Abstract: The Multi-Layered Perceptron (MLP) Neural
networks have been very successful in a number of signal processing
applications. In this work we have studied the possibilities and the
met difficulties in the application of the MLP neural networks for the
prediction of daily solar radiation data. We have used the Polack-Ribière algorithm for training the neural networks. A comparison, in
term of the statistical indicators, with a linear model most used in
literature, is also performed, and the obtained results show that the
neural networks are more efficient and gave the best results.
Abstract: The speech signal conveys information about the
identity of the speaker. The area of speaker identification is
concerned with extracting the identity of the person speaking the
utterance. As speech interaction with computers becomes more
pervasive in activities such as the telephone, financial transactions
and information retrieval from speech databases, the utility of
automatically identifying a speaker is based solely on vocal
characteristic. This paper emphasizes on text dependent speaker
identification, which deals with detecting a particular speaker from a
known population. The system prompts the user to provide speech
utterance. System identifies the user by comparing the codebook of
speech utterance with those of the stored in the database and lists,
which contain the most likely speakers, could have given that speech
utterance. The speech signal is recorded for N speakers further the
features are extracted. Feature extraction is done by means of LPC
coefficients, calculating AMDF, and DFT. The neural network is
trained by applying these features as input parameters. The features
are stored in templates for further comparison. The features for the
speaker who has to be identified are extracted and compared with the
stored templates using Back Propogation Algorithm. Here, the
trained network corresponds to the output; the input is the extracted
features of the speaker to be identified. The network does the weight
adjustment and the best match is found to identify the speaker. The
number of epochs required to get the target decides the network
performance.
Abstract: Choosing the right metadata is a critical, as good
information (metadata) attached to an image will facilitate its
visibility from a pile of other images. The image-s value is enhanced
not only by the quality of attached metadata but also by the technique
of the search. This study proposes a technique that is simple but
efficient to predict a single human image from a website using the
basic image data and the embedded metadata of the image-s content
appearing on web pages. The result is very encouraging with the
prediction accuracy of 95%. This technique may become a great
assist to librarians, researchers and many others for automatically and
efficiently identifying a set of human images out of a greater set of
images.
Abstract: In a particular case of behavioural model reduction by ANNs, a validity domain shortening has been found. In mechanics, as in other domains, the notion of validity domain allows the engineer to choose a valid model for a particular analysis or simulation. In the study of mechanical behaviour for a cantilever beam (using linear and non-linear models), Multi-Layer Perceptron (MLP) Backpropagation (BP) networks have been applied as model reduction technique. This reduced model is constructed to be more efficient than the non-reduced model. Within a less extended domain, the ANN reduced model estimates correctly the non-linear response, with a lower computational cost. It has been found that the neural network model is not able to approximate the linear behaviour while it does approximate the non-linear behaviour very well. The details of the case are provided with an example of the cantilever beam behaviour modelling.
Abstract: In this paper we present an adaptive method for image
compression that is based on complexity level of the image. The
basic compressor/de-compressor structure of this method is a multilayer
perceptron artificial neural network. In adaptive approach
different Back-Propagation artificial neural networks are used as
compressor and de-compressor and this is done by dividing the
image into blocks, computing the complexity of each block and then
selecting one network for each block according to its complexity
value. Three complexity measure methods, called Entropy, Activity
and Pattern-based are used to determine the level of complexity in
image blocks and their ability in complexity estimation are evaluated
and compared. In training and evaluation, each image block is
assigned to a network based on its complexity value. Best-SNR is
another alternative in selecting compressor network for image blocks
in evolution phase which chooses one of the trained networks such
that results best SNR in compressing the input image block. In our
evaluations, best results are obtained when overlapping the blocks is
allowed and choosing the networks in compressor is based on the
Best-SNR. In this case, the results demonstrate superiority of this
method comparing with previous similar works and JPEG standard
coding.
Abstract: In this paper in consideration of each available
techniques deficiencies for speech recognition, an advanced method
is presented that-s able to classify speech signals with the high
accuracy (98%) at the minimum time. In the presented method, first,
the recorded signal is preprocessed that this section includes
denoising with Mels Frequency Cepstral Analysis and feature
extraction using discrete wavelet transform (DWT) coefficients; Then
these features are fed to Multilayer Perceptron (MLP) network for
classification. Finally, after training of neural network effective
features are selected with UTA algorithm.
Abstract: Fluid flow and heat transfer of vertical full cone
embedded in porous media is studied in this paper. Nonlinear
differential equation arising from similarity solution of inverted cone
(subjected to wall temperature boundary conditions) embedded in
porous medium is solved using a hybrid neural network- particle
swarm optimization method.
To aim this purpose, a trial solution of the differential equation is
defined as sum of two parts. The first part satisfies the initial/
boundary conditions and does contain an adjustable parameter and
the second part which is constructed so as not to affect the
initial/boundary conditions and involves adjustable parameters (the
weights and biases) for a multi-layer perceptron neural network.
Particle swarm optimization (PSO) is applied to find adjustable
parameters of trial solution (in first and second part). The obtained
solution in comparison with the numerical ones represents a
remarkable accuracy.
Abstract: Research into the problem of classification of sonar signals has been taken up as a challenging task for the neural networks. This paper investigates the design of an optimal classifier using a Multi layer Perceptron Neural Network (MLP NN) and Support Vector Machines (SVM). Results obtained using sonar data sets suggest that SVM classifier perform well in comparison with well-known MLP NN classifier. An average classification accuracy of 91.974% is achieved with SVM classifier and 90.3609% with MLP NN classifier, on the test instances. The area under the Receiver Operating Characteristics (ROC) curve for the proposed SVM classifier on test data set is found as 0.981183, which is very close to unity and this clearly confirms the excellent quality of the proposed classifier. The SVM classifier employed in this paper is implemented using kernel Adatron algorithm is seen to be robust and relatively insensitive to the parameter initialization in comparison to MLP NN.
Abstract: Development of artificial neural network (ANN) for
prediction of aluminum workpieces' surface roughness in ultrasonicvibration
assisted turning (UAT) has been the subject of the present
study. Tool wear as the main cause of surface roughness was also
investigated. ANN was trained through experimental data obtained
on the basis of full factorial design of experiments. Various
influential machining parameters were taken into consideration. It
was illustrated that a multilayer perceptron neural network could
efficiently model the surface roughness as the response of the
network, with an error less than ten percent. The performance of the
trained network was verified by further experiments. The results of
UAT were compared with the results of conventional turning
experiments carried out with similar machining parameters except for
the vibration amplitude whence considerable reduction was observed
in the built-up edge and the surface roughness.
Abstract: This work presents a novel means of extracting fixedlength parameters from voice signals, such that words can be recognized
in linear time. The power and the zero crossing rate are first
calculated segment by segment from a voice signal; by doing so, two
feature sequences are generated. We then construct an FIR system
across these two sequences. The parameters of this FIR system, used
as the input of a multilayer proceptron recognizer, can be derived by
recursive LSE (least-square estimation), implying that the complexity of overall process is linear to the signal size. In the second part of
this work, we introduce a weighting factor λ to emphasize recent
input; therefore, we can further recognize continuous speech signals.
Experiments employ the voice signals of numbers, from zero to nine, spoken in Mandarin Chinese. The proposed method is verified to
recognize voice signals efficiently and accurately.
Abstract: The objective of this paper is to propose an adaptive multi threshold for image segmentation precisely in object detection. Due to the different types of license plates being used, the requirement of an automatic LPR is rather different for each country. The proposed technique is applied on Malaysian LPR application. It is based on Multi Layer Perceptron trained by back propagation. The proposed adaptive threshold is introduced to find the optimum threshold values. The technique relies on the peak value from the graph of the number object versus specific range of threshold values. The proposed approach has improved the overall performance compared to current optimal threshold techniques. Further improvement on this method is in progress to accommodate real time system specification.
Abstract: This paper describes a new supervised fusion (hybrid)
electrocardiogram (ECG) classification solution consisting of a new
QRS complex geometrical feature extraction as well as a new version
of the learning vector quantization (LVQ) classification algorithm
aimed for overcoming the stability-plasticity dilemma. Toward this
objective, after detection and delineation of the major events of ECG
signal via an appropriate algorithm, each QRS region and also its
corresponding discrete wavelet transform (DWT) are supposed as
virtual images and each of them is divided into eight polar sectors.
Then, the curve length of each excerpted segment is calculated
and is used as the element of the feature space. To increase the
robustness of the proposed classification algorithm versus noise,
artifacts and arrhythmic outliers, a fusion structure consisting of
five different classifiers namely as Support Vector Machine (SVM),
Modified Learning Vector Quantization (MLVQ) and three Multi
Layer Perceptron-Back Propagation (MLP–BP) neural networks with
different topologies were designed and implemented. The new proposed
algorithm was applied to all 48 MIT–BIH Arrhythmia Database
records (within–record analysis) and the discrimination power of the
classifier in isolation of different beat types of each record was
assessed and as the result, the average accuracy value Acc=98.51%
was obtained. Also, the proposed method was applied to 6 number
of arrhythmias (Normal, LBBB, RBBB, PVC, APB, PB) belonging
to 20 different records of the aforementioned database (between–
record analysis) and the average value of Acc=95.6% was achieved.
To evaluate performance quality of the new proposed hybrid learning
machine, the obtained results were compared with similar peer–
reviewed studies in this area.
Abstract: In this paper, the requirement for Coke quality
prediction, its role in Blast furnaces, and the model output is
explained. By applying method of Artificial Neural Networking
(ANN) using back propagation (BP) algorithm, prediction model has
been developed to predict CSR. Important blast furnace functions
such as permeability, heat exchanging, melting, and reducing
capacity are mostly connected to coke quality. Coke quality is further
dependent upon coal characterization and coke making process
parameters. The ANN model developed is a useful tool for process
experts to adjust the control parameters in case of coke quality
deviations. The model also makes it possible to predict CSR for new
coal blends which are yet to be used in Coke Plant. Input data to the
model was structured into 3 modules, for tenure of past 2 years and
the incremental models thus developed assists in identifying the
group causing the deviation of CSR.
Abstract: Human identification at a distance has recently gained
growing interest from computer vision researchers. Gait recognition
aims essentially to address this problem by identifying people based
on the way they walk [1]. Gait recognition has 3 steps. The first step
is preprocessing, the second step is feature extraction and the third
one is classification. This paper focuses on the classification step that
is essential to increase the CCR (Correct Classification Rate).
Multilayer Perceptron (MLP) is used in this work. Neural Networks
imitate the human brain to perform intelligent tasks [3].They can
represent complicated relationships between input and output and
acquire knowledge about these relationships directly from the data
[2]. In this paper we apply MLP NN for 11 views in our database and
compare the CCR values for these views. Experiments are performed
with the NLPR databases, and the effectiveness of the proposed
method for gait recognition is demonstrated.
Abstract: This study presents a hybrid neural network and Gravitational Search Algorithm (HNGSA) method to solve well known Wessinger's equation. To aim this purpose, gravitational search algorithm (GSA) technique is applied to train a multi-layer perceptron neural network, which is used as approximation solution of the Wessinger's equation. A trial solution of the differential equation is written as sum of two parts. The first part satisfies the initial/ boundary conditions and does not contain any adjustable parameters and the second part which is constructed so as not to affect the initial/boundary conditions. The second part involves adjustable parameters (the weights and biases) for a multi-layer perceptron neural network. In order to demonstrate the presented method, the obtained results of the proposed method are compared with some known numerical methods. The given results show that presented method can introduce a closer form to the analytic solution than other numerical methods. Present method can be easily extended to solve a wide range of problems.
Abstract: Purpose of this work is to develop an automatic classification system that could be useful for radiologists in the breast cancer investigation. The software has been designed in the framework of the MAGIC-5 collaboration. In an automatic classification system the suspicious regions with high probability to include a lesion are extracted from the image as regions of interest (ROIs). Each ROI is characterized by some features based generally on morphological lesion differences. A study in the space features representation is made and some classifiers are tested to distinguish the pathological regions from the healthy ones. The results provided in terms of sensitivity and specificity will be presented through the ROC (Receiver Operating Characteristic) curves. In particular the best performances are obtained with the Neural Networks in comparison with the K-Nearest Neighbours and the Support Vector Machine: The Radial Basis Function supply the best results with 0.89 ± 0.01 of area under ROC curve but similar results are obtained with the Probabilistic Neural Network and a Multi Layer Perceptron.
Abstract: Active research is underway on virtual touch screens
that complement the physical limitations of conventional touch
screens. This paper discusses a virtual touch screen that uses a
multi-layer perceptron to recognize and control three-dimensional
(3D) depth information from a time of flight (TOF) camera. This
system extracts an object-s area from the image input and compares it
with the trajectory of the object, which is learned in advance, to
recognize gestures. The system enables the maneuvering of content in
virtual space by utilizing human actions.
Abstract: This paper illustrates the use of a combined neural
network model for classification of electrocardiogram (ECG) beats.
We present a trainable neural network ensemble approach to develop
customized electrocardiogram beat classifier in an effort to further
improve the performance of ECG processing and to offer
individualized health care.
We process a three stage technique for detection of premature
ventricular contraction (PVC) from normal beats and other heart
diseases. This method includes a denoising, a feature extraction and a
classification. At first we investigate the application of stationary
wavelet transform (SWT) for noise reduction of the
electrocardiogram (ECG) signals. Then feature extraction module
extracts 10 ECG morphological features and one timing interval
feature. Then a number of multilayer perceptrons (MLPs) neural
networks with different topologies are designed.
The performance of the different combination methods as well as
the efficiency of the whole system is presented. Among them,
Stacked Generalization as a proposed trainable combined neural
network model possesses the highest recognition rate of around 95%.
Therefore, this network proves to be a suitable candidate in ECG
signal diagnosis systems. ECG samples attributing to the different
ECG beat types were extracted from the MIT-BIH arrhythmia
database for the study.
Abstract: In this paper, the application of neural networks to study the design of short-term load forecasting (STLF) Systems for Illam state located in west of Iran was explored. One important architecture of neural networks named Multi-Layer Perceptron (MLP) to model STLF systems was used. Our study based on MLP was trained and tested using three years (2004-2006) data. The results show that MLP network has the minimum forecasting error and can be considered as a good method to model the STLF systems.
Abstract: Inverse kinematics analysis plays an important role in developing a robot manipulator. But it is not too easy to derive the inverse kinematic equation of a robot manipulator especially robot manipulator which has numerous degree of freedom. This paper describes an application of Artificial Neural Network for modeling the inverse kinematics equation of a robot manipulator. In this case, the robot has three degree of freedoms and the robot was implemented for drilling a printed circuit board. The artificial neural network architecture used for modeling is a multilayer perceptron networks with steepest descent backpropagation training algorithm. The designed artificial neural network has 2 inputs, 2 outputs and varies in number of hidden layer. Experiments were done in variation of number of hidden layer and learning rate. Experimental results show that the best architecture of artificial neural network used for modeling inverse kinematics of is multilayer perceptron with 1 hidden layer and 38 neurons per hidden layer. This network resulted a RMSE value of 0.01474.