Search results for: depth-wise separable convolutional neural network for light-weight GAN architecture for EDGE devices

Authors: Djamel Fawzi Hadj Sadok, Marrone Silvério Melo Dantas Pedro Henrique Dreyer, Gabriel Fonseca Reis de Souza, Daniel Bezerra, Ricardo Souza, Silvia Lins, Judith Kelner

Abstract:

A radio base station (RBS), part of the radio access network, is a particular type of equipment that supports the connection between a wide range of cellular user devices and an operator network access infrastructure. Nowadays, most of the RBS maintenance is carried out manually, resulting in a time consuming and costly task. A suitable candidate for RBS maintenance automation is repairing faulty links between devices caused by missing or unplugged connectors. A suitable candidate for RBS maintenance automation is repairing faulty links between devices caused by missing or unplugged connectors. This paper proposes and compares two deep learning solutions to identify attached RJ45 connectors on network ports. We named connector detection, the solution based on object detection, and connector classification, the one based on object classification. With the connector detection, we get an accuracy of 0:934, mean average precision 0:903. Connector classification, get a maximum accuracy of 0:981 and an AUC of 0:989. Although connector detection was outperformed in this study, this should not be viewed as an overall result as connector detection is more flexible for scenarios where there is no precise information about the environment and the possible devices. At the same time, the connector classification requires that information to be well-defined.

Keywords: radio base station, maintenance, classification, detection, deep learning, automation

Procedia PDF Downloads 180

Authors: Akinbo Razak Yinka, Adesanya Kehinde Kazeem, Oladokun Oluwagbenga Peter

Abstract:

Time series data are sequences of observations collected over a period of time. Such data can be used to predict health outcomes, such as disease progression, mortality, hospitalization, etc. The Statistical approach is based on mathematical models that capture the patterns and trends of the data, such as autocorrelation, seasonality, and noise, while Neural methods are based on artificial neural networks, which are computational models that mimic the structure and function of biological neurons. This paper compared both parametric and non-parametric time series models of patients treated for malaria in Maternal and Child Health Centres in Lagos State, Nigeria. The forecast methods considered linear regression, Integrated Moving Average, ARIMA and SARIMA Modeling for the parametric approach, while Multilayer Perceptron (MLP) and Long Short-Term Memory (LSTM) Network were used for the non-parametric model. The performance of each method is evaluated using the Mean Absolute Error (MAE), R-squared (R2) and Root Mean Square Error (RMSE) as criteria to determine the accuracy of each model. The study revealed that the best performance in terms of error was found in MLP, followed by the LSTM and ARIMA models. In addition, the Bootstrap Aggregating technique was used to make robust forecasts when there are uncertainties in the data.

Keywords: ARIMA, bootstrap aggregation, MLP, LSTM, SARIMA, time-series analysis

Procedia PDF Downloads 62

Authors: Gałka Aleksandra, Jelińska Justyna, Masiak Albert, Walentukiewicz Krzysztof

Abstract:

An offline signature is well-known however not the safest way to verify identity. Nowadays, to ensure proper authentication, i.e. in banking systems, multimodal verification is more widely used. In this paper the online signature analysis based on dynamic time warping (DTW) coupled with machine learning approaches has been presented. In our research signatures made with biometric pens were gathered. Signature features as well as their forgeries have been described. For verification of authenticity various methods were used including convolutional neural networks using DTW matrix and multilayer perceptron using sums of DTW matrix paths. System efficiency has been evaluated on signatures and signature forgeries collected on the same day. Results are presented and discussed in this paper.

Keywords: dynamic time warping, handwritten signature verification, feature-based recognition, online signature

Procedia PDF Downloads 151

Authors: Anthony D. Rhodes, Manan Goel

Abstract:

We provide a high fidelity deep learning algorithm (HyperSeg) for interactive video segmentation tasks using a dense convolutional network with context-aware skip connections and compressed, 'hypercolumn' image features combined with a convolutional tessellation procedure. In order to maintain high output fidelity, our model crucially processes and renders all image features in high resolution, without utilizing downsampling or pooling procedures. We maintain this consistent, high grade fidelity efficiently in our model chiefly through two means: (1) we use a statistically-principled, tensor decomposition procedure to modulate the number of hypercolumn features and (2) we render these features in their native resolution using a convolutional tessellation technique. For improved pixel-level segmentation results, we introduce a boundary loss function; for improved temporal coherence in video data, we include temporal image information in our model. Through experiments, we demonstrate the improved accuracy of our model against baseline models for interactive segmentation tasks using high resolution video data. We also introduce a benchmark video segmentation dataset, the VFX Segmentation Dataset, which contains over 27,046 high resolution video frames, including green screen and various composited scenes with corresponding, hand-crafted, pixel-level segmentations. Our work presents a improves state of the art segmentation fidelity with high resolution data and can be used across a broad range of application domains, including VFX pipelines and medical imaging disciplines.

Keywords: computer vision, object segmentation, interactive segmentation, model compression

Procedia PDF Downloads 106

Authors: Massimo Miccoli, Luca Marangoni, Alberto Aniello Scaringi, Alessandro Marceddu, Alessandro Amicone

Abstract:

The susceptibility of deep neural networks (DNNs) to adversarial manipulations is a recognized challenge within the computer vision domain. Adversarial examples, crafted by adding subtle yet malicious alterations to benign images, exploit this vulnerability. Various defense strategies have been proposed to safeguard DNNs against such attacks, stemming from diverse research hypotheses. Building upon prior work, our approach involves the utilization of autoencoder models. Autoencoders, a type of neural network, are trained to learn representations of training data and reconstruct inputs from these representations, typically minimizing reconstruction errors like mean squared error (MSE). Our autoencoder was trained on a dataset of benign examples; learning features specific to them. Consequently, when presented with significantly perturbed adversarial examples, the autoencoder exhibited high reconstruction errors. The architecture of the autoencoder was tailored to the dimensions of the images under evaluation. We considered various image sizes, constructing models differently for 256x256 and 512x512 images. Moreover, the choice of the computer vision model is crucial, as most adversarial attacks are designed with specific AI structures in mind. To mitigate this, we proposed a method to replace image-specific dimensions with a structure independent of both dimensions and neural network models, thereby enhancing robustness. Our multi-modal autoencoder reconstructs the spectral representation of images across the red-green-blue (RGB) color channels. To validate our approach, we conducted experiments using diverse datasets and subjected them to adversarial attacks using models such as ResNet50 and ViT_L_16 from the torch vision library. The autoencoder extracted features used in a classification model, resulting in an MSE (RGB) of 0.014, a classification accuracy of 97.33%, and a precision of 99%.

Keywords: adversarial attacks, malicious images detector, binary classifier, multimodal transformer autoencoder

Procedia PDF Downloads 93

Authors: Moumita Chanda, Md. Fazlul Karim Patwary

Abstract:

Smoking is one of the most popular recreational drug use behaviors, and it contributes to birth defects, COPD, heart attacks, and erectile dysfunction. To completely eradicate this disease, it is imperative that it be identified and treated. Numerous smoking cessation programs have been created, and they demonstrate how beneficial it may be to help someone stop smoking at the ideal time. A tomography meter is an effective smoking detector. Other wearables, such as RF-based proximity sensors worn on the collar and wrist to detect when the hand is close to the mouth, have been proposed in the past, but they are not impervious to deceptive variables. In this study, we create a machine that can discriminate between smokers and non-smokers in real-time with high sensitivity and specificity by watching and collecting the human lung and analyzing the X-ray data using machine learning. If it has the highest accuracy, this machine could be utilized in a hospital, in the selection of candidates for the army or police, or in university entrance.

Keywords: CNN, smoker detection, non-smoker detection, OpenCV, artificial Intelligence, X-ray Image detection

Procedia PDF Downloads 68

Authors: A. Majeed, P. Ali

Abstract:

Before designing an electrical system, the estimation of load is necessary for unit sizing and demand-generation balancing. The system could be a stand-alone system for a village or grid connected or integrated renewable energy to grid connection, especially as there are non–electrified villages in developing countries. In the classical model, the energy demand was found by estimating the household appliances multiplied with the amount of their rating and the duration of their operation, but in this paper, information exists for electrified villages could be used to predict the demand, as villages almost have the same life style. This paper describes a method used to predict the average energy consumed in each two months for every consumer living in a village by Artificial Neural Network (ANN). The input data are collected using a regional survey for samples of consumers representing typical types of different living, household appliances and energy consumption by a list of information, and the output data are collected from administration office of Piramagrun for each corresponding consumer. The result of this study shows that the average demand for different consumers from four villages in different months throughout the year is approximately 12 kWh/day, this model estimates the average demand/day for every consumer with a mean absolute percent error of 11.8%, and MathWorks software package MATLAB version 7.6.0 that contains and facilitate Neural Network Toolbox was used.

Keywords: artificial neural network, load estimation, regional survey, rural electrification

Procedia PDF Downloads 108

Authors: M. Boudjerda, B. Touaibia, M. K. Mihoubi

Abstract:

In Algeria agricultural irrigation is the primary water consuming sector followed by the domestic and industrial sectors. Economic development in the last decade has weighed heavily on water resources which are relatively limited and gradually decreasing to the detriment of agriculture. The research presented in this paper focuses on the optimization of irrigation water demand. Dynamic Programming-Neural Network (DPNN) method is applied to investigate reservoir optimization. The optimal operation rule is formulated to minimize the gap between water release and water irrigation demand. As a case study, Foum El-Gherza dam’s reservoir system in south of Algeria has been selected to examine our proposed optimization model. The application of DPNN method allowed increasing the satisfaction rate (SR) from 12.32% to 55%. In addition, the operation rule generated showed more reliable and resilience operation for the examined case study.

Keywords: water management, agricultural demand, dam and reservoir operation, Foum el-Gherza dam, dynamic programming, artificial neural network

Procedia PDF Downloads 101

Authors: Ghazala Anjum, Farooq Hussain Bhat, Ravi Kumar

Abstract:

Multiferroic materials are vital for new application and memory devices, not only because of the presence of multiple types of domains but also as a result of cross correlation between coexisting forms of magnetic and electrical orders. In spite of wide studies done on multiferroic bulk ceramic materials their realization in thin film form is yet limited due to some crucial problems. During the last few years, special attention has been devoted to synthesis of thin films like of BiFeO₃. As they allow direct integration of the material into the device technology. Therefore owing to the process of exploration of new multiferroic thin films, preparation, and characterization of La₀.₈Bi₀.₂Fe₀.₇Mn₀.₃O₃ (LBFMO3) thin film on LaAlO₃ (LAO) substrate with LaNiO₃ (LNO) being the buffer layer has been done. The fact that all the electrical and magnetic properties are closely related to the electronic structure makes it inevitable to study the electronic structure of system under study. Without the knowledge of this, one may never be sure about the mechanism responsible for different properties exhibited by the thin film. Literature review reveals that studies on change in atomic and the hybridization state in multiferroic samples are still insufficient except few. The technique of x-ray absorption (XAS) has made great strides towards the goal of providing such information. It turns out to be a unique signature to a given material. In this milieu, it is time honoured to have the electronic structure study of the elements present in the LBFMO₃ multiferroic thin film on LAO substrate with buffer layer of LNO synthesized by RF sputtering technique. We report the electronic structure studies of well characterized LBFMO3 multiferroic thin film on LAO substrate with LNO as buffer layer using near-edge X-ray absorption fine structure (NEXAFS). Present exploration has been performed to find out the valence state and crystal field symmetry of ions present in the system. NEXAFS data of O K- edge spectra reveals a slight shift in peak position along with growth in intensities of low energy feature. Studies of Mn L₃,₂- edge spectra indicates the presence of Mn³⁺/Mn⁴⁺ network apart from very small contribution from Mn²⁺ ions in the system that substantiates the magnetic properties exhibited by the thin film. Fe L₃,₂- edge spectra along with spectra of reference compound reveals that Fe ions are present in +3 state. Electronic structure and valence state are found to be in accordance with the magnetic properties exhibited by LBFMO/LNO/LAO thin film.

Keywords: magnetic, multiferroic, NEXAFS, x-ray absorption fine structure, XMCD, x-ray magnetic circular dichroism

Procedia PDF Downloads 140

Authors: El Miloudi Djelloul

Abstract:

The computer network play an important position for practical implementations of the differently system. To implement a system into network above all is needed to know all the configurations, which is responsible to be a part of the system, and to give adequate information and solution in realtime. So if want to implement this system for example in the school or relevant institutions, the first step is to analyze the types of model which is needed to be configured and another important step is to organize the works in the context of devices, as a part of the general system. Often before configuration, as important point is descriptions and documentations from all the works into the respective process, and then to organize in the aspect of problem-solving. The computer network as critic infrastructure is very specific so the paper present the effectiveness solutions in the structured aspect viewed from one side, and another side is, than the paper reflect the positive aspect in the context of modeling and block schema presentations as an better alternative to solve the specific problem because of continually distortions of the system from the line of devices, programs and signals or packed collisions, which are in movement from one computer node to another nodes.

Keywords: local area networks, LANs, block schema presentations, computer network system, computer node, critical infrastructure packed collisions, structural control system representations, computer network, implementations, modeling structural representations, companies, computers, context, control systems, internet, software

Procedia PDF Downloads 341

Authors: Hussain Sajid, Jung-Hun Shin, Kum-Won Cho

Abstract:

Image super-resolution is the most common computer vision problem with many important applications. Generative adversarial networks (GANs) have promoted remarkable advances in single-image super-resolution (SR) by recovering photo-realistic images. However, high memory requirements of GAN-based SR (mainly generators) lead to performance degradation and increased energy consumption, making it difficult to implement it onto resource-constricted devices. To relieve such a problem, In this paper, we introduce an optimized and highly efficient architecture for SR-GAN (generator) model by utilizing model compression techniques such as Knowledge Distillation and pruning, which work together to reduce the storage requirement of the model also increase in their performance. Our method begins with distilling the knowledge from a large pre-trained model to a lightweight model using different loss functions. Then, iterative weight pruning is applied to the distilled model to remove less significant weights based on their magnitude, resulting in a sparser network. Knowledge Distillation reduces the model size by 40%; pruning then reduces it further by 18%. To accelerate the learning process, we employ the Horovod framework for distributed training on a cluster of 2 nodes, each with 8 GPUs, resulting in improved training performance and faster convergence. Experimental results on various benchmarks demonstrate that the proposed compressed model significantly outperforms state-of-the-art methods in terms of peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), and image quality for x4 super-resolution tasks.

Keywords: single-image super-resolution, generative adversarial networks, knowledge distillation, pruning

Procedia PDF Downloads 71

Authors: Somaiyeh Dehghan, Mohammad Reza Kheirkhahan Haghighi

Abstract:

The rapid growth of domains that have data with a large number of features, while the number of samples is limited has caused difficulty in constructing strong classifiers. To reduce the dimensionality of the feature space becomes an essential step in classification task. Random subspace method (or attribute bagging) is an ensemble classifier that consists of several classifiers that each base learner in ensemble has subset of features. In the present paper, we introduce Random Subspace Ensemble of CMAC neural network (RSE-CMAC), each of which has training with subset of features. Then we use this model for classification task. For evaluation performance of our model, we compare it with bagging algorithm on 36 UCI datasets. The results reveal that the new model has better performance.

Keywords: classification, random subspace, ensemble, CMAC neural network

Procedia PDF Downloads 315

Authors: Farid Arya, Trevor Hyde

Abstract:

The development of vacuum glazing represents a significant advancement in the area of low heat loss glazing systems with the potential to substantially reduce building heating and cooling loads. Vacuum glazing consists of two or more glass panes hermetically sealed together around the edge with a vacuum gap between the panes. To avoid the glass panes from collapsing and touching each other under the influence of atmospheric pressure an array of support pillars is provided between the glass panes. A high level of thermal insulation is achieved by evacuating the spaces between the glass panes to a very low pressure which greatly reduces conduction and convection within the space; therefore heat transfer through this kind of glazing is significantly lower when compared with conventional insulating glazing. However, vacuum glazing is subject to inherent stresses due to atmospheric pressure and temperature differentials which can lead to fracture of the glass panes and failure of the edge seal. A flexible edge seal has been proposed to minimise the impact of these issues. In this paper, vacuum glazing system with rigid and flexible edge seals is theoretically studied and their advantages and disadvantages are discussed.

Keywords: flexible edge seal, stress, support pillar, vacuum glazing

Procedia PDF Downloads 223

Authors: Mohammad Yahaghifar

Abstract:

Nowadays Wireless system designers have been facing the continuously increasing demand for high data rates and mobility required by new wireless applications. Evolving future communication network generation cellular wireless networks are envisioned to overcome the fundamental challenges of existing cellular networks, for example, higher data rates, excellent end-to-end performance, and user coverage in hot-spots and crowded areas with lower latency,energy consumption and cost per information transfer. In this paper we propose a potential cellular architecture that separates indoor and outdoor scenarios and discuss various promising technologies for future wireless communication systemssystems, such as massive MIMO, energy-efficient communications,cognitive radio networks, and visible light communications and we disscuse about 5G that is next generation of wireless networks.

Keywords: future challenges in networks, cellur architecture, visible light communication, 5G wireless technologies, spatial modulation, massiva mimo, cognitive radio network, green communications

Procedia PDF Downloads 471

Authors: Maria Pavlova

Abstract:

In our days, it is possible to put the camera on different vehicles like quadcopter, train, airplane and etc. The camera also can be the input sensor in many different systems. That means the object recognition like non separate part of monitoring control can be key part of the most intelligent systems. The aim of this paper is to focus of the object recognition process during vehicles movement. During the vehicle’s movement the camera takes pictures from the environment without storage in Data Base. In case the camera detects a special object (for example human or animal), the system saves the picture and sends it to the work station in real time. This functionality will be very useful in emergency or security situations where is necessary to find a specific object. In another application, the camera can be mounted on crossroad where do not have many people and if one or more persons come on the road, the traffic lights became the green and they can cross the road. In this papers is presented the system has solved the aforementioned problems. It is presented architecture of the object recognition system includes the camera, Raspberry platform, GPS system, neural network, software and Data Base. The camera in the system takes the pictures. The object recognition is done in real time using the OpenCV library and Raspberry microcontroller. An additional feature of this library is the ability to display the GPS coordinates of the captured objects position. The results from this processes will be sent to remote station. So, in this case, we can know the location of the specific object. By neural network, we can learn the module to solve the problems using incoming data and to be part in bigger intelligent system. The present paper focuses on the design and integration of the image recognition like a part of smart systems.

Keywords: camera, object recognition, OpenCV, Raspberry

Procedia PDF Downloads 208

Authors: Igor Astrov, Natalya Berezovski

Abstract:

This paper focuses on a critical component of the situational awareness (SA), the control of autonomous vertical flight for vectored thrust aerial vehicle (VTAV). With the SA strategy, we proposed a neural network motion control procedure to address the dynamics variation and performance requirement difference of flight trajectory for a VTAV. This control strategy with using of NARMA-L2 neurocontroller for chosen model of VTAV has been verified by simulation of take-off and forward maneuvers using software package Simulink and demonstrated good performance for fast stabilization of motors, consequently, fast SA with economy in energy can be asserted during search-and-rescue operations.

Keywords: NARMA-L2 neurocontroller, situational awareness, vectored thrust aerial vehicle, aviation

Procedia PDF Downloads 405

Authors: Mehrdad Nouri Khajavi, Gollamhassan Payganeh, Mohsen Fallah Tafti

Abstract:

Fault diagnosis is an important aspect of maintaining rotating machinery health and increasing productivity. Many researches has been done in this regards. Many faults such as unbalance, misalignment, looseness, bearing faults, etc. have been considered and diagnosed with different techniques. Most of the researches in fault diagnosis of rotating machinery deal with single fault. Where as in reality faults usually occur simultaneously and it is, therefore, necessary to recognize them at the same time. In this research, two of the most common faults namely unbalance and misalignment have been considered simultaneously with different intensity and then identified and classified with the use of Multi-Layer Perception Neural Network (MLPNN). Processed Vibration signals are used as the input to the MLPNN, and the class of mixed unbalancy, and misalignment is the output of the NN.

Keywords: unbalance, parallel misalignment, combined faults, vibration signals

Procedia PDF Downloads 338

Authors: Leong Tatt Loh, Ming Kun Yew, Ming Chian Yew, Lip Huat Saw, Jing Han Beh, Siong Kang Lim, Foo Wei Lee

Abstract:

Oil palm shell (OPS) is the solid waste product from the palm oil sector of the agricultural industry and can be used as alternative coarse aggregates to substitute depleting conventional raw materials. This research aims to investigate the incorporation of various high-performance polypropylene (HPP) fibres with different geometry to enhance the mechanical properties and thermal conductivity of OPS lightweight concrete. The effect of different volume fractions (Vf) (0.05%, 0.10% and 0.15%) were studied for each fibre. The results reveal that the effectiveness of HPP fibres to increase the compressive strength at later ages was more pronounced than at early age. It is found that the use of HPP fibres reinforced OPS lightweight concrete (LWC) induced the advantageous of improving mechanical properties (compressive strength, flexural strength and splitting tensile strength) and thermal conductivity. Hence, this HPP fibres is a promising alternative solution to compensate lower mechanical properties as well as contribute to energy efficiency building material in the construction industry.

Keywords: oil palm shell, high performance polypropylene fibre, lightweight concrete, mechanical properties, thermal conductivity

Procedia PDF Downloads 190

Authors: Piyangkun Kukutapan, Siridech Boonsang

Abstract:

The article proposes maximum power point tracking without mechanical sensor using Multilayer Perceptron Neural Network (MLPNN). The aim of article is to reduce the cost and complexity but still retain efficiency. The experimental is that duty cycle is generated maximum power, if it has suitable qualification. The measured data from DC generator, voltage (V), current (I), power (P), turnover rate of power (dP), and turnover rate of voltage (dV) are used as input for MLPNN model. The output of this model is duty cycle for driving the converter. The experiment implemented using Arduino Uno board. This diagram is compared to MPPT using MLPNN and P&O control (Perturbation and Observation control). The experimental results show that the proposed MLPNN based approach is more efficiency than P&O algorithm for this application.

Keywords: maximum power point tracking, multilayer perceptron netural network, optimal duty cycle, DC generator

Procedia PDF Downloads 313

Authors: Jingpeng Gao, Xinrui Mao, Zhibin Deng

Abstract:

Aiming at the low recognition rate on the composite signal modulation in low signal to noise ratio (SNR), this paper proposes a modulation recognition algorithm based on ICanny-CNN. Firstly, the radar signal is transformed into the time-frequency image by Choi-Williams Distribution (CWD). Secondly, we propose an image processing algorithm using the Guided Filter and the threshold selection method, which is combined with the hole filling and the mask operation. Finally, the shallow convolutional neural network (CNN) is combined with the idea of the depth-wise convolution (Dw Conv) and the point-wise convolution (Pw Conv). The proposed CNN is designed to complete image classification and realize modulation recognition of radar signal. The simulation results show that the proposed algorithm can reach 90.83% at 0dB and 71.52% at -8dB. Therefore, the proposed algorithm has a good classification and anti-noise performance in radar signal modulation recognition and other fields.

Keywords: modulation recognition, image processing, composite signal, improved Canny algorithm

Procedia PDF Downloads 176

Authors: Mohammad M. Garmabia, Peyman Shahia, Jimi Tjonga, Mohini Saina

Abstract:

Flame retardant composite filaments with functional carbon in the composition were fabricated, and printed parts showed enhancedcrash resistance pproperties and imporved EMI shielding. The negligible mass difference after prolonged immersion in automobile chemicals revealed the outstanding performance of parts for under-the-hood high-temperature applications.

Keywords: FDM, crash worthy, EMI Shield, lightweight, automotive parts

Procedia PDF Downloads 85

Authors: Nadia Samantha Zuñiga-Peña, Norberto Hernández-Romero, Omar Aguilar-Mejia, Salatiel García-Nava

Abstract:

Using unmanned aerial vehicles (UAVs) for load transport has gained significant importance in various sectors due to their ability to improve efficiency, reduce costs, and access hard-to-reach areas. Although UAVs offer numerous advantages for load transport, several complications and challenges must be addressed to exploit their potential fully. Complexity relays on UAVs are underactuated, non-linear systems with a high degree of coupling between their variables and are subject to forces with uncertainty. One of the biggest challenges is modeling and controlling the system formed by UAVs carrying a load. In order to solve the controller problem, in this work, a hybridization of Neural Network and Hunger Games Search (HGS) metaheuristic algorithm is developed and implemented to find the parameters of the Super Twisting Sliding Mode Controller for the 8 degrees of freedom model of UAV with payload. The optimized controller successfully tracks the UAV through the three-dimensional desired path, demonstrating the effectiveness of the proposed solution. A comparison of performance shows the superiority of the neural network HGS (NNHGS) over the HGS algorithm, minimizing the tracking error by 57.5 %.

Keywords: neural networks, hunger games search, super twisting sliding mode controller, UAVs.

Procedia PDF Downloads 9

Authors: Jesús Antonio Puente Fernández, Luis Javier Garcia Villalba

Abstract:

New trends on streaming videos such as series or films require a high demand of network resources. This fact results in a huge problem within traditional IP networks due to the rigidity of its architecture. In this way, Software Defined Networks (SDN) is a new concept of network architecture that intends to be more flexible and it simplifies the management in networks with respect to the existing ones. These aspects are possible due to the separation of control plane (controller) and data plane (switches). Taking the advantage of this separated control, it is easy to deploy a monitoring tool independent of device vendors since the existing ones are dependent on the installation of specialized and expensive hardware. In this paper, we propose a framework that optimizes the traffic monitoring in SDN networks that decreases the number of monitoring queries to improve the network traffic and also reduces the overload. The performed experiments (with and without the optimization) using a video streaming delivery between two hosts demonstrate the feasibility of our monitoring proposal.

Keywords: optimization, monitoring, software defined networking, statistics, query

Procedia PDF Downloads 315

Authors: Cheng Fang, Lingwei Quan, Cunyue Lu

Abstract:

Human pose estimation and tracking are to accurately identify and locate the positions of human joints in the video. It is a computer vision task which is of great significance for human motion recognition, behavior understanding and scene analysis. There has been remarkable progress on human pose estimation in recent years. However, more researches are needed for human pose tracking especially for online tracking. In this paper, a framework, called PoseSRPN, is proposed for online single-person pose estimation and tracking. We use Siamese network attaching a pose estimation branch to incorporate Single-person Pose Tracking (SPT) and Visual Object Tracking (VOT) into one framework. The pose estimation branch has a simple network structure that replaces the complex upsampling and convolution network structure with deconvolution. By augmenting the loss of fully convolutional Siamese network with the pose estimation task, pose estimation and tracking can be trained in one stage. Once trained, PoseSRPN only relies on a single bounding box initialization and producing human joints location. The experimental results show that while maintaining the good accuracy of pose estimation on COCO and PoseTrack datasets, the proposed method achieves a speed of 59 frame/s, which is superior to other pose tracking frameworks.

Keywords: computer vision, pose estimation, pose tracking, Siamese network

Procedia PDF Downloads 139

Authors: Chi-Yan Chu, Kai-Chi Chuang, Huang-Chung Cheng

Abstract:

In this study, the RRAM devices with the TiN/Ti/HfO_x/TiN structure were fabricated, then the electrical characteristics of the devices without annealing and after 400 °C and 500 °C of the furnace annealing (FA) temperature processes were compared. The RRAM devices after the FA’s 400 °C showed the lower forming, set and reset voltages than the other devices without annealing. However, the RRAM devices after the FA’s 500 °C did not show any electrical characteristics because the TiN/Ti/HfO_x/TiN device was oxidized, as shown in the XPS analysis. From these results, the RRAM devices after the FA’s 400 °C showed the best electrical characteristics.

Keywords: RRAM, furnace annealing (FA), forming, set and reset voltages, XPS

Procedia PDF Downloads 360

Authors: Shunzuo Wu, Xudong Luo, Yuanxiu Liao

Abstract:

Recently, the rapid development of deep learning makes artificial intelligence (AI) penetrate into many fields, replacing manual work there. In particular, AI systems also become a research focus in the field of automatic office. To meet real needs in automatic officiating, in this paper we develop an automatic form filling system. Specifically, it uses two classical neural network models and several word embedding models to classify various relevant information elicited from the Internet. When training the neural network models, we use less noisy and balanced data for training. We conduct a series of experiments to test my systems and the results show that our system can achieve better classification results.

Keywords: artificial intelligence and office, NLP, deep learning, text classification

Procedia PDF Downloads 174

Authors: Vivek Kumar Singh, Tripti Singh

Abstract:

An ad hoc wireless network consists of mobile networks which creates an underlying architecture for communication without the help of traditional fixed-position routers. Ad-hoc On-demand Distance Vector (AODV) is a routing protocol used for Mobile Ad hoc Network (MANET). Nevertheless, the architecture must maintain communication routes although the hosts are mobile and they have limited transmission range. There are different protocols for handling the routing in the mobile environment. Routing protocols used in fixed infrastructure networks cannot be efficiently used for mobile ad-hoc networks, so that MANET requires different protocols. This paper presents the performance analysis of the routing protocols used various parameter-patterns with Two-ray model.

Keywords: AODV, packet transmission rate, pause time, ZRP, QualNet 6.1

Procedia PDF Downloads 810

Authors: Meriem Mossaddek, El Mehdi Laadissi, El Mehdi Loualid, Chouaib Ennawaoui, Sohaib Bouzaid, Abdelowahed Hajjaji

Abstract:

An electrochemical system is a subset of mechatronic systems that includes a wide variety of batteries and nickel-cadmium, lead-acid batteries, and lithium-ion. Those structures have several non-linear behaviors and uncertainties in their running range. This paper studies an effective technique for modeling Lithium-Ion (Li-Ion) batteries using a Nonlinear Auto-Regressive model with exogenous input (NARX). The Artificial Neural Network (ANN) is trained to employ the data collected from the battery testing process. The proposed model is implemented on a Li-Ion battery cell. Simulation of this model in MATLAB shows good accuracy of the proposed model.

Keywords: lithium-ion battery, neural network, energy storage, battery model, nonlinear models

Procedia PDF Downloads 90

Authors: Hrishi Rakshit, Pooneh Bagheri Zadeh

Abstract:

This paper investigates the performance of drone classification methods using conventional DCNN with different hyperparameters, when additional drone audio data is embedded in the dataset for training and further classification. In this paper, first a custom dataset is created using different images of drones from University of South California (USC) datasets and Leeds Beckett university datasets with embedded drone audio signal. The three well-known DCNN architectures namely, Resnet50, Darknet53 and Shufflenet are employed over the created dataset tuning their hyperparameters such as, learning rates, maximum epochs, Mini Batch size with different optimizers. Precision-Recall curves and F1 Scores-Threshold curves are used to evaluate the performance of the named classification algorithms. Experimental results show that Resnet50 has the highest efficiency compared to other DCNN methods.

Keywords: drone classifications, deep convolutional neural network, hyperparameters, drone audio signal

Procedia PDF Downloads 86

Authors: Vijaya Yuvaram Singh V M, Kameshwar Rao J V

Abstract:

The challenge with development of neural network based methods for medical is the availability of data. Anatomical landmark detection in the medical domain is a process to find points on the x-ray scan report of the patient. Most of the time this task is done manually by trained professionals as it requires precision and domain knowledge. Traditionally object detection based methods are used for landmark detection. Here, we utilize reinforcement learning and query based method to train a single agent capable of detecting multiple landmarks. A deep Q network agent is trained to detect single and multiple landmarks present on hip and shoulder from x-ray scan of a patient. Here a single agent is trained to find multiple landmark making it superior to having individual agents per landmark. For the initial study, five images of different patients are used as the environment and tested the agents performance on two unseen images.

Keywords: reinforcement learning, medical landmark detection, multi target detection, deep neural network

Procedia PDF Downloads 130