Search results for: A complexity-based approach in image compression using neural networks

Authors: Norbik Bashah, Idris Bharanidharan Shanmugam, Abdul Manan Ahmed

Abstract:

Intrusion Detection Systems are increasingly a key part of systems defense. Various approaches to Intrusion Detection are currently being used, but they are relatively ineffective. Artificial Intelligence plays a driving role in security services. This paper proposes a dynamic model Intelligent Intrusion Detection System, based on specific AI approach for intrusion detection. The techniques that are being investigated includes neural networks and fuzzy logic with network profiling, that uses simple data mining techniques to process the network data. The proposed system is a hybrid system that combines anomaly, misuse and host based detection. Simple Fuzzy rules allow us to construct if-then rules that reflect common ways of describing security attacks. For host based intrusion detection we use neural-networks along with self organizing maps. Suspicious intrusions can be traced back to its original source path and any traffic from that particular source will be redirected back to them in future. Both network traffic and system audit data are used as inputs for both.

Keywords: Intrusion Detection, Network Security, Data mining, Fuzzy Logic.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2139

Authors: Mbainaibeye Jérôme, Noureddine Ellouze

Abstract:

We study in this paper the effect of the scene changing on image sequences coding system using Embedded Zerotree Wavelet (EZW). The scene changing considered here is the full motion which may occurs. A special image sequence is generated where the scene changing occurs randomly. Two scenarios are considered: In the first scenario, the system must provide the reconstruction quality as best as possible by the management of the bit rate (BR) while the scene changing occurs. In the second scenario, the system must keep the bit rate as constant as possible by the management of the reconstruction quality. The first scenario may be motivated by the availability of a large band pass transmission channel where an increase of the bit rate may be possible to keep the reconstruction quality up to a given threshold. The second scenario may be concerned by the narrow band pass transmission channel where an increase of the bit rate is not possible. In this last case, applications for which the reconstruction quality is not a constraint may be considered. The simulations are performed with five scales wavelet decomposition using the 9/7-tap filter bank biorthogonal wavelet. The entropy coding is performed using a specific defined binary code book and EZW algorithm. Experimental results are presented and compared to LEAD H263 EVAL. It is shown that if the reconstruction quality is the constraint, the system increases the bit rate to obtain the required quality. In the case where the bit rate must be constant, the system is unable to provide the required quality if the scene change occurs; however, the system is able to improve the quality while the scene changing disappears.

Keywords: Image Sequence Compression, Wavelet Transform, Scene Changing, Zero Tree, Bit Rate, Quality.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1360

Authors: Alexandre Boum, Salomon Madinatou

Abstract:

This paper aims at bringing a scientific contribution to the cardiac arrhythmia biomedical diagnosis systems; more precisely to the study of the amelioration of cardiac arrhythmia classification performance using artificial neural network, adaptive neuro-fuzzy and fuzzy inference systems classifiers. The purpose of this amelioration is to enable cardiologists to make reliable diagnosis through automatic cardiac arrhythmia analyzes and classifications based on high confidence classifiers. In this study, six classes of the most commonly encountered arrhythmias are considered: the Right Bundle Branch Block, the Left Bundle Branch Block, the Ventricular Extrasystole, the Auricular Extrasystole, the Atrial Fibrillation and the Normal Cardiac rate beat. From the electrocardiogram (ECG) extracted parameters, we constructed a matrix (360x360) serving as an input data sample for the classifiers based on neural networks and a matrix (1x6) for the classifier based on fuzzy logic. By varying three parameters (the quality of the neural network learning, the data size and the quality of the input parameters) the automatic classification permitted us to obtain the following performances: in terms of correct classification rate, 83.6% was obtained using the fuzzy logic based classifier, 99.7% using the neural network based classifier and 99.8% for the adaptive neuro-fuzzy based classifier. These results are based on signals containing at least 360 cardiac cycles. Based on the comparative analysis of the aforementioned three arrhythmia classifiers, the classifiers based on neural networks exhibit a better performance.

Keywords: Adaptive neuro-fuzzy, artificial neural network, cardiac arrythmias, fuzzy inference systems.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 720

Authors: H.Mohammadi Majd, M.Jalali Azizpour

Abstract:

In this paper back-propagation artificial neural network (BPANN )with Levenberg–Marquardt algorithm is employed to predict the deformation of the upsetting process. To prepare a training set for BPANN, some finite element simulations were carried out. The input data for the artificial neural network are a set of parameters generated randomly (aspect ratio d/h, material properties, temperature and coefficient of friction). The output data are the coefficient of polynomial that fitted on barreling curves. Neural network was trained using barreling curves generated by finite element simulations of the upsetting and the corresponding material parameters. This technique was tested for three different specimens and can be successfully employed to predict the deformation of the upsetting process

Keywords: Back-propagation artificial neural network(BPANN), prediction, upsetting

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1797

Authors: Vipin Pal, Girdhari Singh, R P Yadav

Abstract:

As wireless sensor networks are energy constraint networks so energy efficiency of sensor nodes is the main design issue. Clustering of nodes is an energy efficient approach. It prolongs the lifetime of wireless sensor networks by avoiding long distance communication. Clustering algorithms operate in rounds. Performance of clustering algorithm depends upon the round time. A large round time consumes more energy of cluster heads while a small round time causes frequent re-clustering. So existing clustering algorithms apply a trade off to round time and calculate it from the initial parameters of networks. But it is not appropriate to use initial parameters based round time value throughout the network lifetime because wireless sensor networks are dynamic in nature (nodes can be added to the network or some nodes go out of energy). In this paper a variable round time approach is proposed that calculates round time depending upon the number of active nodes remaining in the field. The proposed approach makes the clustering algorithm adaptive to network dynamics. For simulation the approach is implemented with LEACH in NS-2 and the results show that there is 6% increase in network lifetime, 7% increase in 50% node death time and 5% improvement over the data units gathered at the base station.

Keywords: Wireless Sensor Network, Clustering, Energy Efficiency, Round Time.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1793

Authors: El Asnaoui Khalid, Aksasse Brahim, Ouanan Mohammed

Abstract:

This paper presents an unsupervised color image segmentation method. It is based on a hierarchical analysis of 2-D histogram in RGB color space. This histogram minimizes storage space of images and thus facilitates the operations between them. The improved segmentation approach shows a better identification of objects in a color image and, at the same time, the system is fast.

Keywords: Image segmentation, hierarchical analysis, 2-D histogram, Classification.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1633

Authors: K.Mala, V.Sadasivam, S.Alagappan

Abstract:

Advances in clinical medical imaging have brought about the routine production of vast numbers of medical images that need to be analyzed. As a result an enormous amount of computer vision research effort has been targeted at achieving automated medical image analysis. Computed Tomography (CT) is highly accurate for diagnosing liver tumors. This study aimed to evaluate the potential role of the wavelet and the neural network in the differential diagnosis of liver tumors in CT images. The tumors considered in this study are hepatocellular carcinoma, cholangio carcinoma, hemangeoma and hepatoadenoma. Each suspicious tumor region was automatically extracted from the CT abdominal images and the textural information obtained was used to train the Probabilistic Neural Network (PNN) to classify the tumors. Results obtained were evaluated with the help of radiologists. The system differentiates the tumor with relatively high accuracy and is therefore clinically useful.

Keywords: Fuzzy c means clustering, texture analysis, probabilistic neural network, LVQ neural network.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2994

Authors: Balimu Mwiya, Mundia Muya, Chabota Kaliba, Peter Mukalula

Abstract:

Factors affecting construction unit cost vary depending on a country’s political, economic, social and technological inclinations. Factors affecting construction costs have been studied from various perspectives. Analysis of cost factors requires an appreciation of a country’s practices. Identified cost factors provide an indication of a country’s construction economic strata. The purpose of this paper is to identify the essential factors that affect unit cost estimation and their breakdown using artificial neural networks. Twenty five (25) identified cost factors in road construction were subjected to a questionnaire survey and employing SPSS factor analysis the factors were reduced to eight. The 8 factors were analysed using neural network (NN) to determine the proportionate breakdown of the cost factors in a given construction unit rate. NN predicted that political environment accounted 44% of the unit rate followed by contractor capacity at 22% and financial delays, project feasibility and overhead & profit each at 11%. Project location, material availability and corruption perception index had minimal impact on the unit cost from the training data provided. Quantified cost factors can be incorporated in unit cost estimation models (UCEM) to produce more accurate estimates. This can create improvements in the cost estimation of infrastructure projects and establish a benchmark standard to assist the process of alignment of work practises and training of new staff, permitting the on-going development of best practises in cost estimation to become more effective.

Keywords: Construction cost factors, neural networks, roadworks, Zambian Construction Industry.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3834

Authors: A. Mitra, Y. V. Subba Rao, S. R. M. Prasanna

Abstract:

This paper proposes a new approach for image encryption using a combination of different permutation techniques. The main idea behind the present work is that an image can be viewed as an arrangement of bits, pixels and blocks. The intelligible information present in an image is due to the correlations among the bits, pixels and blocks in a given arrangement. This perceivable information can be reduced by decreasing the correlation among the bits, pixels and blocks using certain permutation techniques. This paper presents an approach for a random combination of the aforementioned permutations for image encryption. From the results, it is observed that the permutation of bits is effective in significantly reducing the correlation thereby decreasing the perceptual information, whereas the permutation of pixels and blocks are good at producing higher level security compared to bit permutation. A random combination method employing all the three techniques thus is observed to be useful for tactical security applications, where protection is needed only against a casual observer.

Keywords: Encryption, Permutation, Good key, Combinationalpermutation, Pseudo random index generator.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2240

Authors: M.R. Mustafa, M.H. Isa, R.B. Rezaur

Abstract:

Prediction of highly non linear behavior of suspended sediment flow in rivers has prime importance in the field of water resources engineering. In this study the predictive performance of two Artificial Neural Networks (ANNs) namely, the Radial Basis Function (RBF) Network and the Multi Layer Feed Forward (MLFF) Network have been compared. Time series data of daily suspended sediment discharge and water discharge at Pari River was used for training and testing the networks. A number of statistical parameters i.e. root mean square error (RMSE), mean absolute error (MAE), coefficient of efficiency (CE) and coefficient of determination (R2) were used for performance evaluation of the models. Both the models produced satisfactory results and showed a good agreement between the predicted and observed data. The RBF network model provided slightly better results than the MLFF network model in predicting suspended sediment discharge.

Keywords: ANN, discharge, modeling, prediction, suspendedsediment,

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1732

Authors: Wajdi Bellil, Chokri Ben Amar, Adel M. Alimi

Abstract:

An application of Beta wavelet networks to synthesize pass-high and pass-low wavelet filters is investigated in this work. A Beta wavelet network is constructed using a parametric function called Beta function in order to resolve some nonlinear approximation problem. We combine the filter design theory with wavelet network approximation to synthesize perfect filter reconstruction. The order filter is given by the number of neurons in the hidden layer of the neural network. In this paper we use only the first derivative of Beta function to illustrate the proposed design procedures and exhibit its performance.

Keywords: Beta wavelets, Wavenet, multiresolution analysis, perfect filter reconstruction, salient point detect, repeatability.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1669

Authors: Jalal Karam

Abstract:

Years of extensive research in the field of speech processing for compression and recognition in the last five decades, resulted in a severe competition among the various methods and paradigms introduced. In this paper we include the different representations of speech in the time-frequency and time-scale domains for the purpose of compression and recognition. The examination of these representations in a variety of related work is accomplished. In particular, we emphasize methods related to Fourier analysis paradigms and wavelet based ones along with the advantages and disadvantages of both approaches.

Keywords: Time-Scale, Wavelets, Time-Frequency, Compression, Recognition.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2337

Authors: Christian Mayr, Holger Eisenreich, Stephan Henker, René Schüffny

Abstract:

Image convolution similar to the receptive fields found in mammalian visual pathways has long been used in conventional image processing in the form of Gabor masks. However, no VLSI implementation of parallel, multi-layered pulsed processing has been brought forward which would emulate this property. We present a technical realization of such a pulsed image processing scheme. The discussed IC also serves as a general testbed for VLSI-based pulsed information processing, which is of interest especially with regard to the robustness of representing an analog signal in the phase or duration of a pulsed, quasi-digital signal, as well as the possibility of direct digital manipulation of such an analog signal. The network connectivity and processing properties are reconfigurable so as to allow adaptation to various processing tasks.

Keywords: Neural image processing, pulse computation application, pulsed Gabor convolution, VLSI pulse routing.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1401

Authors: Brian A. Smith, Ronald W. McClendon, Gerrit Hoogenboom

Abstract:

The mitigation of crop loss due to damaging freezes requires accurate air temperature prediction models. An improved model for temperature prediction in Georgia was developed by including information on seasonality and modifying parameters of an existing artificial neural network model. Alternative models were compared by instantiating and training multiple networks for each model. The inclusion of up to 24 hours of prior weather information and inputs reflecting the day of year were among improvements that reduced average four-hour prediction error by 0.18°C compared to the prior model. Results strongly suggest model developers should instantiate and train multiple networks with different initial weights to establish appropriate model parameters.

Keywords: Time-series forecasting, weather modeling.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1873

Authors: Hazem M. El-Bakry

Abstract:

In this paper, a new technique for fast painting with different colors is presented. The idea of painting relies on applying masks with different colors to the background. Fast painting is achieved by applying these masks in the frequency domain instead of spatial (time) domain. New colors can be generated automatically as a result from the cross correlation operation. This idea was applied successfully for faster specific data (face, object, pattern, and code) detection using neural algorithms. Here, instead of performing cross correlation between the input input data (e.g., image, or a stream of sequential data) and the weights of neural networks, the cross correlation is performed between the colored masks and the background. Furthermore, this approach is developed to reduce the computation steps required by the painting operation. The principle of divide and conquer strategy is applied through background decomposition. Each background is divided into small in size subbackgrounds and then each sub-background is processed separately by using a single faster painting algorithm. Moreover, the fastest painting is achieved by using parallel processing techniques to paint the resulting sub-backgrounds using the same number of faster painting algorithms. In contrast to using only faster painting algorithm, the speed up ratio is increased with the size of the background when using faster painting algorithm and background decomposition. Simulation results show that painting in the frequency domain is faster than that in the spatial domain.

Keywords: Fast Painting, Cross Correlation, Frequency Domain, Parallel Processing

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1803

Authors: Mohammad Heidari

Abstract:

Horizontal wells are proven to be better producers because they can be extended for a long distance in the pay zone. Engineers have the technical means to forecast the well productivity for a given horizontal length. However, experiences have shown that the actual production rate is often significantly less than that of forecasted. It is a difficult task, if not impossible to identify the real reason why a horizontal well is not producing what was forecasted. Often the source of problem lies in the drilling of horizontal section such as permeability reduction in the pay zone due to mud invasion or snaky well patterns created during drilling. Although drillers aim to drill a constant inclination hole in the pay zone, the more frequent outcome is a sinusoidal wellbore trajectory. The two factors, which play an important role in wellbore tortuosity, are the inclination and side force at bit. A constant inclination horizontal well can only be drilled if the bit face is maintained perpendicular to longitudinal axis of bottom hole assembly (BHA) while keeping the side force nil at the bit. This approach assumes that there exists no formation force at bit. Hence, an appropriate BHA can be designed if bit side force and bit tilt are determined accurately. The Artificial Neural Network (ANN) is superior to existing analytical techniques. In this study, the neural networks have been employed as a general approximation tool for estimation of the bit side forces. A number of samples are analyzed with ANN for parameters of bit side force and the results are compared with exact analysis. Back Propagation Neural network (BPN) is used to approximation of bit side forces. Resultant low relative error value of the test indicates the usability of the BPN in this area.

Keywords: Artificial Neural Network, BHA, Horizontal Well, Stabilizer.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1985

Authors: Abiodun M. Aibinu, Momoh J. E. Salami, Amir A. Shafie, Athaur Rahman Najeeb

Abstract:

Autoregressive Moving average (ARMA) is a parametric based method of signal representation. It is suitable for problems in which the signal can be modeled by explicit known source functions with a few adjustable parameters. Various methods have been suggested for the coefficients determination among which are Prony, Pade, Autocorrelation, Covariance and most recently, the use of Artificial Neural Network technique. In this paper, the method of using Artificial Neural network (ANN) technique is compared with some known and widely acceptable techniques. The comparisons is entirely based on the value of the coefficients obtained. Result obtained shows that the use of ANN also gives accurate in computing the coefficients of an ARMA system.

Keywords: Autoregressive moving average, coefficients, back propagation, model parameters, neural network, weight.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2298

Authors: S. Zappacosta, C. Bove, M. Carmela Marinelli, P. di Lauro, K. Spasenovic, L. Ostano, G. Aiello, M. Pietrosanto

Abstract:

This paper proposes a neural network approach for assessing flood hazard mapping. The core of the model is a machine learning component fed by frequency ratios, namely statistical correlations between flood event occurrences and a selected number of topographic properties. The classification capability was compared with the flood hazard mapping River Basin Plans (Piani Assetto Idrogeologico, acronimed as PAI) designed by the Italian Institute for Environmental Research and Defence, ISPRA (Istituto Superiore per la Protezione e la Ricerca Ambientale), encoding four different increasing flood hazard levels. The study area of Piemonte, an Italian region, has been considered without loss of generality. The frequency ratios may be used as a standalone block to model the flood hazard mapping. Nevertheless, the mixture with a neural network improves the classification power of several percentage points, and may be proposed as a basic tool to model the flood hazard map in a wider scope.

Keywords: flood modeling, hazard map, neural networks, hydrogeological risk, flood risk assessment

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 747

Authors: Rana Yousef, Khalil el Hindi

Abstract:

The behavior of Radial Basis Function (RBF) Networks greatly depends on how the center points of the basis functions are selected. In this work we investigate the use of instance reduction techniques, originally developed to reduce the storage requirements of instance based learners, for this purpose. Five Instance-Based Reduction Techniques were used to determine the set of center points, and RBF networks were trained using these sets of centers. The performance of the RBF networks is studied in terms of classification accuracy and training time. The results obtained were compared with two Radial Basis Function Networks: RBF networks that use all instances of the training set as center points (RBF-ALL) and Probabilistic Neural Networks (PNN). The former achieves high classification accuracies and the latter requires smaller training time. Results showed that RBF networks trained using sets of centers located by noise-filtering techniques (ALLKNN and ENN) rather than pure reduction techniques produce the best results in terms of classification accuracy. The results show that these networks require smaller training time than that of RBF-ALL and higher classification accuracy than that of PNN. Thus, using ALLKNN and ENN to select center points gives better combination of classification accuracy and training time. Our experiments also show that using the reduced sets to train the networks is beneficial especially in the presence of noise in the original training sets.

Keywords: Radial basis function networks, Instance-based reduction, PNN.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1694

Authors: Fatma El-zahraa El-taher, Ayman Taha, Jane Courtney, Susan Mckeever

Abstract:

Understanding road networks plays an important role in navigation applications such as self-driving vehicles and route planning for individual journeys. Intersections of roads are essential components of road networks. Understanding the features of an intersection, from a simple T-junction to larger multi-road junctions is critical to decisions such as crossing roads or selecting safest routes. The identification and profiling of intersections from satellite images is a challenging task. While deep learning approaches offer state-of-the-art in image classification and detection, the availability of training datasets is a bottleneck in this approach. In this paper, a labelled satellite image dataset for the intersection recognition  problem is presented. It consists of 14,692 satellite images of Washington DC, USA. To support other users of the dataset, an automated download and labelling script is provided for dataset replication. The challenges of construction and fine-grained feature labelling of a satellite image dataset are examined, including the issue of how to address features that are spread across multiple images. Finally, the accuracy of detection of intersections in satellite images is evaluated.

Keywords: Satellite images, remote sensing images, data acquisition, autonomous vehicles, robot navigation, route planning, road intersections.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 791

Authors: Fatemeh Behjati Ardakani, Fatemeh Khademian, Abbas Nowzari Dalini, Reza Ebrahimpour

Abstract:

Human activity is a major concern in a wide variety of applications, such as video surveillance, human computer interface and face image database management. Detecting and recognizing faces is a crucial step in these applications. Furthermore, major advancements and initiatives in security applications in the past years have propelled face recognition technology into the spotlight. The performance of existing face recognition systems declines significantly if the resolution of the face image falls below a certain level. This is especially critical in surveillance imagery where often, due to many reasons, only low-resolution video of faces is available. If these low-resolution images are passed to a face recognition system, the performance is usually unacceptable. Hence, resolution plays a key role in face recognition systems. In this paper we introduce a new low resolution face recognition system based on mixture of expert neural networks. In order to produce the low resolution input images we down-sampled the 48 × 48 ORL images to 12 × 12 ones using the nearest neighbor interpolation method and after that applying the bicubic interpolation method yields enhanced images which is given to the Principal Component Analysis feature extractor system. Comparison with some of the most related methods indicates that the proposed novel model yields excellent recognition rate in low resolution face recognition that is the recognition rate of 100% for the training set and 96.5% for the test set.

Keywords: Low resolution face recognition, Multilayered neuralnetwork, Mixture of experts neural network, Principal componentanalysis, Bicubic interpolation, Nearest neighbor interpolation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1732

Authors: Hongyuan Li, Guang Liu, Yuewei Dai, Zhiquan Wang

Abstract:

Digital watermarking is the process of embedding information into a digital signal which can be used in DRM (digital rights managements) system. The visible watermark (often called logo) can indicate the owner of the copyright which can often be seen in the TV program and protects the copyright in an active way. However, most of the schemes do not consider the visible watermark removing process. To solve this problem, a visible watermarking scheme with embedding and removing process is proposed under the control of a secure template. The template generates different version of watermarks which can be seen visually the same for different users. Users with the right key can completely remove the watermark and recover the original image while the unauthorized user is prevented to remove the watermark. Experiment results show that our watermarking algorithm obtains a good visual quality and is hard to be removed by the illegally users. Additionally, the authorized users can completely remove the visible watermark and recover the original image with a good quality.

Keywords: digital watermarking, visible and removablewatermark, secure template, JPEG compression

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1541

Authors: Shuen-Tai Wang, Fang-An Kuo, Chau-Yi Chou, Yu-Bin Fang

Abstract:

2016 has become the year of the Artificial Intelligence explosion. AI technologies are getting more and more matured that most world well-known tech giants are making large investment to increase the capabilities in AI. Machine learning is the science of getting computers to act without being explicitly programmed, and deep learning is a subset of machine learning that uses deep neural network to train a machine to learn  features directly from data. Deep learning realizes many machine learning applications which expand the field of AI. At the present time, deep learning frameworks have been widely deployed on servers for deep learning applications in both academia and industry. In training deep neural networks, there are many standard processes or algorithms, but the performance of different frameworks might be different. In this paper we evaluate the running performance of two state-of-the-art distributed deep learning frameworks that are running training calculation in parallel over multi GPU and multi nodes in our cloud environment. We evaluate the training performance of the frameworks with ResNet-50 convolutional neural network, and we analyze what factors that result in the performance among both distributed frameworks as well. Through the experimental analysis, we identify the overheads which could be further optimized. The main contribution is that the evaluation results provide further optimization directions in both performance tuning and algorithmic design.

Keywords: Artificial Intelligence, machine learning, deep learning, convolutional neural networks.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1269

Authors: Z. Zainuddin, N. Mahat, Y. Abu Hassan

Abstract:

Since the presentation of the backpropagation algorithm, a vast variety of improvements of the technique for training a feed forward neural networks have been proposed. This article focuses on two classes of acceleration techniques, one is known as Local Adaptive Techniques that are based on weightspecific only, such as the temporal behavior of the partial derivative of the current weight. The other, known as Dynamic Adaptation Methods, which dynamically adapts the momentum factors, α, and learning rate, η, with respect to the iteration number or gradient. Some of most popular learning algorithms are described. These techniques have been implemented and tested on several problems and measured in terms of gradient and error function evaluation, and percentage of success. Numerical evidence shows that these techniques improve the convergence of the Backpropagation algorithm.

Keywords: Backpropagation, Dynamic Adaptation Methods, Local Adaptive Techniques, Neural networks.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2179

Authors: M. Nahid, A. Bajit, A. Tamtaoui, E. H. Bouyakhf

Abstract:

To model the human visual system (HVS) in the region of interest, we propose a new objective metric evaluation adapted to wavelet foveation-based image compression quality measurement, which exploits a foveation setup filter implementation technique in the DWT domain, based especially on the point and region of fixation of the human eye. This model is then used to predict the visible divergences between an original and compressed image with respect to this region field and yields an adapted and local measure error by removing all peripheral errors. The technique, which we call foveation wavelet visible difference prediction (FWVDP), is demonstrated on a number of noisy images all of which have the same local peak signal to noise ratio (PSNR), but visibly different errors. We show that the FWVDP reliably predicts the fixation areas of interest where error is masked, due to high image contrast, and the areas where the error is visible, due to low image contrast. The paper also suggests ways in which the FWVDP can be used to determine a visually optimal quantization strategy for foveation-based wavelet coefficients and to produce a quantitative local measure of image quality.

Keywords: Human Visual System, Image Quality, ImageCompression, foveation wavelet, region of interest ROI.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1504

Authors: Kyoung Hoon Kim

Abstract:

Turbine blade cooling is considered as the most effective way of maintaining high operating temperature making use of the available materials, and turbine systems with wet compression have a potential for future power generation because of high efficiency and high specific power with a relatively low cost. In this paper performance analysis of wet-compression gas turbine cycle with turbine blade cooling is carried out. The wet compression process is analytically modeled based on non-equilibrium droplet evaporation. Special attention is paid for the effects of pressure ratio and water injection ratio on the important system variables such as ratio of coolant fluid flow, fuel consumption, thermal efficiency and specific power. Parametric studies show that wet compression leads to insignificant improvement in thermal efficiency but significant enhancement of specific power in gas turbine systems with turbine blade cooling.

Keywords: Water injection, wet compression, gas turbine, turbine blade cooling.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3413

Authors: D. Singh, R. Yousefi, M. Boroushaki

Abstract:

Intelligent deep-drawing is an instrumental research field in sheet metal forming. A set of 28 different experimental data have been employed in this paper, investigating the roles of die radius, punch radius, friction coefficients and drawing ratios for axisymmetric workpieces deep drawing. This paper focuses an evolutionary neural network, specifically, error back propagation in collaboration with genetic algorithm. The neural network encompasses a number of different functional nodes defined through the established principles. The input parameters, i.e., punch radii, die radii, friction coefficients and drawing ratios are set to the network; thereafter, the material outputs at two critical points are accurately calculated. The output of the network is used to establish the best parameters leading to the most uniform thickness in the product via the genetic algorithm. This research achieved satisfactory results based on demonstration of neural networks.

Keywords: Deep-drawing, Neural network, Genetic algorithm, Sheet metal forming.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2211

Authors: El Mahdi Barrah, Said Safi, Abdessamad Malaoui

Abstract:

In this paper, we present an algorithm for reconstruction from incomplete 2D scans for tifinagh characters. This algorithm is based on using correlation between the lost block and its neighbors. This system proposed contains three main parts: pre-processing, features extraction and recognition. In the first step, we construct a database of tifinagh characters. In the second step, we will apply “shape analysis algorithm”. In classification part, we will use Neural Network. The simulation results demonstrate that the proposed method give good results.

Keywords: Tifinagh character recognition, Neural networks, Local cost computation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1291

Authors: Mohamed Abdallah, Mostafa Warith, Roberto Narbaitz, Emil Petriu, Kevin Kennedy

Abstract:

Heterogeneity of solid waste characteristics as well as the complex processes taking place within the landfill ecosystem motivated the implementation of soft computing methodologies such as artificial neural networks (ANN), fuzzy logic (FL), and their combination. The present work uses a hybrid ANN-FL model that employs knowledge-based FL to describe the process qualitatively and implements the learning algorithm of ANN to optimize model parameters. The model was developed to simulate and predict the landfill gas production at a given time based on operational parameters. The experimental data used were compiled from lab-scale experiment that involved various operating scenarios. The developed model was validated and statistically analyzed using F-test, linear regression between actual and predicted data, and mean squared error measures. Overall, the simulated landfill gas production rates demonstrated reasonable agreement with actual data. The discussion focused on the effect of the size of training datasets and number of training epochs.

Keywords: Adaptive neural fuzzy inference system (ANFIS), gas production, landfill

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2421

Authors: A. Yazdanmehr, H. Jahed

Abstract:

Being made with the lightest commercially available industrial metal, Magnesium (Mg) alloys are of interest for light-weighting. Expanding their application to different material processing methods requires Mg properties at large strains. Several room-temperature processes such as shot and laser peening and hole cold expansion need compressive large strain data. Two methods have been proposed in the literature to obtain the stress-strain curve at high strains: 1) anti-buckling guides and 2) small cubic samples. In this paper, an anti-buckling fixture is used with the help of digital image correlation (DIC) to obtain the compression-tension (C-T) of AZ31B-H24 rolled sheet at large strain values of up to 10.5%. The effect of the anti-bucking fixture on stress-strain curves is evaluated experimentally by comparing the results with those of the compression tests of cubic samples. For testing cubic samples, a new fixture has been designed to increase the accuracy of testing cubic samples with DIC strain measurements. Results show a negligible effect of anti-buckling on stress-strain curves, specifically at high strain values.

Keywords: Large strain, compression-tension, loading-unloading, Mg alloys.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 790