检索结果-内蒙古大学图书馆

IEEE Region 10 Conference (TENCON)

作者： Chumuang, Naruemol Ketcham, Mahasak King Mongkuts Univ Technol North Bangkok Dept Informat Technol Fac Informat Technol Bangkok Thailand King Mongkuts Univ Technol North Bangkok Dept Informat Technol Management Fac Informat Technol Bangkok Thailand

ISBN: (纸本)9781479940752

This paper proposes an Intelligent Handwriting Thai Signature Recognition System base on Multilayer Perceptron and Radial Basis Network. The proposed system compose of three main processes, i.e. image pre-processing, feature extraction and Thai signature recognition. In the recognition processes the neural network is used into two stage. First, Multilayer Perceptron (MLP) and Radial Basis Function (RBF) is used to learning handwriting Thai signature and then the trained network will be used for recognizing. Later, RBF is used to decision in final stage. There are 600 images from 10 writers in this experiment then the experimental results show that the proposed method yielded the satisfied results.

关键词： component Signature Recognition Neural network Multilayer Perceptron back propagation algorithm Radial Basis Function

来源：评论

学校读者我要写书评

暂无评论

Dynamic Training Rate for backpropagation Learning algorithm

Dynamic Training Rate for Backpropagation Learning Algorithm

引用

IEEE 11th Malaysia International Conference on Communications (MICC)

作者： Al Duais, M. S. Yaakub, A. R. Yusoff, N. Univ Utara Malaysia Coll Arts & Sci Sch Comp Sintok 06010 Kedah Malaysia

ISBN: (纸本)9781479915323

In this paper, we created a dynamic function training rate for the back propagation learning algorithm to avoid the local minimum and to speed up training. The back propagation with dynamic training rate (BPDR) algorithm uses the sigmoid function. The 2-dimensional XOR problem and iris data were used as benchmarks to test the effects of the dynamic training rate formulated in this paper. The results of these experiments demonstrate that the BPDR algorithm is advantageous with regards to both generalization performance and training speed. The stop training or limited error was determined by 1.0(e-5)

关键词： Artificial neural networks back propagation algorithm adaptive training dynamic training rate

来源：评论

学校读者我要写书评

暂无评论

Random input based prediction and transfer of heat in soil temperature using artificial neural network

Random input based prediction and transfer of heat in soil t...

引用

International Conference on Advances in Materials Research (ICAMR)

作者： Jebamalar, S. Christopher, Jaslin J. Ajisha, M. Angelina Thanga Nesamony Mem Christian Coll Dept Phys Marthandam 629165 India MS Univ Constituent Coll Dept Chem Palkulam 629401 India Francis Xavier Engn Coll Dept Civil Engn Tirunelveli 627003 India

The soil temperature affects the climate in a great manner and the soil temperature data can be used to make predictions about how the ecosystem is affected by heat or cold in global temperature. Various measurement techniques have been employed in analyzing the soil temperature. Among them, Artificial Neural Network (ANN) is used to perform mathematical formulation to learn patterns and relationships in the data. The present study provides a tool to predict the soil temperature for the year 2004 using random inputs taken between 1993 and 2003 under different cases at two different soil depths 10 cm and 20 cm using ANN. The analysis was carried out for both the annual and the seasonal waves. Thereby, heat gets depleted during the monsoon seasons and gets accumulated in the annual and the other season which help to study the atmospheric changes. The statistical validation of the predictions possesses an insignificant error when compared with that of the observed soil temperature. (c) 2021 Elsevier Ltd. All rights reserved. Selection and peer-review under responsibility of the scientific committee of the International Conference on Advances in Materials Research-2019.

关键词： Soil temperature Artificial neural network Seasonal variation back propagation algorithm Statistical validation

来源：评论

学校读者我要写书评

暂无评论

A Neural Network Approach for Selection of Powder Metallurgy Process Parameter of Rapidly Solidified White Cast Iron

A Neural Network Approach for Selection of Powder Metallurgy...

引用

World Congress on Nature and Biologically Inspired Computing

作者： Salwan, G. Mishra, D. Pani, S. VSS Univ Technol Burla India

ISBN: (纸本)9781424450534

Powder Metallurgy (P/M) involves multiple input and output which are non-linearly related for which statistical optimization methods are not suitable. These considerations lead to adoption of neural network (NN) for proper selection of P/M process parameter. In the present work, white cast iron powder is taken as the work material and NN approach is employed which allows specification of multiple input and generation of multiple output recommendations. The NN model developed for the purpose is based on three-layer resilient back propagation learning algorithm with the help of Mat lab NN Toolbox. Supervised training has been adopted to train the network which helps in prediction of process parameter such as sintered density and growth % at different compaction pressure and sintering temperature of consolidated water atomized rapidly solidified white cast iron. Training data are collected by the experimental setup in laboratory. The density and growth % predicted by NN model coincides well with the experimental data with a tolerable error of 1 x 10(-10) which confirms its capability over the standard design procedures. The mathematical model developed here can be used as knowledge based system for a number of P/M products.

关键词： powder metallurgy neural network back propagation algorithm supervised learning

来源：评论

学校读者我要写书评

暂无评论

Modelling volatility and forecasting of exchange rate of British pound sterling and Indian rupee

引用

JOURNAL OF MODELLING IN MANAGEMENT 2016年第2期11卷 389-404页

作者： Gupta, Sanjeev Kashyap, Sachin Cent Univ Himachal Pradesh Sch Business & Management Studies Dept Accounting & Finance TAB Shahpur Kangra India

Purpose - The paper aims to analyse the extent of volatility and generating forecasts of exchange rates of British pound and Indian rupees in US terms. Design/methodology/approach - This study applies different combinations of GARCH and EGARCH models suggested in the Econometric literature to capture the extent of volatility. The forecast of exchange rates of British Pound and Indian Rupees in US terms are generated applying artificial neural network (ANN) technique using different combination of networks with hyperbolic tangent function at hidden and output stage of the model. Findings - The presence of volatility depicts that there is noise and chaos in the forex market. Prediction of exchange rate of the respective currencies underscores that exchange rates will increase marginally in near future. Practical Implications - The results proposed in this study will be benchmark for the hedgers, investors, bankers, practitioners and economists to foresee the exchange rate in the presence of volatility and design policies accordingly. Originality/value - In literature, no study has applied ANN for forecasting exchange rate after measuring the extent of volatility. The present study is a unique contribution in the existing pool of literature to forecasts the concerned variable(s) after ascertaining the noise and chaos in the data by applying GARCH family models.

关键词： Forecasting Time series Exchange rate Neural networks back propagation algorithm Heteroscedasticity

来源：评论

学校读者我要写书评

暂无评论

A novel architecture of a personalized on-line testing system based on mobile agent and improved genetic algorithm

A novel architecture of a personalized on-line testing syste...

引用

International Symposium on Computer Science and Technology

作者： Wang, Hui Sun, Tieli NE Normal Univ Sch Comp Sci Changchun 130024 Peoples R China

ISBN: (纸本)0972147950

Many existing on-line testing systems cannot create adaptive examination papers that are fit for every student respectively. For traditional algorithms, the process of creating examination questions from question bank is lack of high efficiency and occupies too much memory and CPU time. The speed of data transmission is very slow when the flow of information in network is very high. In order to solve these problems, a novel architecture of a personalized on-line testing system is proposed in this paper. Firstly, the mobile agent technology is used in this system architecture to mitigate the burden of server and the flow of information in network. The tasks in different phrases of an testing procedure are assigned to several agents: master agent, paper generation agent, mobile agent, interface agent, testing agent, evaluation agent and result agent respectively. These agents can collaborate with each other during the exam period, which arranges the testing procedure in an orderly way. Secondly, we use the back propagation algorithm to evaluate the student's teaming status roundly. Thirdly, we present an improved genetic algorithm. By means of these algorithms, the system can provide objective assessments and personalized examination papers for students.

关键词： mobile agent genetic algorithm back propagation algorithm

来源：评论

学校读者我要写书评

暂无评论

Development of a hybrid neural network system for prediction of process parameters in injection moulding

Development of a hybrid neural network system for prediction...

引用

International conference on Advances in Materials and Processing Technology

作者： Yarlagadda, PKDV Khong, CAT Queensland Univ Technol Sch Mech Mfg & Med Engn Brisbane Qld Australia

In this paper, the attempts made by the authors to develop an artificial neural network system for prediction of injection moulding process parameters is presented. In this work, attempts have been made to determine the process parameters that could affect injection moulding process based on governing equations of the filling process. Focus is then directed to parameters that require the use of trial and error methods or other complex software to determine the process parameters. The two parameters that are predicted from the developed network are injection time and injection pressure. In this work, the training data are generated by simulation using C-MOLD flow simulation software. A total of 114 data were collected out of which 94 were used to train the network using MATLAB and the remaining 20 for testing the network. Two algorithms are used during the training phase, namely the error-back-propagation algorithm and the Levenberg-Marquardt approximation algorithm. Results showed that the latter algorithm is more suitable for this a plication since the Leverberg's algorithm converged rapidly with lesser training cycles when compared to the error-back-propagation algorithm. The accuracy of the developed network has been tested by predicting the injection pressure and injection time for few engineering components and found that the overall error is 0.93% with a deviation of 3.93%. (C) 2001 Elsevier Science B.V. All rights reserved.

关键词： artificial intelligence hybrid neural networks injection moulding back propagation algorithm

来源：评论

学校读者我要写书评

暂无评论

Hamming Code Performance Evaluation using Artificial Neural Network Decoder 15

Hamming Code Performance Evaluation using Artificial Neural ...

引用

15th International Conference on Engineering of Modern Electric Systems (EMES)

作者： Vaz, Aldrin Claytus Nayak, C. Gurudas Nayak, Dayananda Manipal Acad Higher Educ Manipal Inst Technol Dept Instrumentat & Control Enginering Manipal Karnataka India

ISBN: (纸本)9781728107738

With the increase in the connectivity among various electronic devices day-by-day, the technology has stepped up into a new era of Internet-of-Things. To ensure the accuracy, integrity and fault-tolerance in the transmitted data, Error Correcting Codes are used. Various techniques are available to decode the received data and correct the errors. In this paper, an approach based on Artificial Neural Networks (ANN) is been used to decode the received data because of their real-time operation, self-organization and adaptive learning. back propagation algorithm for feed forward ANN has been simulated using MATLAB for (7, 4) Hamming Code. The synaptic weights are updated during each training cycle. The designed ANN is trained for all possible combination of code words such that it can detect and correct 1-bit error. The Bit Error rate performance of the proposed ANN based method is compared with the syndrome decoding.

关键词： Artificial Neural Network back propagation algorithm Error Correcting Code Hamming Code

来源：评论

学校读者我要写书评

暂无评论

A face recognition method using artificial neural networks

A face recognition method using artificial neural networks

引用

2nd International Conference on Digital Image Processing

作者： Omidvar, Mohammadnabi Dezfouli, Mashaallah Abbasi Rahmani, Amirmasoud Azad Univ Dept Comp Sci Khozestan Iran

ISBN: (纸本)9780819479426

The present paper aims to introduce a new method of face recognition based on integrating the results of three different neural networks and discuss the final outcome from a fuzzy point of view (recognition classifier). The first merit of this method is that it is not relying on the positions of eyes and lip on an individual's face. The second is that even if the face is partially covered, the method appears fault tolerant. All the experiments of the study were carried out based on the ORL (Olivetti Research Laboratory) database with 5 training images. For the selected numbers of 20, 30, and 40 subjects, we came to the results of 94%, 92.5%, and 90.25% respectively.

关键词： Face recognition neural network back propagation algorithm neuron

来源：评论

学校读者我要写书评

暂无评论

Off-line Odia Handwritten Numeral Recognition Using Neural Network: A Comparative Analysis

Off-line Odia Handwritten Numeral Recognition Using Neural N...

引用

IEEE International Conference on Computing, Communication and Automation (ICCCA)

作者： Sethy, Abhisek Patra, Prashanta Kumar Coll Engn & Technol Dept Comp Sci & Engn Bhubaneswar Orissa India

ISBN: (纸本)9781509016662

Character recognition is one of the most interesting and challenging research areas in the field of image processing. The recognition rate of handwritten character is still limited due to the presence of large variety of shape, scale and format in hand written characters. A sophisticated handwritten character recognition system demands a better feature extraction technique that would take care of such variety of handwriting. The work proposed in this paper is an attempt to develop recognizer for Odia handwritten numeral digits based on Binarization and discrete cosine transform (DCT) scheme. The recognizer puts emphasis on exploiting the inherit characteristics of Odia numeral images. The system first employs the techniques like thinning, foreground and background noise removal, cropping and size normalization etc., to preprocess the character images. Binarization and DCT technique is employed separately to extract the features from the images. Subsequently, these feature vectors are sent to the neural network classifier. Extensive simulations show that the results are very promising over a standard dataset and the recognition rate for Binarization and DCT are 80.2% and 90%, respectively.

关键词： Character Recognition (CR) Binarization Discrete cosine transform (DCT) Feature Extraction back propagation algorithm Neural Network

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：