Error correction codes such as low density parity check (LDPC) codes are popularly used to enhance the performance of digital communication systems. The current decoding framework relies on exchanging beliefs over a T...
详细信息
ISBN:
(纸本)9781728119854
Error correction codes such as low density parity check (LDPC) codes are popularly used to enhance the performance of digital communication systems. The current decoding framework relies on exchanging beliefs over a Tanner graph, which the encoder and decoder are aware of. However, this information may not be available readily, for example in covert communication. The main idea of this paper is to build a neural network to learn the encoder mappings in the absence of knowledge of the Tanner graph. We propose a scheme to learn the mappings using the back propagation algorithm. We investigate into the choice of different cost functions and the number of hidden neurons for learning the encoding function. The proposed scheme is capable of learning the parity check equations over a binary field towards identifying the validity of a codeword. Simulation results over synthetic data show that our algorithm is indeed capable of learning the encoder mappings and identifying the parity check equations.
Building an effective methodology to detect characters from images with less error rate is the great task. Our aim is to furnish such an algorithm that will be able to generate error free recognition of text from the ...
详细信息
Building an effective methodology to detect characters from images with less error rate is the great task. Our aim is to furnish such an algorithm that will be able to generate error free recognition of text from the given input image which will help in document digitizing and prevention to the hand written text recognition. OCR has been in the intensive research topic for more than 4 decades, it is probably the most time consuming and labor intensive work of inputting the data through keyboard. This paper discuss about mechanical or electronic conversion of scanned images, text which contain graphics, image captured by camera, scanned images and the recognition of images where characters may be broken or smeared. The optical character recognition is the desktop based application developed using Java IDE and mysql as a database. We have gain 91.82% accuracy when applied on different data sets, in pre-processing we used different techniques to remove noise from the image in post processing we used dictionary for the characters which are not recognized during classification, in classification we have used the back propagation algorithm for the training of neural network, feature extraction has been performed by template matching and hamming distance. All the algorithms have been developed in java technology.
Recently a commonly used method for Recognition of Handwritten Digit Application based on backpropagation Neural Network (BPNN) has been widely applied. However, the original algorithm and its modifications contains ...
详细信息
ISBN:
(纸本)9781467348430
Recently a commonly used method for Recognition of Handwritten Digit Application based on backpropagation Neural Network (BPNN) has been widely applied. However, the original algorithm and its modifications contains a number of free parameters which affect particular networks differently and the slight error rate on the selection of these parameters can cause problems. Thus, this paper presents the effect of input parameters on BPNN with three different structures including Simple backpropagation, backpropagation with momentum terms and backpropagation using conjugate gradient descent methods. To do so, this paper determined different parameters such as learning rate, momentum term or even the number of units in the hidden layer that exist in each structure. The data of UCI database is used for experiment in MATLAB program. The result showed that the backpropagation with momentum term could perform very well leading to a recognition rate of 99%. The Simple algorithm obtained high recognition rate but it needed to increase learning rate, while backpropagation using conjugate gradient descent could provide high result in case of improving hidden neural nodes. Thus, the result confirmed that adjustment of the relevant parameters are significant to obtain better recognition effect and higher accuracy.
This paper presents the forecasting of wind energy using neural networks based on the speed and the rotational speed of the rotor. The wind power mainly depends on the velocity of the wind, density, and swept area of ...
详细信息
ISBN:
(纸本)9781538693469
This paper presents the forecasting of wind energy using neural networks based on the speed and the rotational speed of the rotor. The wind power mainly depends on the velocity of the wind, density, and swept area of the wind mill which depends on the radius of the rotor. The power also depends on the efficiency of the motor, gear mechanisms. Hence the power varies according to the various parameters which makes the model non-linear. In this paper neural network is used to predict the output from the previous set of data. A new set of data is chosen and tested for the accuracy. It can be seen that the back propagation algorithm in neural network is able to classify the power output based on the speed of the rotor and the wind velocity.
In the past years modern arithmetical methods for image investigation have led to a rebellion in many fields, from computer vision to scientific imaging. Though, some recently developed image processing techniques suc...
详细信息
ISBN:
(纸本)9781509012855
In the past years modern arithmetical methods for image investigation have led to a rebellion in many fields, from computer vision to scientific imaging. Though, some recently developed image processing techniques successfully oppressed by other sections have been infrequently, if ever, experimented on celestial observations. Here we present a new idea of super resolution of Astronomical objects using back propagation algorithm."Super-resolution " is efficient in improving the excellence of analysis of diffused sources formerly unobserved by the background noise, efficiently rising the depth of obtainable observations. Higher-resolution image out of a set of low resolution frames can be obtained through super-resolution. Super-resolution is viable only for point sources which have negligible dimensions, then for wide-ranging objects the knowledge about intensity vacillation at angular prevalence is irreversibly mislaid. Again obtaining super resolved image for extended sources(e.g. comets, meteoroids, etc) is a new challenge if the speed of the object is very high. Acquiring High resolution images of celestial objects from ground based telescopes is intricate and often requires computational post processing techniques to remove blur caused by atmospheric commotion. Even images obtained through satellite imaging are compressed and sent to earth. So there is need for Super resolution of those compressed or noisy images. So, here we simply implement Super-resolution for Astronomical objects using back propagation algorithm to overcome lost information and challenges for high speedy celestial objects. The purpose is to super resolve high speedy celestial objects whose analysis may in future help to prevent collisions of such celestial objects with earth and also avoid future solar system damage
The back propagation algorithm has wide range of applications for training of feed forward neural networks. Over the years, many researchers have used back propagation algorithm to train their neural network based sys...
详细信息
ISBN:
(纸本)9781509052561
The back propagation algorithm has wide range of applications for training of feed forward neural networks. Over the years, many researchers have used back propagation algorithm to train their neural network based systems without emphasizing on how to fine tune the parameters of the algorithm. The paper throws the light on how researchers can manipulate and experiment with the parameters of the back propagation algorithm to achieve the optimum learning performance. The paper presents the results of the laboratory experiments of fine tuning the parameters of the back propagation algorithm. The process of fine tuning the parameters was applied on the neural network based expert system prototype. The prototype aims to analyze and design customized motivational strategies based on employees' perspective. The laboratory experiments were conducted on the following parameters of back propagation algorithm: learning rate, momentum rate and activation functions. Learning performance are measured and recorded. At the same time, the impact of activation function on the final output is also measured. Based on the results, the values of the above parameters which provide the optimum learning performance is chosen for the full scale system implementation.
This paper proposes a new method using back propagation algorithm and equal area criterion for fast transient stability assessment. The basic concept of the proposed method is shown for a one-machine with infinite-bus...
详细信息
This paper proposes a new method using back propagation algorithm and equal area criterion for fast transient stability assessment. The basic concept of the proposed method is shown for a one-machine with infinite-bus system. Then, examples of transient stability assessment are shown to demonstrate the powerfulness of the method in three categories, i.e., stability assessment, stability limits and parametric study. Finally, the example of stability limits calculation using back propagation algorithm is described in detail.
The ANN with back propagation algorithm is a multi-layer feed-forward neural network, which is suitable to study unsteady frost formation with multiple factors. The backpropagation ANN algorithm is used to study fros...
详细信息
The ANN with back propagation algorithm is a multi-layer feed-forward neural network, which is suitable to study unsteady frost formation with multiple factors. The backpropagation ANN algorithm is used to study frost layer growth on cold flat surface, where four feature variables including temperature of cold flat sur-face, the velocity, relative humidity, and temperature of air are adopted. The frost growth experiment generates the database, which is good for training frost growth due to its fast speed and high precision based on Levenberg-Marquardt learning rule. The establishment of neural network model in this paper can quickly and accurately predict the frost layer height on cold flat surface of different control variables, which is helpful for the implementation of defrosting.
X-ray diffractometry is a unique technique and that the X-ray diffraction patterns which depict the structure of the steel sheets during processing with the features extracted, that they serve directly as a signature ...
详细信息
X-ray diffractometry is a unique technique and that the X-ray diffraction patterns which depict the structure of the steel sheets during processing with the features extracted, that they serve directly as a signature which is very complicated. X-ray diffraction (XRD) techniques are a type of non-destructive method of investigation to identify the flaws during the fabrication of steel sheets. X-ray diffraction is comparatively simple and can be effectively used for the examination and identification of flaws during the rolling process of steel sheets. XRD technique finds application in various fields like textile industry, forensic, qualitative and quantitative phase analysis of poly crystalline material, to infer overall properties of the fiber and measure the degree of crystalline nature. It is extensively used to explore areas like material science, chemistry and in industry for research and quality control. This effective method gains novelty by combining the signal processing algorithms like multiple threshold based Fast Fourier Transform (FFT) and Artificial neural network (ANN) trained with back propagation algorithm (BPA) thereby offering an automated system for online monitoring during fabrication of flawless metal sheets. The hot rolled steel sheets for three categories namely, flawless, moderate flaw and extreme flaw conditions are obtained from the XRD pattern. Then multiple thresholds are incorporated to identify the peak position, peak width and peak intensity. The FFT algorithm computes the power spectrum which is used as features to identify the flaws in the steel sheets during cold rolling process. The extracted features are used as inputs to train the ANN with BPA whose performance is evaluated to be 90% efficient.
A multilayer perceptron neural network system is established to support the diagnosis for five most common heart diseases (coronary heart disease, rheumatic valvular heart disease, hypertension, chronic cor pulmonale ...
详细信息
A multilayer perceptron neural network system is established to support the diagnosis for five most common heart diseases (coronary heart disease, rheumatic valvular heart disease, hypertension, chronic cor pulmonale and congenital heart disease). Momentum term, adaptive learning rate, the forgetting mechanics, and conjugate gradients method are introduced to improve the basic BP algorithm aiming to speed up the convergence of the BP algorithm and enhance the accuracy for diagnosis. A heart disease database consisting of 352 samples is applied to the training and testing courses of the system. The performance of the system is assessed by cross-validation method. It is found that as the basic BP algorithm is improved step by step, the convergence speed and the classification accuracy of the network are enhanced, and the system has great application prospect in supporting heart diseases diagnosis.
暂无评论