Search results for: deep layer

Authors: Susan Diamond

Abstract:

Machine Learning workloads have traditionally been run in high-performance computing (HPC) environments, where users log in to dedicated machines and utilize the attached GPUs to run training jobs on huge datasets. Training of large neural network models is very resource intensive, and even after exploiting parallelism and accelerators such as GPUs, a single training job can still take days. Consequently, the cost of hardware is a barrier to entry. Even when upfront cost is not a concern, the lead time to set up such an HPC environment takes months from acquiring hardware to set up the hardware with the right set of firmware, software installed and configured. Furthermore, scalability is hard to achieve in a rigid traditional lab environment. Therefore, it is slow to react to the dynamic change in the artificial intelligent industry. Watson Deep Learning as a service, a cloud-based deep learning platform that mitigates the long lead time and high upfront investment in hardware. It enables robust and scalable sharing of resources among the teams in an organization. It is designed for on-demand cloud environments. Providing a similar user experience in a multi-tenant cloud environment comes with its own unique challenges regarding fault tolerance, performance, and security. Watson Deep Learning as a service tackles these challenges and present a deep learning stack for the cloud environments in a secure, scalable and fault-tolerant manner. It supports a wide range of deep-learning frameworks such as Tensorflow, PyTorch, Caffe, Torch, Theano, and MXNet etc. These frameworks reduce the effort and skillset required to design, train, and use deep learning models. Deep Learning as a service is used at IBM by AI researchers in areas including machine translation, computer vision, and healthcare. 

Keywords: deep learning, machine learning, cognitive computing, model training

Procedia PDF Downloads 209

Authors: Bowen Ding, Yihao Kuang

Abstract:

In recent years, with the development of large-scale pre-trained lan-guage models, building vector representations of text through deep neural network models has become a standard practice for natural language processing tasks. From the performance on downstream tasks, we can know that the text representation constructed by these models contains linguistic information, but its encoding mode and extent are unclear. In this work, a structural probe is proposed to detect whether the vector representation produced by a deep neural network is embedded with a syntax tree. The probe is trained with the deep metric learning method, so that the distance between word vectors in the metric space it defines encodes the distance of words on the syntax tree, and the norm of word vectors encodes the depth of words on the syntax tree. The experiment results on ELMo and BERT show that the syntax tree is encoded in their parameters and the word representations they produce.

Keywords: deep metric learning, syntax tree probing, natural language processing, word representations

Procedia PDF Downloads 68

Authors: Faouzi Ahtchi-Ali, Marcus Vitiello

Abstract:

A study was conducted in the east region of the Middle East to assess the constructability of a shoring system for a 12-meter deep excavation. Several shoring systems were considered in this study including secant concrete piling, contiguous concrete piling, and sheet-piling. The excavation was carried out in a very dense sand with the groundwater level located at 3 meters below ground surface. The study included conducting a pilot test for each shoring system listed above. The secant concrete piling included overlapping concrete piles to a depth of 16 meters. Drilling method with full steel casing was utilized to install the concrete piles. The verticality of the piles was a concern for the overlap. The contiguous concrete piling required the installation of micro-piles to seal the gap between the concrete piles. This method revealed that the gap between the piles was not fully sealed as observed by the groundwater penetration to the excavation. The sheet-piling method required pre-drilling due to the high blow count of the penetrated layer of saturated sand. This study concluded that the sheet-piling method with pre-drilling was the most cost effective and recommended a method for the shoring system.

Keywords: excavation, shoring system, middle east, Drilling method

Procedia PDF Downloads 468

Authors: Ratnadip Natwarbhai Bhoi

Abstract:

The objective of this research paper is to study waste tire crumb rubber granules as a partial concrete replacement by the different percentages of facing layer thickness and without facing layer in the production of rubber cement composite paver block. The physical properties of RCCRP compressive strength, flexural strength, abrasion strength density, and water absorption testing by the IS 15658:2006 method. All these physical properties depend upon the ratio of crumb rubber uses. The result showed that the with facing layer at 15 mm, 25 mm, totally rubberized and without facing layer had little effect on compressive strength, flexural strength and abrasion resistance properties. Water absorption is also important for the service life of the product. The crumb rubber paver block also performed quite well in both compressive strength and abrasion resistance. The rubber cement composite rubber paver block is suitable for nonstructural purposes, such as being lightweight and easy installation for the walkway, sidewalks, and playing area applications.

Keywords: rubber cement, crumb rubber, composite, layer

Procedia PDF Downloads 98

Authors: Mayank Raj, V. G. Narendra

Abstract:

Ever since the DARPA challenge on autonomous vehicles in 2005, there has been a lot of buzz about ‘Autonomous Vehicles’ amongst the major tech giants such as Google, Uber, and Tesla. Numerous approaches have been adopted to solve this problem, which can have a long-lasting impact on mankind. In this paper, we have used Deep Learning techniques and TensorFlow framework with the goal of building a neural network model to predict (speed, acceleration, steering angle, and brake) features needed for navigation of autonomous vehicles. The Deep Neural Network has been trained on images and sensor data obtained from the comma.ai dataset. A heatmap was used to check for correlation among the features, and finally, four important features were selected. This was a multivariate regression problem. The final model had five convolutional layers, followed by five dense layers. Finally, the calculated values were tested against the labeled data, where the mean squared error was used as a performance metric.

Keywords: autonomous vehicles, deep learning, computer vision, artificial intelligence

Procedia PDF Downloads 158

Authors: Sang Hoon Lee, Tae Il Lee, Su Jeong Lee, Jae Min Myoung

Abstract:

In order to manufacture short gap single Si nanowire (NW) field effect transistor (FET) by imprinting and transferring method, we introduce the method using Al2O3 sacrificial layer. The diameters of cylindrical Si NW addressed between Au electrodes by dielectrophoretic (DEP) alignment method are controlled to 106, 128, and 148 nm. After imprinting and transfer process, cylindrical Si NW is embedded in PVP adhesive and dielectric layer. By curing transferred cylindrical Si NW and Au electrodes on PVP-coated p++ Si substrate with 200nm-thick SiO2, 3μm gap Si NW FET fabrication was completed. As the diameter of embedded Si NW increases, the mobility of FET increases from 80.51 to 121.24 cm2/V•s and the threshold voltage moves from –7.17 to –2.44 V because the ratio of surface to volume gets reduced.

Keywords: Al2O3 sacrificial transfer layer, cylindrical silicon nanowires, dielectrophorestic alignment, field effect transistor

Procedia PDF Downloads 457

Authors: A. Gopala Krishna, M. Lakshmi Chaitanya, V. Kalyana Manohar

Abstract:

In the existent analysis, laser micro machining (LMM) of Silicon carbide (SiCp) reinforced Aluminum 7075 Metal Matrix Composite (Al7075/SiCp MMC) was studied. While machining, Because of the intense heat generated, A layer gets formed on the work piece surface which is called recast layer and this layer is detrimental to the surface quality of the component. The recast layer needs to be as small as possible for precise applications. Therefore, The height of recast layer and the depth of groove which are conflicting in nature were considered as the significant manufacturing criteria, Which determines the pursuit of a machining process obtained in LMM of Al7075/10%SiCp composite. The present work formulates the depth of groove and height of recast layer in relation to the machining parameters using the Response Surface Methodology (RSM) and correspondingly, The formulated mathematical models were put to use for optimization. Since the effect of machining parameters on the depth of groove and height of recast layer was contradictory, The problem was explicated as a multi objective optimization problem. Moreover, An evolutionary Non-dominated sorting genetic algorithm (NSGA-II) was employed to optimize the model established by RSM. Subsequently this algorithm was also adapted to achieve the Pareto optimal set of solutions that provide a detailed illustration for making the optimal solutions. Eventually experiments were conducted to affirm the results obtained from RSM and NSGA-II.

Keywords: Laser Micro Machining (LMM), depth of groove, Height of recast layer, Response Surface Methodology (RSM), non-dominated sorting genetic algorithm

Procedia PDF Downloads 345

Authors: G. Sivakumar, T. Pratyusha, D. Gupta, W. Shen

Abstract:

To get ohmic contact between the cathode and organic semiconductor, transport layers are introduced between the active layer and the electrodes. Generally zinc oxide or titanium dioxide are used as electron transport layer. When electron transport layer is doped with lithium, the resultant film exhibited superior electronic properties, which enables faster electron transport. Doping is accomplished by heat treatment of films with Lithium salts. Li-doped films. We fabricated organic solar cell using PTB7(poly(3-hexylthiopene-2,5- diyl):PCBM(phenyl-C61-butyric acid methyl ester) and found that the solar cells prepared using Li doped films had better performance in terms of efficiency when compared to the undoped transport layers.

Keywords: electron transport layer, higher efficiency, lithium doping, ohmic contact

Procedia PDF Downloads 513

Authors: Fatemeh Sadat Sharifi

Abstract:

In the present study, the capabilities of WAVEWATCH-III and MIKE21-SW for predicting the characteristics of wind waves in Hormuz Strait are evaluated. The GFS wind data (Global Forecast System) were derived. The bathymetry of gride with 2 arc-minute resolution, also were extracted from the ETOPO1. WAVEWATCH-III findings illustrate more valid prediction of wave features comparing to the MIKE-21 SW in deep water. Apparently, in shallow area, the MIKE-21 provides more uniformities with altimetry measurements. This may be due to the merits of the unstructured grid which are used in MIKE-21, leading to better representations of the coastal area. The findings on the direction of waves generated by wind in the modeling area indicate that in some regions, despite the increase in wind speed, significant wave height stays nearly unchanged. This is fundamental because of swift changes in wind track over the Strait of Hormuz. After discussing wind-induced waves in the region, the impact of instability of the surface layer on wave growth has been considered. For this purpose, the average monthly mean air temperature has been used. The results in cold months, when the surface layer is unstable, indicates an acceptable increase in the accuracy of prediction of the indicator wave height.

Keywords: numerical modeling, WAVEWATCH-III, Strait of Hormuz, MIKE21-SW

Procedia PDF Downloads 207

Authors: Chaima Babi, Said Gadri

Abstract:

The rapid increase in fake news continues to grow at a very fast rate; this requires implementing efficient techniques that allow testing the re-liability of online content. For that, the current research strives to illuminate the fake news problem using deep learning DL and machine learning ML ap-proaches. We have developed the traditional LSTM (Long short-term memory), and the bidirectional BiLSTM model. A such process is to perform a training task on almost of samples of the dataset, validate the model on a subset called the test set to provide an unbiased evaluation of the final model fit on the training dataset, then compute the accuracy of detecting classifica-tion and comparing the results. For the programming stage, we used Tensor-Flow and Keras libraries on Python to support Graphical Processing Units (GPUs) that are being used for developing deep learning applications.

Keywords: machine learning, deep learning, natural language, fake news, Bi-LSTM, LSTM, multiclass classification

Procedia PDF Downloads 95

Authors: A. Soualem

Abstract:

The design of multi stage deep drawing processes requires the evaluation of many process parameters such as the intermediate die geometry, the blank shape, the sheet thickness, the blank holder force, friction, lubrication etc..These process parameters have to be determined for the optimum forming conditions before the process design. In general sheet metal forming may involve stretching drawing or various combinations of these basic modes of deformation. It is important to determine the influence of the process variables in the design of sheet metal working process. Especially, the punch and die corner for deep drawing will affect the formability. At the same time the prediction of sheet metals springback after deep drawing is an important issue to solve for the control of manufacturing processes. Nowadays, the importance of this problem increases because of the use of steel sheeting with high stress and also aluminum alloys. The aim of this paper is to give a better understanding of the springback and its effect in various sheet metals forming process such as expansion and restraint deep drawing in the cup drawing process, by varying radius die, lubricant for two commercially available materials e.g. galvanized steel and Aluminum sheet. To achieve these goals experiments were carried out and compared with other results. The original of our purpose consist on tests which are ensured by adapting a U-type stretching-bending device on a tensile testing machine, where we studied and quantified the variation of the springback.

Keywords: springback, deep drawing, expansion, restricted deep drawing

Procedia PDF Downloads 454

Authors: Sumaiya Amir Khan, Shayma Al-Sharif, Samiha Mazher, Neha Intikhab Khan

Abstract:

This research paper aims to evaluate the effectiveness of sleep-tracking AI applications in smart-watches. It focuses on comparing the sleep analyses of two different smartwatch brands, Samsung and Fitbit, and measuring sleep at three different stages – REM (Rapid-Eye-Movement), NREM (Non-Rapid-Eye-Movement), and deep sleep. The methodology involves the participation of different users and analyzing their sleep data. The results reveal that although light sleep is the longest stage, deep sleep is higher than average in the participants. The study also suggests that light sleep is not uniform, and getting higher levels of deep sleep can prevent debilitating health conditions. Based on the findings, it is recommended that individuals should aim to achieve higher levels of deep sleep to maintain good health. Overall, this research contributes to the growing literature on the effectiveness of sleep-tracking AI applications and their potential to improve sleep quality.

Keywords: sleep tracking, lifestyle, accuracy, health, AI, AI features, ML

Procedia PDF Downloads 79

Authors: Ryu Wonjin, Choi Hyeok, Park Joonhong

Abstract:

Overlay welding for improving surface properties is a method of the surface treatments which improve surface properties of material by welding materials of alloy having corrosion resistance on the basic material surface. Overlay welding affects contents of chemical components and weld hardness from different parts by dilution of the lamination layer thickness, and it determines surface properties. Therefore, overlay welding has to take into account thickness of the lamination layers with the process. As a result in this study examined contents of Fe, weldability of the base metal and monel materials, hardness and surface flatness from different parts according to each the lamination layer parameters by overlay welding monel materials with corrosion resources to the base material of carbon steel. Through this, evaluated effect by the lamination layer parameters of welding and presented decision methods of the lamination layer parameters of the overlay welding by the purpose of use.

Keywords: clad pipe, lamination layer parameters, monel, overlay welding

Procedia PDF Downloads 273

Authors: Shi-Uk Lee, Chae Young Lim

Abstract:

Objective: The aim of this study was to evaluate the changes of lumbar deep muscle thickness and cross-sectional area using ultrasonography with magnetic stimulation. Methods: To evaluate the changes of lumbar deep muscle by using magnetic stimulation, 12 healthy volunteers (39.6±10.0 yrs) without low back pain during 3 months participated in this study. All the participants were checked with X-ray and electrophysiologic study to confirm that they had no problems with their back. Magnetic stimulation was done on the L5 and S1 root with figure-eight coil as previous study. To confirm the proper motor root stimulation, the surface electrode was put on the tibialis anterior (L5) and abductor hallucis muscles (S1) and the hot spots of magnetic stimulation were found with 50% of maximal magnetic stimulation and determined the stimulation threshold lowering the magnetic intensity by 5%. Ultrasonography was used to assess the changes of L5 and S1 lumbar multifidus (superficial and deep) cross-sectional area and thickness with maximal magnetic stimulation. Cross-sectional area (CSA) and thickness was evaluated with image acquisition program, ImageJ software (National Institute of Healthy, USA). Wilcoxon signed-rank was used to compare outcomes between before and after stimulations. Results: The mean minimal threshold was 29.6±3.8% of maximal stimulation intensity. With minimal magnetic stimulation, thickness of L5 and S1 deep multifidus (DM) were increased from 1.25±0.20, 1.42±0.23 cm to 1.40±0.27, 1.56±0.34 cm, respectively (P=0.005, P=0.003). CSA of L5 and S1 DM were also increased from 2.26±0.18, 1.40±0.26 cm2 to 2.37±0.18, 1.56±0.34 cm2, respectively (P=0.002, P=0.002). However, thickness of L5 and S1 superficial multifidus (SM) were not changed from 1.92±0.21, 2.04±0.20 cm to 1.91±0.33, 1.96±0.33 cm (P=0.211, P=0.199) and CSA of L5 and S1 were also not changed from 4.29±0.53, 5.48±0.32 cm2 to 4.42±0.42, 5.64±0.38 cm2. With maximal magnetic stimulation, thickness of L5, S1 of DM and SM were increased (L5 DM, 1.29±0.26, 1.46±0.27 cm, P=0.028; L5 SM, 2.01±0.42, 2.24±0.39 cm, P=0.005; S1 DM, 1.29±0.19, 1.67±0.29 P=0.002; S1 SM, 1.90±0.36, 2.30±0.36, P=0.002). CSA of L5, S1 of DM and SM were also increased (all P values were 0.002). Conclusions: Deep lumbar muscles could be stimulated with lumbar motor root magnetic stimulation. With minimal stimulation, thickness and CSA of lumbosacral deep multifidus were increased in this study. Further studies are needed to confirm whether the similar results in chronic low back pain patients are represented. Lumbar magnetic stimulation might have strengthening effect of deep lumbar muscles with no discomfort.

Keywords: magnetic stimulation, lumbar multifidus, strengthening, ultrasonography

Procedia PDF Downloads 371

Authors: Sam Khozama, Ali M. Mayya

Abstract:

Breast cancer is one of the most common types in women. Early prediction of breast cancer helps physicians detect cancer in its early stages. Big cancer data needs a very powerful tool to analyze and extract predictions. Machine learning and deep learning are two of the most efficient tools for predicting cancer based on textual data. In this study, we developed a fusion model of two machine learning and deep learning models. To obtain the final prediction, Long-Short Term Memory (LSTM) and ensemble learning with hyper parameters optimization are used, and score-level fusion is used. Experiments are done on the Breast Cancer Surveillance Consortium (BCSC) dataset after balancing and grouping the class categories. Five different training scenarios are used, and the tests show that the designed fusion model improved the performance by 3.3% compared to the individual models.

Keywords: machine learning, deep learning, cancer prediction, breast cancer, LSTM, fusion

Procedia PDF Downloads 163

Authors: Yan Bachtiar Muslih, Hangga Wijaya, Trio Fani, Putri Agustin

Abstract:

Demand of Energy in Indonesia always increases 5-6% a year, but production of conventional energy always decreases 3-5% a year, it means that conventional energy in 20-40 years ahead will not able to complete all energy demand in Indonesia, one of the solve way is using unconventional energy that is gas hydrate, gas hydrate is gas that form by biogenic process, gas hydrate stable in condition with extremely depth and low temperature, gas hydrate can form in two condition that is in pole condition and in deep-sea condition, wherein this research will focus in gas hydrate that association with methane form methane hydrate in deep-sea condition and usually form in depth between 150-2000 m, this research will focus in process of methane hydrate formation that is biogenic process and the important role of deep-sea sediment so can produce accumulation of methane hydrate, methane hydrate usually will be accumulated in find sediment in deep-sea environment with condition high-pressure and low-temperature this condition too usually make methane hydrate change into white nodule, methodology of this research is geology field work and laboratory analysis, from geology field work will get sample data consist of 10-15 samples from Kerek Formation outcrops as random for imagine the condition of deep-sea environment that influence the methane hydrate formation and also from geology field work will get data of measuring stratigraphy in outcrops Kerek Formation too from this data will help to imagine the process in deep-sea sediment like energy flow, supply sediment, and etc, and laboratory analysis is activity to analyze all data that get from geology field work, the result of this research can used to exploration activity of methane hydrate in another prospect deep-sea environment in Indonesia.

Keywords: methane hydrate, deep-sea sediment, kerek formation, sub-basin of kendeng, central java, Indonesia

Procedia PDF Downloads 462

Authors: Yuanchao Sang, Zhihao Gong, Longsheng Chen, Long Chen

Abstract:

Clothing fashion represents a human’s aesthetic appreciation towards everyday outfits and appetite for fashion, and it reflects the development of status in society, humanity, and economics. However, modelling fashion by machine is extremely challenging because fashion is too abstract to be efficiently described by machines. Even human beings can hardly reach a consensus about fashion. In this paper, we are dedicated to answering a fundamental fashion-related problem: what image feature best describes clothing fashion? To address this issue, we have designed and evaluated various image features, ranging from traditional low-level hand-crafted features to mid-level style awareness features to various current popular deep neural network-based features, which have shown state-of-the-art performance in various vision tasks. In summary, we tested the following 9 feature representations: color, texture, shape, style, convolutional neural networks (CNNs), CNNs with distance metric learning (CNNs&DML), AutoEncoder, CNNs with multiple layer combination (CNNs&MLC) and CNNs with dynamic feature clustering (CNNs&DFC). Finally, we validated the performance of these features on two publicly available datasets. Quantitative and qualitative experimental results on both intra-domain and inter-domain fashion clothing image retrieval showed that deep learning based feature representations far outweigh traditional hand-crafted feature representation. Additionally, among all deep learning based methods, CNNs with explicit feature clustering performs best, which shows feature clustering is essential for discriminative fashion feature representation.

Keywords: convolutional neural network, feature representation, image processing, machine modelling

Procedia PDF Downloads 139

Authors: Jenny Radeva, Anke-Gundula Roth, Christian Goebbert, Robert Niestroj-Pahl, Lars Daehne, Axel Wolfram, Juergen Wiese

Abstract:

Ceramic membranes for water purification combine excellent stability with long-life characteristics and high chemical resistance. Layer-by-Layer coating is a well-known technique for customization and optimization of filtration properties of membranes but is mostly used on polymeric membranes. Ceramic membranes comprising a metal oxide filtration layer of Al2O3 or TiO2 are charged and therefore highly suitable for polyelectrolyte adsorption. The high stability of the membrane support allows efficient backwash and chemical cleaning of the membrane. The presented study reports metal oxide/organic composite membrane with an increased rejection of bivalent salts like MgSO4 and the organic micropollutant Diclofenac. A self-build apparatus was used for applying the polyelectrolyte multilayers on the ceramic membrane. The device controls the flow and timing of the polyelectrolytes and washing solutions. As support for the Layer-by-Layer coat, ceramic mono-channel membranes were used with an inner capillary of 8 mm diameter, which is connected to the coating device. The inner wall of the capillary is coated subsequently with polycat- and anions. The filtration experiments were performed with a feed solution of MgSO4 and Diclofenac. The salt content of the permeate was detected conductometrically and Diclofenac was measured with UV-Adsorption. The concluded results show retention values of magnesium sulfate of 70% and diclofenac retention of 60%. Further experimental research studied various parameters of the composite membrane-like Molecular Weight Cut Off and pore size, Zeta potential and its mechanical and chemical robustness.

Keywords: water purification, polyelectrolytes, membrane modification, layer-by-layer coating, ceramic membranes

Procedia PDF Downloads 245

Authors: Mohamed Abd Elfattah Ibraheem Elghrbawy

Abstract:

Icons are a significant group of cultural heritage objects that deserve to be maintained and conserved, as these ions are performed according to religious standards and norms. The ideal structure of icons is five strata, the lower layer is a wood plate, and the upper layer is the varnish layer that is exposed to photo-oxidation, that is turned into a fragile yellow layer. In addition, the components of the icons are important in dating these ions, so X-Ray Fluorescence (XRF), X-Ray Diffraction (XRD), and Scanning Electron Microscopy with energy dispersive X-ray spectroscopy (SEM-EDX) patterns were used. SEM-EDX pattern revealed that the red pigment was vermillion (HgS), that was used in the late period, with a slight difference from the synthesized pigment. Pigments were subjected to chromatic alteration due to different agents, such as microbial agents and pollutants, in particular SO₂, whereas the pigment-based pigments are more sensitive. Moreover, cleaning, varnish removal, and retouching are important processes in the conservation of icons.

Keywords: conservation, cultural heritage, Egyptian icon, pigments

Procedia PDF Downloads 79

Authors: Cahit Perkgöz

Abstract:

Artificial intelligence has brought new approaches to solving problems in almost every research field in recent years. One of these topics is shape design and optimization, which has the possibility of applications in many fields, such as nanotechnology and electronics. A properly constructed cost function can eliminate the need for labeled data required in deep learning and create desired shapes. In this work, the network parameters are optimized differentially, which differs from traditional approaches. The methods are tested for physics-related structures and successful results are obtained. This work is supported by Eskişehir Technical University scientific research project (Project No: 20ADP090)

Keywords: deep learning, shape design, optimization, artificial intelligence

Procedia PDF Downloads 152

Authors: Noureddine Mohtaram, Jeremy Patrix, Jerome Verny

Abstract:

As an enormous number of online services have been developed into web applications, security problems based on web applications are becoming more serious now. Most intrusion detection systems rely on each request to find the cyber-attack rather than on user behavior, and these systems can only protect web applications against known vulnerabilities rather than certain zero-day attacks. In order to detect new attacks, we analyze the HTTP protocols of web servers to divide them into two categories: normal attacks and malicious attacks. On the other hand, the quality of the results obtained by deep learning (DL) in various areas of big data has given an important motivation to apply it to cybersecurity. Deep learning for attack detection in cybersecurity has the potential to be a robust tool from small transformations to new attacks due to its capability to extract more high-level features. This research aims to take a new approach, deep learning to cybersecurity, to classify these two categories to eliminate attacks and protect web servers of the defense sector which encounters different web traffic compared to other sectors (such as e-commerce, web app, etc.). The result shows that by using a machine learning method, a higher accuracy rate, and a lower false alarm detection rate can be achieved.

Keywords: anomaly detection, HTTP protocol, logs, cyber attack, deep learning

Procedia PDF Downloads 211

Authors: Nubwimana Rachel, Mugabowindekwe Maurice

Abstract:

Trees play a crucial role in maintaining biodiversity and providing various ecological services. Traditional methods of counting trees are time-consuming, and there is a need for more efficient techniques. However, deep learning makes it feasible to identify the multi-scale elements hidden in aerial imagery. This research focuses on the application of deep learning techniques for tree detection and counting in both forest and non-forest areas through the exploration of the deep learning application for automated tree detection and counting using satellite imagery. The objective is to identify the most effective model for automated tree counting. We used different deep learning models such as YOLOV7, SSD, and UNET, along with Generative Adversarial Networks to generate synthetic samples for training and other augmentation techniques, including Random Resized Crop, AutoAugment, and Linear Contrast Enhancement. These models were trained and fine-tuned using satellite imagery to identify and count trees. The performance of the models was assessed through multiple trials; after training and fine-tuning the models, UNET demonstrated the best performance with a validation loss of 0.1211, validation accuracy of 0.9509, and validation precision of 0.9799. This research showcases the success of deep learning in accurate tree counting through remote sensing, particularly with the UNET model. It represents a significant contribution to the field by offering an efficient and precise alternative to conventional tree-counting methods.

Keywords: remote sensing, deep learning, tree counting, image segmentation, object detection, visualization

Procedia PDF Downloads 71

Authors: Nazmul Shahadat, Anthony S. Maida

Abstract:

In recent years, hypercomplex inspired neural networks improved deep CNN architectures due to their ability to share weights across input channels and thus improve cohesiveness of representations within the layers. The work described herein studies the effect of replacing existing layers in an Axial Attention ResNet with their quaternion variants that use cross-channel weight sharing to assess the effect on image classification. We expect the quaternion enhancements to produce improved feature maps with more interlinked representations. We experiment with the stem of the network, the bottleneck layer, and the fully connected backend by replacing them with quaternion versions. These modifications lead to novel architectures which yield improved accuracy performance on the ImageNet300k classification dataset. Our baseline networks for comparison were the original real-valued ResNet, the original quaternion-valued ResNet, and the Axial Attention ResNet. Since improvement was observed regardless of which part of the network was modified, there is a promise that this technique may be generally useful in improving classification accuracy for a large class of networks.

Keywords: axial attention, representational networks, weight sharing, cross-channel correlations, quaternion-enhanced axial attention, deep networks

Procedia PDF Downloads 83

Authors: Ramzi Rihane, Yassine Benayed

Abstract:

Epilepsy is a chronic neurological disorder characterized by recurrent, unprovoked seizures resulting from abnormal electrical activity in the brain. Timely and accurate detection of these seizures is essential for improving patient care. In this study, we leverage the UK Bonn University open-source EEG dataset and employ advanced deep-learning techniques to automate the detection of epileptic seizures. By extracting key features from both time and frequency domains, as well as Spectrogram features, we enhance the performance of various deep learning models. Our investigation includes architectures such as Long Short-Term Memory (LSTM), Bidirectional LSTM (Bi-LSTM), 1D Convolutional Neural Networks (1D-CNN), and hybrid CNN-LSTM and CNN-BiLSTM models. The models achieved impressive accuracies: LSTM (98.52%), Bi-LSTM (98.61%), CNN-LSTM (98.91%), CNN-BiLSTM (98.83%), and CNN (98.73%). Additionally, we utilized a data augmentation technique called SMOTE, which yielded the following results: CNN (97.36%), LSTM (97.01%), Bi-LSTM (97.23%), CNN-LSTM (97.45%), and CNN-BiLSTM (97.34%). These findings demonstrate the effectiveness of deep learning in capturing complex patterns in EEG signals, providing a reliable and scalable solution for real-time seizure detection in clinical environments.

Keywords: electroencephalogram, epileptic seizure, deep learning, LSTM, CNN, BI-LSTM, seizure detection

Procedia PDF Downloads 12

Authors: Muhammad Zubair Khan, Yugyung Lee

Abstract:

Deep learning has recently achieved enormous response in semantic image segmentation. The previously developed U-Net inspired architectures operate with continuous stride and pooling operations, leading to spatial data loss. Also, the methods lack establishing long-term pixels connection to preserve context knowledge and reduce spatial loss in prediction. This article developed encoder-decoder architecture with bi-directional LSTM embedded in long skip-connections and densely connected convolution blocks. The network non-linearly combines the feature maps across encoder-decoder paths for finding dependency and correlation between image pixels. Additionally, the densely connected convolutional blocks are kept in the final encoding layer to reuse features and prevent redundant data sharing. The method applied batch-normalization for reducing internal covariate shift in data distributions. The empirical evidence shows a promising response to our method compared with other semantic segmentation techniques.

Keywords: deep learning, semantic segmentation, image analysis, pixels connection, convolution neural network

Procedia PDF Downloads 102

Authors: Larisa Liamina, Vasily Lavrushin, Salvatore Inguaggiato

Abstract:

This study presents and summarizes the results of researching the isotopic signature of helium in the deep fluid eastern part of the Southern slope of the Greater Caucasus and the Lesser Caucasus (Azerbaijan and Armenia) for the period from 2010 to 2016. The results of isotope ratios of 3He/4He in 59 samples of the gas phase of geothermal fluids and mud volcanoes are presented. New data have been obtained not only on the isotopic ratios of helium, but also neon and argon. The R/Ra ratio was analyzed along the Ankara-Sevan ophiolite structure. The patterns of lateral variations of the 3He/4He ratio of different geological structural elements of the studied region are revealed.

Keywords: isotopes helium, deep fluids, tectonic structures, Caucasus

Procedia PDF Downloads 45

Authors: Xinxin Ma, Guangze Tang, Shuxin Yang, Jinguang He, Fan Zhang, Peiling Sun, Ming Liu, Minyu Sun, Liqin Wang

Abstract:

As a traditional surface hardening technique, shot peening is widely used in industry. By using shot peening, a residual compressive stress is formed in the surface which is beneficial for improving the fatigue life of metal materials. At the same time, very fine grains and high density defects are generated in the surface layer which enhances the surface hardness, either. However, most of the processes are carried out at room temperature. For high strength steel, such as M50, the thickness of the strengthen layer is limited. In order to obtain a thick strengthen surface layer, elevated temperature shot peening was carried out in this work by using Φ1mm cast ion balls with a speed of 80m/s. Considering the tempering temperature of M50 steel is about 550 oC, the processing temperature was in the range from 300 to 500 oC. The effect of processing temperature and processing time of shot peening on distribution of residual stress and surface hardness was investigated. As we known, the working temperature of M50 steel can be as high as 315 oC. Because the defects formed by shot peening are unstable when the working temperature goes higher, it is worthy to understand what happens during the shot peening process, and what happens when the strengthen samples were kept at a certain temperature. In our work, the shot peening time was selected from 2 to 10 min. And after the strengthening process, the samples were annealed at various temperatures from 200 to 500 oC up to 60 h. The results show that the maximum residual compressive stress is near 900 MPa. Compared with room temperature shot peening, the strengthening depth of 500 oC shot peening sample is about 2 times deep. The surface hardness increased with the processing temperature, and the saturation peening time decreases. After annealing, the residual compressive stress decreases, however, for 500 oC peening sample, even annealing at 500 oC for 20 h, the residual compressive stress is still over 600 MPa. However, it is clean to see from SEM that the grain size of surface layers is still very small.

Keywords: shot peening, M50 steel, residual compressive stress, elevated temperature

Procedia PDF Downloads 456

Authors: Farshad Zeinali, Sajedeh Norouzi, Nader Mokari, Eduard Jorswieck

Abstract:

The capacity of fifth-generation (5G) vehicle-to-everything (V2X) networks poses significant challenges. To ad- dress this challenge, this paper utilizes New Radio (NR) and New Radio Unlicensed (NR-U) networks to develop a heterogeneous vehicular network (HetNet). We propose a new framework, named joint BS assignment and resource allocation (JBSRA) for mobile V2X users and also consider coexistence schemes based on flexible duty cycle (DC) mechanism for unlicensed bands. Our objective is to maximize the average throughput of vehicles while guaranteeing the WiFi users' throughput. In simulations based on deep reinforcement learning (DRL) algorithms such as deep deterministic policy gradient (DDPG) and deep Q network (DQN), our proposed framework outperforms existing solutions that rely on fixed DC or schemes without consideration of unlicensed bands.

Keywords: vehicle-to-everything (V2X), resource allocation, BS assignment, new radio (NR), new radio unlicensed (NR-U), coexistence NR-U and WiFi, deep deterministic policy gradient (DDPG), deep Q-network (DQN), joint BS assignment and resource allocation (JBSRA), duty cycle mechanism

Procedia PDF Downloads 103

Authors: Shibo Wei, Ting Jiang

Abstract:

Convolutional-recurrent neural networks (CRN) have achieved much success recently in the speech enhancement field. The common processing method is to use the convolution layer to compress the feature space by multiple upsampling and then model the compressed features with the LSTM layer. At last, the enhanced speech is obtained by deconvolution operation to integrate the global information of the speech sequence. However, the feature space compression process may cause the loss of information, so we propose to model the upsampling result of each step with the residual LSTM layer, then join it with the output of the deconvolution layer and input them to the next deconvolution layer, by this way, we want to integrate the global information of speech sequence better. The experimental results show the network model (RES-CRN) we introduce can achieve better performance than LSTM without residual and overlaying LSTM simply in the original CRN in terms of scale-invariant signal-to-distortion ratio (SI-SNR), speech quality (PESQ), and intelligibility (STOI).

Keywords: convolutional-recurrent neural networks, speech enhancement, residual LSTM, SI-SNR

Procedia PDF Downloads 200

Authors: Imen Elleuch, Wael Ouarda, Gargouri Bilel

Abstract:

We proposes in this paper a Deep Learning based approach to classify text in order to enrich an Arabic ontology based on the objects classes of Gaston Gross. Those object classes are defined by taking into account the syntactic and semantic features of the treated language. Thus, our proposed approach is a hybrid one. In fact, it is based on the one hand on the object classes that represents a knowledge based-approach on classification of text and in the other hand it uses the deep learning approach that use the word embedding-based-approach to classify text. We have applied our proposed approach on a corpus constructed from an Arabic dictionary. The obtained semantic classification of text will enrich the Arabic objects classes ontology. In fact, new classes can be added to the ontology or an expansion of the features that characterizes each object class can be updated. The obtained results are compared to a similar work that treats the same object with a classical linguistic approach for the semantic classification of text. This comparison highlight our hybrid proposed approach that can be ameliorated by broaden the dataset used in the deep learning process.

Keywords: deep-learning approach, object-classes, semantic classification, Arabic

Procedia PDF Downloads 88