scholarly journals Effect of Learning Rate on the Recognition of Images

1996 ◽  
Vol 19 (1) ◽  
pp. 1-12 ◽  
Author(s):  
M. Hamed ◽  
A. El Desouky
Keyword(s):  
Neural Network ◽  
Texture Classification ◽  
Back Propagation ◽  
Learning Rate ◽  
Computational Time ◽  
Rate Parameter ◽  
Modern Computer ◽  
The Neural Network ◽  
Hidden Layer ◽  
Propagation Technique

This paper presents a study for the effect of learning rate on an approach for texture classification and detection based on the neural network principle. This neural network consists of three layers, which are input, output, and hidden layers. The back propagation technique is considered. A computer algorithm is deduced and applied. In this work, the synthetic textures are generated. The results are taken for the modern computer of AT 486 type. The mathematical analysis is summarized in order to illustrate the effect of learning rate parameter on the exact discrimination during processing. This effect is studied through applications. The minimum consumed time for the computational time of classification in industry is correlated to correspond only the use of only 2 units in the hidden layer of a neural network for real images instead of 11 units.

scholarly journals Quranic Sign Language for Deaf People: Quranic Recitation Classification and Verification

2018 ◽  
Vol 4 (1) ◽  
pp. 7-11
Author(s):  
Mohamed Ali Mahmod ◽  
Akram Zeki
Keyword(s):  
Neural Network ◽  
Sign Language ◽  
Back Propagation ◽  
Deaf People ◽  
The Neural Network ◽  
Holy Quran ◽  
Video Recognition ◽  
The Holy Quran ◽  
Propagation Technique

This paper provides an overview of the techniques used in image and video recognition for sign language through following hand motions and translating it to the text of the Holy Quran. It also provides a proposal for a system that will be capable of identifying errors in Quran recitation depending on alphabets of Arabic and Quranic sign language and be able to show where exactly errors have occurred. In addition, this system will identify and classify location of verse (Ayah) and names of Souras depending on Back Propagation technique of the neural network.

scholarly journals Development of design system for crack patterns on cup surface based on KANSEI

2018 ◽  
Vol 5 (4) ◽  
pp. 435-441 ◽  
Author(s):  
Mai Misaka ◽  
Hideki Aoyama
Keyword(s):  
Neural Network ◽  
Computer Aided Design ◽  
Back Propagation ◽  
Transition Function ◽  
Design System ◽  
Sigmoid Function ◽  
Crack Patterns ◽  
The Neural Network ◽  
Hidden Layer ◽  
Aided Design

Abstract With the development of manufacturing technology in recent years, as well as with the industrial product development, differentiation in the design aspect is becoming effective, not in terms of performance or quality. In addition, as consumers seek products that match their own sensibilities (KANSEI), designers are required to propose designs that highly conform to concept presented by client, while understanding the KANSEI of diversified consumers; therefore, their burden is increasing. To address these issues, the support of the development of computer-aided design has advanced; however, it is difficult to reflect human KANSEI or to generate a design that induces a natural impression through computers. The purpose of this research is to develop a system that incorporates the KANSEI of users, and emits a pattern design that induces a natural impression using a computer. This work is focused on crack patterns that can be observed on pottery surfaces, and a method for generating crack patterns on a cup surface is suggested. In this study, a Bézier curved surface and fluctuation were employed in order to induce a natural impression. In addition, by using the neural network, the crack patterns were associated with user KANSEI. The neural network was composed of three layers, namely the input layer, the hidden layer, and the output layer; it adopted the sigmoid function as the transition function and the back propagation as the learning method. As a result, a system was constructed, in which a crack pattern that satisfied the input produces an output according to the desired impression of the user. Finally, an evaluation questionnaire was distributed, and the usefulness of the system was confirmed. Highlights A method for creating crack patterns using a computer is proposed. The relationship between KANSEI and crack patterns is modeled by neural network. A system is developed that outputs crack patterns that satisfy the inputted KANSEI.

Flow Stress Prediction of High-Nb TiAl Alloys under High Temperature Deformation

2012 ◽  
Vol 510 ◽  
pp. 723-728 ◽  
Author(s):  
Liang Cheng ◽  
Hui Chang ◽  
Bin Tang ◽  
Hong Chao Kou ◽  
Jin Shan Li
Keyword(s):  
Neural Network ◽  
High Temperature ◽  
Flow Stress ◽  
Back Propagation ◽  
Three Dimensional ◽  
High Temperature Deformation ◽  
Temperature Deformation ◽  
Ann Model ◽  
The Neural Network ◽  
Hidden Layer

In this work, a back propagation artificial neural network (BP-ANN) model is conducted to predict the flow behaviors of high-Nb TiAl (TNB) alloys during high temperature deformation. The inputs of the neural network are deformation temperature, log strain rate and strain whereas flow stress is the output. There is a single hidden layer with 7 neutrons in the network, and the weights and bias of the network were optimized by Genetic Algorithm (GA). The comparison result suggests a very good correlation between experimental and predicted data. Besides, the non-experimental flow stress predicted by the network is shown to be in good agreement with the results calculated by three dimensional interpolation, which confirmed a good generalization capability of the proposed network.

scholarly journals PENDETEKSIAN DERAU CITRA SECARA OTOMATIS MENGGUNAKAN TEKNIK JARINGAN SYARAF TIRUAN

2009 ◽  
Vol 1 (2) ◽  
pp. 73-86
Author(s):  
Julsam Julsam
Keyword(s):  
Neural Network ◽  
Back Propagation ◽  
Convolution Operation ◽  
Output Layer ◽  
Output Neuron ◽  
Input Layer ◽  
Image Blurring ◽  
Hidden Layer ◽  
Propagation Technique ◽  
Blurring Effect

This research is application of neural network technique to optimize convolution operation using mask 3x3 to omit the image blurring effect. This neural network consists of three layers.  They are input layer (9 neuron inputs), output layer (1 output neuron) and hidden layer. Each layer is applied to 3, 5 and 7 neuron using back propagation technique. The result shows the using five neurons to hidden layer give the highest value of sound pixel recognizing (76.47%)

A Neural Network-Based Prediction of the Weld Lines in Resin Transfer Molded Parts

1999 ◽  
Author(s):  
Faezeh Soltani ◽  
Souran Manoochehri
Keyword(s):  
Neural Network ◽  
Back Propagation ◽  
Simulation Models ◽  
Optimization Methods ◽  
Computational Time ◽  
Back Propagation Algorithm ◽  
Rtm Process ◽  
The Neural Network ◽  
Molded Parts ◽  
The Neural Networks

Abstract A model is developed to predict the weld lines in Resin Transfer Molding (RTM) process. In this model, the preforms are assumed to be thin flat with isotropic and orthotropic permeabilities. The position of the weld lines formed by multiple specified inlet ports are predicted using a neural network-based back propagation algorithm. The neural network was trained with data obtained from simulation and actual molding experimentation. Part geometry is decomposed into smaller sections based on the position of the weld lines. The variety of preforms and processing conditions are used to verify the model. Applying the neural networks reduced the amount of computational time by several orders of magnitude compared with simulations. The models developed in this study can be effectively utilized in iterative optimization methods where use of numerical simulation models is cumbersome.

Signature Verification using Normalized Static Features and Neural Network Classification

2016 ◽  
Vol 6 (6) ◽  
pp. 2665
Author(s):  
Manish Trikha ◽  
Manas Singhal ◽  
Maitreyee Dutta
Keyword(s):  
Neural Network ◽  
Back Propagation ◽  
Rejection Rate ◽  
Signature Verification ◽  
Online Systems ◽  
Biometric Verification ◽  
Online Mode ◽  
Verification System ◽  
The Neural Network ◽  
Propagation Technique

<p>Signature verification is very widely used in verification of the identity of any person. Now a days other biometric verification system has been evolved very widely like figure print, iris etc., but signature verification through computer system is still in development phase. The verification system is either through offline mode or online mode in online systems the dynamic information of a signature captured at the time the signature is made while in offline systems based on the scanned image of a signature. In this paper, a method is presented for Offline signatures Verification, for this verification system signature image is first pre-processed and converted into binary image of same size with 200x200 Pixels and then different features are extracted from the image like Eccentricity, Kurtosis, Skewness etc. and that features are used to train the neural network using back-propagation technique. For this verification system 6 different user signatures are taken to make database of the feature and results are analysed. The result demonstrate the efficiency of the proposed methodology when compared with other existing studies. The proposed algorithm gives False Acceptance Rate (FAR) as 5.05% and False Rejection rate (FRR) as 4.25%.</p>

The Effects of ZigBee Position with Different Hidden Layers in Back-Propagation Neural Network

2011 ◽  
Vol 189-193 ◽  
pp. 1761-1767
Author(s):  
Fang Tsung Liu ◽  
Ceweng Erh Weng ◽  
Chien Ming Huang ◽  
Chang Yan Yang ◽  
Huang Chu Huang
Keyword(s):  
Neural Network ◽  
Wireless Sensor Networks ◽  
Sensor Networks ◽  
Back Propagation ◽  
Signal Strength ◽  
Back Propagation Neural Network ◽  
Wireless Sensor ◽  
Measured Signal ◽  
The Neural Network ◽  
Hidden Layer

In this paper, the research topic is that the expert experience is established by the size of the measured signal strength of wireless sensor networks and put the strength of the actual collection of historical data into the neural network model. In order to get the minimize error we use the errors to modify the weights and threshold of the neural network links. We compare the differences of hidden layer neural network and the experimental results. We set up a wireless sensor networks environment to collect the measurement values of signal strength (RSSI) and develop an indoor positioning system.

Set Up of a Robust Neural Network for Gas Turbine Simulation

2004 ◽  
Author(s):  
R. Bettocchi ◽  
M. Pinelli ◽  
P. R. Spina ◽  
M. Venturini ◽  
M. Burgio
Keyword(s):  
Neural Network ◽  
Gas Turbine ◽  
Gas Turbines ◽  
Learning Algorithm ◽  
General Procedure ◽  
Back Propagation ◽  
Computational Time ◽  
Time Simulation ◽  
Hidden Layer ◽  
Set Up

In this paper, Neural Network (NN) models for the real-time simulation of gas turbines are studied and developed. The analyses carried out are aimed at the selection of the most appropriate NN structure for gas turbine simulation, in terms of both computational time of the NN training phase and accuracy and robustness with respect to measurement uncertainty. In particular, feed-forward NNs, with a single hidden layer and different numbers of neurons, trained by using a back-propagation learning algorithm are considered and tested. Finally, a general procedure for the validation of computational codes is adapted and applied to the validation of the developed NN models.

Sign in / Sign up

Export Citation Format

Share Document