinery monitoring in projects.
keywords: هوش مصنوعی (AI) | سیستم های فیزیکی سایبری | معیارهای ارزیابی خطا | طراحی و آزمایش تجربی | تخمین ژست کامل بدن | صنعت و ساخت 4.0 | الگوریتم های یادگیری ماشین | معماری های ستون فقرات شبکه | Artificial intelligence (AI) | Cyber physical systems | Error evaluation metrics | Experimental design and testing | Full body pose estimation | Industry and construction 4.0 | Machine learning algorithms | Network backbone architectures
مقاله انگلیسی
4 PortiK: A computer vision based solution for real-time automatic solid waste characterization – Application to an aluminium stream
PortiK: یک راه حل مبتنی بر بینایی کامپیوتری برای شناسایی خودکار زباله جامد در زمان واقعی - کاربرد در جریان آلومینیوم-2022
In Material Recovery Facilities (MRFs), recyclable municipal solid waste is turned into a precious commodity. However, effective recycling relies on effective waste sorting, which is still a challenge to sustainable develop- ment of our society. To help the operations improve and optimise their process, this paper describes PortiK, a solution for automatic waste analysis. Based on image analysis and object recognition, it allows for continuous, real-time, non-intrusive measurements of mass composition of waste streams. The end-to-end solution is detailed with all the steps necessary for the system to operate, from hardware specifications and data collection to su- pervisory information obtained by deep learning and statistical analysis. The overall system was tested and validated in an operational environment in a material recovery facility. PortiK monitored an aluminium can stream to estimate its purity. Aluminium cans were detected with 91.2% precision and 90.3% recall, respectively, resulting in an underestimation of the number of cans by less than 1%. Regarding contaminants (i.e. other types of waste), precision and recall were 80.2% and 78.4%, respectively, giving an 2.2% underestimation. Based on five sample analyses where pieces of waste were counted and weighed per batch, the detection results were used to estimate purity and its confidence level. The estimation error was calculated to be within ±7% after 5 minutes of monitoring and ±5% after 8 hours. These results have demon- strated the feasibility and the relevance of the proposed solution for online quality control of aluminium can stream.
keywords: امکانات بازیابی مواد | شناسایی مواد زائد جامد | یادگیری عمیق | شبکه عصبی عمیق | بینایی کامپیوتر | Material recovery facilities | MRF | Solid waste characterization | Deep-learning | Deep neural network | Computer vision
مقاله انگلیسی
5 A computer vision-based method to identify the international roughness index of highway pavements
یک روش مبتنی بر بینایی کامپیوتری برای شناسایی شاخص ناهمواری بین‌المللی روسازی بزرگراه-2022
The International Roughness Index (IRI) is one of the most critical parameters in the field of pavement performance management. Traditional methods for the measurement of IRI rely on expensive instrumented vehicles and well-trained professionals. The equipment and labor costs of traditional measurement methods limit the timely updates of IRI on the pavements. In this article, a novel imaging-based Deep Neural Network (DNN) model, which can use pavement photos to directly identify the IRI values, is proposed. This model proved that it is possible to use 2-dimensional (2D) images to identify the IRI other than the typically used vertical accelerations or 3-dimensional (3D) images. Due to the fast growth in photography equipment, small and convenient sports action cameras such as the GoPro Hero series are able to capture smooth videos at a high framerate with built-in electronic image stabilization systems. These significant improvements make it not only more convenient to collect high-quality 2D images, but also easier to process them than vibrations or accelerations. In the proposed method, 15% of the imaging data were randomly selected for testing and had never been touched during the training steps. The testing results showed an averaged coefficient of determination (R square) of 0.6728 and an averaged root mean square error (RMSE) of 0.50.
keywords: شاخص بین المللی زبری | شبکه عصبی عمیق | بینایی کامپیوتر | ارزیابی وضعیت روسازی | International roughness index | Deep neural network | Computer vision | Pavement condition assessment
مقاله انگلیسی
6 Deep Q learning based secure routing approach for OppIoT networks
رویکرد مسیریابی ایمن مبتنی بر یادگیری Q برای شبکه های OppIoT-2022
Opportunistic IoT (OppIoT) networks are a branch of IoT where the human and machines collaborate to form a network for sharing data. The broad spectrum of devices and ad-hoc nature of connections, further alleviate the problem of network and data security. Traditional approaches like trust based approaches or cryptographic approaches fail to preemptively secure these networks. Machine learning (ML) approaches, mainly deep reinforcement learning (DRL) methods can prove to be very effective in ensuring the security of the network as they are profoundly capable of solving complex and dynamic problems. Deep Q-learning (DQL) incorporates deep neural network in the Q learning process for dealing with high-dimensional data. This paper proposes a routing approach for OppIoT, DQNSec, based on DQL for securing the network against attacks viz. sinkhole, hello flood and distributed denial of service attack. The actor–critic approach of DQL is utilized and OppIoT is modeled as a Markov decision process (MDP). Extensive simulations prove the efficiency of DQNSec in comparison to other ML based routing protocols, viz. RFCSec, RLProph, CAML and MLProph.
Keywords: OppIoT | Reinforcement learning | Deep learning | Deep Q-learning | Markov decision process | Sinkhole attack | Hello flood attack | Distributed denial of service attack
مقاله انگلیسی
7 Evaluating computing performance of deep neural network models with different backbones on IoT-based edge and cloud platforms
ارزیابی عملکرد محاسباتی مدل‌های شبکه عصبی عمیق با ستون فقرات مختلف در بسترهای لبه و ابری مبتنی بر اینترنت اشیا-2022
This paper focuses on evaluating and predicting the computing performance of different archi- tectures of deep neural network models (DNNs) in cross-platform and cross-inference frame- works. We test nearly 30 typical DNN models for image recognition on Google Colab cloud computing platform and Intel neural compute stick 2 embedded edge computing platform and record the computational performance metrics i.e. the Top-N accuracy, model complexity, computational complexity, inference time, memory usage, and so on. We compare and analyze these performance parameters with the previous workstation equipped with NVIDIA Titan X Pascal and an embedded system based on NVIDIA Jetson TX1 board to evaluate the inference efficiency of different DNN models using different inference frameworks. The methods of ANOVA are adopted to quantify the differences between the models. A combination method of cluster analysis and regression analysis is proposed to find the similar inference time variation processes across models, which can be used to predict the inference results of unknown models. These presented results will contribute to better deployment and application of resource-constrained DNN models on the heterogeneous high-performance computing platform.
keywords: شبکه های عصبی عمیق | سکوهای متقابل | چارچوب های استنتاج متقابل | تشخیص تصویر | Deep neural networks | Cross platforms | Cross-inference frameworks | Image recognition
مقاله انگلیسی
8 A survey: Intelligent system for imposter detection
یک مرور: سیستم هوشمند برای تشخیص جعل کننده-2021
This study aims the impostor is a very cunning person who reaches an obsessive stage to perfection in impersonating someone in actual life, concentrates on his biometric. He analyzes the controls, restrictions, and obstacles that he will face to overcome them. The technologies biometric recognition performs a greatly important role in impostor detection. Biometrics properties refer to the automatic recognition of persons depending on their behavioral and physiological characteristics. Biometrics comprises face recognition, fingerprint, voice recognition, retinal scanning, and so on. Biometrics may increment the reliability of an ID card system. In this paper, a review of the concepts mentioned above will be provided. At first, a presentation about a procedural overview of biometric recognition technologies, ID card systems. Then dissection will be presented for the review of the most recent techniques. A description of each concept will be given and a comparison study is achieved with formal discussion and analysis for each approach result introduces in this study. Finally, a summary of the research results is given.© 2021 Elsevier Ltd. All rights reserved. Selection and peer-review under responsibility of the scientific committee of the Emerging Trends in Materials Science, Technology and Engineering.
Keywords: Face recognition | Voice recognition | Finger print | Biometric systems | ID card | Person identification | Impostor detection | Machine learning | Deep neural networks
مقاله انگلیسی
9 PlexNet: A fast and robust ECG biometric system for human recognition
PlexNet: یک سیستم بیومتریک ECG سریع و قوی برای تشخیص انسان-2021
Researchers have explored the potential of electrocardiogram (ECG) to use as biometrics from past two decades. ECG has the inherent feature of vitality for securing the biometric system from fraudulent attacks. This paper proposes a novel ensemble of the state-of-the- art pre-trained deep neural networks i.e., ResNet and DenseNet for ECG biometric recognition. The principle of transfer learning is utilized to prepare fine-tuned models. The gathered knowledge of four fine-tuned models is fused to prepare one stacking model i.e., ‘PlexNet’. The PlexNet takes advantage of transfer learning along with ensemble learning, thus making a novel model for ECG biometrics that is robust and secure than other met ods using deep networks. Two public datasets PTB and CYBHI are tested on the proposed ensemble for human identification. The experimental results demonstrate the efficacy of the model with identification accuracy reported the best as 99.66% on healthy and unhealthy subjects. Finally, the proposed ECG biometric method proves its robustness from signal acquisition methods, size of datasets, and subject health statuses.© 2021 Elsevier Inc. All rights reserved.
Keywords: Biometric Identification | Electrocardiogram | PlexNet | ResNet | DenseNet
مقاله انگلیسی
10 063-S0893608020304470
063-S0893608020304470-2021
Deep Neural Networks (DNNs) have become popular for various applications in the domain of image and computer vision due to their well-established performance attributes. DNN algorithms involve powerful multilevel feature extractions resulting in an extensive range of parameters and memory footprints. However, memory bandwidth requirements, memory footprint and the associated power consumption of models are issues to be addressed to deploy DNN models on embedded platforms for real time vision-based applications. We present an optimized DNN model for memory and accuracy for vision-based applications on embedded platforms. In this paper we propose Quantization Friendly MobileNet (QF-MobileNet) architecture. The architecture is optimized for inference accuracy and reduced resource utilization. The optimization is obtained by addressing the redundancy and quantization loss of the existing baseline MobileNet architectures. We verify and validate the per- formance of the QF-MobileNet architecture for image classification task on the ImageNet dataset. The proposed model is tested for inference accuracy and resource utilization and compared to the baseline MobileNet architecture. The inference accuracy of the proposed QF-MobileNetV2 float model attained 73.36% and the quantized model has 69.51%. The MobileNetV3 float model attained an inference accuracy of 68.75% and the quantized model has 67.5% respectively. The proposed model saves 33% of time complexity for QF-MobileNetV2 and QF-MobileNetV3 models against the baseline models. The QF-MobileNet also showed optimized resource utilization with 32% fewer tunable parameters, 30% fewer MAC’s operations per image and reduced inference quantization loss by approximately 5% compared to the baseline models. The model is ported onto the android application using TensorFlow API. The android application performs inference on the native devices viz. smartphones, tablets and handheld devices. Future work is focused on introducing channel-wise and layer-wise quantization schemes to the proposed model. We intend to explore quantization aware training of DNN algorithms to achieve optimized resource utilization and inference accuracy.© 2020 Elsevier Ltd. All rights reserved.
Keywords: Deep Neural Network | Classification | MobileNet | Computer vision | Embedded platform | Quantization
مقاله انگلیسی
rss مقالات ترجمه شده rss مقالات انگلیسی rss کتاب های انگلیسی rss مقالات آموزشی
logo-samandehi
بازدید امروز: 2612 :::::::: بازدید دیروز: 3097 :::::::: بازدید کل: 36879 :::::::: افراد آنلاین: 58