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).
ردیف | عنوان | نوع |
---|---|---|
1 |
Deep convolutional neural networks-based Hardware–Software on-chip system for computer vision application
سیستم سختافزار-نرمافزار روی تراشه مبتنی بر شبکههای عصبی عمیق برای کاربرد بینایی ماشین-2022 Embedded vision systems are the best solutions for high-performance and lightning-fast inspection tasks. As everyday life evolves, it becomes almost imperative to harness artificial
intelligence (AI) in vision applications that make these systems intelligent and able to make
decisions close to or similar to humans. In this context, the AI’s integration on embedded
systems poses many challenges, given that its performance depends on data volume and
quality they assimilate to learn and improve. This returns to the energy consumption and
cost constraints of the FPGA-SoC that have limited processing, memory, and communication
capacity. Despite this, the AI algorithm implementation on embedded systems can drastically
reduce energy consumption and processing times, while reducing the costs and risks associated
with data transmission. Therefore, its efficiency and reliability always depend on the designed
prototypes. Within this range, this work proposes two different designs for the Traffic Sign
Recognition (TSR) application based on the convolutional neural network (CNN) model,
followed by three implantations on PYNQ-Z1. Firstly, we propose to implement the CNN-based
TSR application on the PYNQ-Z1 processor. Considering its runtime result of around 3.55 s,
there is room for improvement using programmable logic (PL) and processing system (PS) in a
hybrid architecture. Therefore, we propose a streaming architecture, in which the CNN layers
will be accelerated to provide a hardware accelerator for each layer where direct memory
access (DMA) interface is used. Thus, we noticed efficient power consumption, decreased
hardware cost, and execution time optimization of 2.13 s, but, there was still room for design
optimizations. Finally, we propose a second co-design, in which the CNN will be accelerated
to be a single computation engine where BRAM interface is used. The implementation results
prove that our proposed embedded TSR design achieves the best performances compared to the
first proposed architectures, in terms of execution time of about 0.03 s, computation roof of
about 36.6 GFLOPS, and bandwidth roof of about 3.2 GByte/s.
keywords: CNN | FPGA | Acceleration | Co-design | PYNQ-Z1 |
مقاله انگلیسی |
2 |
Retargetable Optimizing Compilers for Quantum Accelerators via a Multilevel Intermediate Representation
کامپایلرهای بهینه سازی مجدد قابل هدف گیری برای شتاب دهنده های کوانتومی از طریق یک نمایش میانی چند سطحی-2022 We present a multilevel quantum–classical intermediate representation (IR) that
enables an optimizing, retargetable compiler for available quantum languages.
Our work builds upon the multilevel intermediate representation (MLIR)
framework and leverages its unique progressive lowering capabilities to map
quantum languages to the low-level virtual machine (LLVM) machine-level IR.
We provide both quantum and classical optimizations via the MLIR pattern
rewriting subsystem and standard LLVM optimization passes, and demonstrate
the programmability, compilation, and execution of our approach via standard
benchmarks and test cases. In comparison to other standalone language and
compiler efforts available today, our work results in compile times that are
1,000 faster than standard Pythonic approaches, and 5–10 faster than
comparative standalone quantum language compilers. Our compiler provides
quantum resource optimizations via standard programming patterns that result
in a 10 reduction in entangling operations, a common source of program
noise. We see this work as a vehicle for rapid quantum compiler prototyping.
|
مقاله انگلیسی |
3 |
A real-time tennis level evaluation and strokes classification system based on the Internet of Things
یک سیستم ارزیابی سطح تنیس در زمان واقعی و طبقه بندی ضربه ها بر اساس اینترنت اشیا-2022 In this study a single wearable inertial measurement unit (IMU) and machine learning method-
ologies were used to conduct player level evaluation and classification five prototype tennis
strokes in real-time. The International Tennis Number (ITN) test was used to verify the accuracy
of this IoT system in evaluating participant level. We conducted the ITN test on thirty-six par-
ticipants and conducted one-way ANOVA on the ITN test results using IBM SPSS 26. The IMU in
this study contained a tri-axis accelerometer (± 16 g) and tri-axis gyroscope (± 2000◦ /s) worn on
the participants’ wrist connected to a wireless low-energy Bluetooth smart-phone with data sent
to the computer terminal by cloud storage. Data processing including preprocessing, segmenta-
tion, feature extraction, dimensionality reduction and classification using Support Vector Ma-
chines (SVM), K-nearest neighbor (K-NN) and Naive Bayes (NB) algorithms. One-way ANOVA
analysis predicting participants’ ITN level and ITN field test scores yielded p < 0.001 at the three
different skill levels tested. SVM (MinMax), SVM (Standardiser) and SVM (MaxAbsScaler) clas-
sified unique tennis strokes precision and recall factors at the three different skill levels reliably
yielded in f1-scores above 0.90 for serve, forehand and backhand, with f1-scores for forehand and
backhand volley scores falling below that. The results of this study suggest using a single six-axial
50 Hz IMU in combination with SVM and SVM + PCA represents a significant step towards a more
reliable wearable tennis stroke performance and skill level real-time evaluation and feedback
technology. keywords: اینترنت اشیا | جمع آوری داده ها | پردازش داده ها | یادگیری ماشین | اپلیکیشن موبایل | تنیس | سنسورهای پوشیدنی | ارتباطات بی سیم | Internet of Things | Data collection | Data processing | Machine learning | Mobile application | Tennis | Wearable sensors | Wireless communication |
مقاله انگلیسی |
4 |
A simplified method to account for vertical human-structure interaction
یک روش ساده برای تعامل با ساختار عمودی انسان-2021 To account for vertical human-structure interaction (HSI) in the vibration serviceability analysis, the contact
force between the pedestrian and the structure can be modelled as the superposition of the force induced by the
pedestrian on a rigid surface and the force resulting from the mechanical interaction between the structure and
the human body. For the case of large crowds, this approach leads to (time-variant) models with a very high
number of degrees of freedom (DOFs). To simplify analysis, this paper investigates the performance of an
equivalent single-degree-of-freedom approach whereby the effect of HSI is translated into an effective natural
frequency and modal damping ratio for each mode of the supporting structure. First, the numerical study con-
siders a footbridge structure that is modelled as a simply-supported beam for which only the fundamental
vertical bending mode is taken into account. For a relevant range of structure and crowd parameters, the
comparison is made between the structural response predicted by the simplified model and the more accurate
reference model that accounts for all DOFs of the coupled crowd-structure model. Where the simplified model is
found to underestimate the structural response, although to a limited extent, this is compensated for by intro-
ducing a correction factor for the effective damping ratio. Second, the performance of the simplified method is
evaluated through the application on a real footbridge. The results show that the simplified method allows for a
good and mildly conservative estimate of the structural acceleration response that is within 10–20% of the
predictions of the reference crowd-structure model. keywords: ارتعاشات ناشی از انسان | تعامل ساختار انسانی | پایه | قابلیت ارتعاش | Human-induced vibrations | Human-structure interaction | Footbridge | Vibration serviceability |
مقاله انگلیسی |
5 |
Civil engineering stability inspection based on computer vision and sensors
بازرسی پایداری مهندسی عمران بر اساس بینایی ماشین و حسگرها-2021 A computer that combines the purchase of vision technology and remote cameras and drones offers a promising non-contact solution for the state evaluation of civil infrastructure. This system’s ultimate goal is too automatically and reliably converted to actionable information image or video data. This white paper provides an overview of computer vision technology’s latest development and applies it to the state evaluation of private infrastructure. Deep learning has been applied to various computer vision; deep learning course covers most of the application. Each application has its architecture, such as the input image and labels data loss function. To explain computer vision architecture in the following figure. Review of the work can be divided into two types: application checks and application monitoring. Review inspection applications include context identifiers, local and global features, visible damage, and changes in the reference image. Monitoring applications described herein include static and dynamic strain modal analysis measurement and displacement measurement. Next, several key challenges continue to move towards civilian infrastructure automation and monitoring of vision- based. Finally, aim to address some of the ongoing challenges in our work. Keywords: Monitoring applications | Computer vision | Accelerometer | Non-destructive evaluation | Conventional-contact displacement sensors |
مقاله انگلیسی |
6 |
یک مدل برای شبیهسازی و طرحریزی پویای مسیر تعویض باند مبتنی بر تابع پارامتری جدید
سال انتشار: 2021 - تعداد صفحات فایل pdf انگلیسی: 16 - تعداد صفحات فایل doc فارسی: 22 مسألهی تعویض باند (LC) میتواند موجب تصادفات شدید شده و ترافیک آزاردهندهای را در جادههای چندبانده ایجاد نماید. مدل موجود برای شبیهسازی LC با یک سری محدودیتها (انطباق کم، فقدان مشخصههای سرعت و شتاب، انحنای زیاد) با استفاده از منحنی مسیرهای شناختهشدهای همچون منحنی مماس هایپربولیک (HTC)، منحنی مبتنی بر سینوس (SC)، و منحنی چندجملهای (PC) ایجاد شد. در این مقاله، یک منحنی پارامتری جدید با استفاده از دستگاه مختصات خمیدهخطی ارائه و با پایگاه دادهی واقعی شبیهسازی نسل آتی (NGSIM) انطباق داده شد. سپس مشخصههای جدید سرعت و شتاب با استفاده از منحنی مسیر LC پیشنهاد شدند. انحنای مدل پیشنهادی در هر دو نقطهی آغاز و پایان LC، انحنای مبتنی بر صفر بود. این انحنای پیشنهادی با دو مدل همانند HTC و SC مقایسه شد. خطای متوسط جذر میانگین مربعات مدل پیشنهادی در مقایسه با مدل HTC، برای LC چپ به میزان 1.84% و برای LC راست به میزان 15.48% و در مقایسه با مدل SC به میزان 1.74% برای LC چپ و به میزان 15.60% برای LC راست کاهش مییابد. بطور مشابه، مدل پیشنهادی برای مشخصههای سرعت و شتاب نسبت به مدل PC تا حد زیادی بهبود مییابد. منحنی پارامتری پیشنهادی، نقاط فاصله و برخورد خودروی LC با یک خودروی جلویی و خودروی پشتی در باند هدف را حل میکند و میتوان از آن در برنامهریزی مسیر LC واقعی استفاده کرد.
کلیدواژه ها: مشخصههای شتاب | منحنی پارامتری | سرعت | برنامهریزی مسیر |
مقاله ترجمه شده |
7 |
Benchmarking vision kernels and neural network inference accelerators on embedded platforms
محک زدن هسته بینایی و شتاب دهنده های استنتاج شبکه عصبی بر روی سیستم عامل های توکار-2021 Developing efficient embedded vision applications requires exploring various algorithmic optimization trade- offs and a broad spectrum of hardware architecture choices. This makes navigating the solution space and finding the design points with optimal performance trade-offs a challenge for developers. To help provide a fair baseline comparison, we conducted comprehensive benchmarks of accuracy, run-time, and energy efficiency of a wide range of vision kernels and neural networks on multiple embedded platforms: ARM57 CPU, Nvidia Jetson TX2 GPU and Xilinx ZCU102 FPGA. Each platform utilizes their optimized libraries for vision kernels (OpenCV, Vision Works and xfOpenCV) and neural networks (OpenCV DNN, TensorRT and Xilinx DPU). Forvision kernels, our results show that the GPU achieves an energy/frame reduction ratio of 1.1–3.2× compared to the others for simple kernels. However, for more complicated kernels and complete vision pipelines, the FPGA outperforms the others with energy/frame reduction ratios of 1.2–22.3×. For neural networks [Inception-v2 and ResNet-50, ResNet-18, Mobilenet-v2 and SqueezeNet], it shows that the FPGA achieves a speed up of [2.5, 2.1, 2.6, 2.9 and 2.5]× and an EDP reduction ratio of [1.5, 1.1, 1.4, 2.4 and 1.7]× compared to the GPU FP16 implementations, respectively. Keywords: Benchmarks | CPUs | GPUs | FPGAs | Embedded vision | Neural networks |
مقاله انگلیسی |
8 |
Are social incubators different from other incubators? Evidence from Italy
آیا دستگاه های جوجه کشی اجتماعی با سایر دستگاه های جوجه کشی فرق دارند؟ مدارکی از ایتالیا-2020 This paper defines and analyses incubators that mainly support start-ups with a significant social impact. In 2016, a survey was conducted on the 162 incubators active in Italy, and a total of 88 responses were received. An analysis of the literature and of this dataset led to the identification of three types of incubators: Business, Mixed, and Social. Thirty of the respondents sent information on their tenants. Thanks to the data regarding 247 tenants, it was possible to analyze the impact of the three different types of incubators (Business, Mixed, and Social) on the tenants’ growth through OLS regression analyses. A Social Incubator is here defined as an incubator that supports more than 50% of start-ups that aim to introduce a positive social impact. The study shows that Social Incubators perceive social impact measurement and training/consulting on business ethics and CSR as being more important services than other incubator types. The regression analyses explain that Social Incubators are as efficient as other incubators, in terms of tenants’ economic growth, notwithstanding the focus of Social Incubators on start-ups that do not pursue only economic objectives. Finally, this study indicates that policy- makers can foster Social Incubators to support social entrepreneurship. Keywords: Incubators | Accelerators | Social start-up | Social entrepreneurship | Social innovation | Entrepreneurship |
مقاله انگلیسی |
9 |
A pressure- or velocity-dependent acceleration rate law for the shock-to-detonation transition process in PBX 9502 high explosive
قانون میزان شتاب فشار یا وابسته به سرعت برای روند انتقال شوک به انفجار در مواد منفجره PBX 9502-2020 Shock-to-detonation transition profiles of PBX 9502 explosive are analyzed to develop a rate law for shock acceleration. The shock motion profiles are seen to follow a common trend in the shock acceleration–velocity frame, aside from an early time transient that is dependent on the initiating shock strength. The duration of the early time transient is seen to correlate with the initial shock strength. The com- mon shock acceleration profile is seen to be Arrhenius-like with respect to the local particle velocity or pressure. A dual-rate pressure-dependent Arrhenius-type rate law is developed with the duration of the early rate set by the initial shock strength. The rate law is able to predict the shock motion for all tests well in both particle velocity and pressure space. In addition to directly measuring commonalities in the acceleration profiles of the experimental shock motion, this work provides insight into the functional form of the reaction rate laws for this TATB-based high explosive. The rate law also supports the concept that shock-driven reaction in heterogenous high explosives is driven by localized ignition and growth of hotspots. Keywords: Detonation | Shock-to-detonation | transition SDT | Detonation reaction rate | High explosive |
مقاله انگلیسی |
10 |
A different sleep apnea classification system with neural network based on the acceleration signals
یک سیستم طبقه بندی sleep apnea متفاوت با شبکه عصبی مبتنی بر سیگنال های شتاب-2020 Background and objective: The apnea syndrome is characterized by an abnormal breath pause or reduction
in the airflow during sleep. It is reported in the literature that it affects 2% of middle-aged women
and 4% of middle-aged men, approximately. This study has vital importance, especially for the elderly,
the disabled, and pediatric sleep apnea patients.
Methods: In this study, a new diagnostic method is developed to detect the apnea event by using a microelectromechanical
system (MEMS) based acceleration sensor. It records the value of acceleration by measuring
the movements of the diaphragm in three axes during the respiratory. The measurements are
carried out simultaneously, a medical spirometer (Fukuda Sangyo), to test the validity of measurement
results. An artificial neural network model was designed to determine the apnea event. For the number
of neurons in the hidden layer, 1-3-5-10-18-20-25 values were tried, and the network with three hidden
neurons giving the most suitable result was selected. In the designed ANN, three layers were formed that
three neurons in the hidden layer, the two neurons at the input, and two neurons at the output layer.
Results: A study group was formed of 5 patients (having different characteristics (age, height, and body
weight)). The patients in the study group have sleep apnea (SA) in different grades. Several 12.723 acceleration
data (ACC) in the XYZ-axis from 5 different patients are recorded for apnea event training and
detection. The measured accelerometer (ACC) data from one of the patients (called H1) are used to train
an ANN. During the training phase, MSE is used to calculate the fitness value of the apnea event. Then
Apnea event is detected successfully for the other patients by using ANN trained only with H1’s ACC data.
Conclusions: The sleep apnea event detection system is presented by using ANN from directly acceleration
values. Measurements are performed by the MEMS-based accelerometer and Industrial
Spirometer simultaneously. A total of 12723 acceleration data is measured from 5 different patients.
The best result in 7000 iterations was reached (the number of iterations was tried up to 10.000 with
1000 steps). 605 data of only H1 measurements are used to train ANN, and then all data used to check
the performance of the ANN as well as H2, H3, H4, and H5 measurement results. MSE performance benchmark
shows us that trained ANN successfully detects apnea events. One of the contributions of this study
to literature is that only ACC data are used in the ANN training step. After training for one patient, the
ANN system can monitor the apnea event situation on-line for others. Keywords: Sleep apnea | Acceleration sensor | Acceleration data | Artificial neural network | Medical decision making |
مقاله انگلیسی |