Abstract: A multilayer self organizing neural neural network
(MLSONN) architecture for binary object extraction, guided by a beta
activation function and characterized by backpropagation of errors
estimated from the linear indices of fuzziness of the network output
states, is discussed. Since the MLSONN architecture is designed to
operate in a single point fixed/uniform thresholding scenario, it does
not take into cognizance the heterogeneity of image information in
the extraction process. The performance of the MLSONN architecture
with representative values of the threshold parameters of the beta
activation function employed is also studied. A three layer bidirectional
self organizing neural network (BDSONN) architecture
comprising fully connected neurons, for the extraction of objects from
a noisy background and capable of incorporating the underlying image
context heterogeneity through variable and adaptive thresholding,
is proposed in this article. The input layer of the network architecture
represents the fuzzy membership information of the image scene to
be extracted. The second layer (the intermediate layer) and the final
layer (the output layer) of the network architecture deal with the self
supervised object extraction task by bi-directional propagation of the
network states. Each layer except the output layer is connected to the
next layer following a neighborhood based topology. The output layer
neurons are in turn, connected to the intermediate layer following
similar topology, thus forming a counter-propagating architecture
with the intermediate layer. The novelty of the proposed architecture
is that the assignment/updating of the inter-layer connection weights
are done using the relative fuzzy membership values at the constituent
neurons in the different network layers. Another interesting feature
of the network lies in the fact that the processing capabilities of
the intermediate and the output layer neurons are guided by a beta
activation function, which uses image context sensitive adaptive
thresholding arising out of the fuzzy cardinality estimates of the
different network neighborhood fuzzy subsets, rather than resorting to
fixed and single point thresholding. An application of the proposed
architecture for object extraction is demonstrated using a synthetic
and a real life image. The extraction efficiency of the proposed
network architecture is evaluated by a proposed system transfer index
characteristic of the network.
Abstract: Investigation of soil properties like Cation Exchange
Capacity (CEC) plays important roles in study of environmental
reaserches as the spatial and temporal variability of this property
have been led to development of indirect methods in estimation of
this soil characteristic. Pedotransfer functions (PTFs) provide an
alternative by estimating soil parameters from more readily available
soil data. 70 soil samples were collected from different horizons of
15 soil profiles located in the Ziaran region, Qazvin province, Iran.
Then, multivariate regression and neural network model (feedforward
back propagation network) were employed to develop a
pedotransfer function for predicting soil parameter using easily
measurable characteristics of clay and organic carbon. The
performance of the multivariate regression and neural network model
was evaluated using a test data set. In order to evaluate the models,
root mean square error (RMSE) was used. The value of RMSE and
R2 derived by ANN model for CEC were 0.47 and 0.94 respectively,
while these parameters for multivariate regression model were 0.65
and 0.88 respectively. Results showed that artificial neural network
with seven neurons in hidden layer had better performance in
predicting soil cation exchange capacity than multivariate regression.
Abstract: In this paper, an approach to reduce the computation steps required by fast neural networksfor the searching process is presented. The principle ofdivide and conquer strategy is applied through imagedecomposition. Each image is divided into small in sizesub-images and then each one is tested separately usinga fast neural network. The operation of fast neuralnetworks based on applying cross correlation in thefrequency domain between the input image and theweights of the hidden neurons. Compared toconventional and fast neural networks, experimentalresults show that a speed up ratio is achieved whenapplying this technique to locate human facesautomatically in cluttered scenes. Furthermore, fasterface detection is obtained by using parallel processingtechniques to test the resulting sub-images at the sametime using the same number of fast neural networks. Incontrast to using only fast neural networks, the speed upratio is increased with the size of the input image whenusing fast neural networks and image decomposition.
Abstract: In this paper, we study the application of Extreme
Learning Machine (ELM) algorithm for single layered feedforward
neural networks to non-linear chaotic time series problems. In this
algorithm the input weights and the hidden layer bias are randomly
chosen. The ELM formulation leads to solving a system of linear
equations in terms of the unknown weights connecting the hidden
layer to the output layer. The solution of this general system of
linear equations will be obtained using Moore-Penrose generalized
pseudo inverse. For the study of the application of the method we
consider the time series generated by the Mackey Glass delay
differential equation with different time delays, Santa Fe A and
UCR heart beat rate ECG time series. For the choice of sigmoid,
sin and hardlim activation functions the optimal values for the
memory order and the number of hidden neurons which give the
best prediction performance in terms of root mean square error are
determined. It is observed that the results obtained are in close
agreement with the exact solution of the problems considered
which clearly shows that ELM is a very promising alternative
method for time series prediction.
Abstract: The development of Artificial Neural Networks
(ANNs) is usually a slow process in which the human expert has to
test several architectures until he finds the one that achieves best
results to solve a certain problem. This work presents a new
technique that uses Genetic Programming (GP) for automatically
generating ANNs. To do this, the GP algorithm had to be changed in
order to work with graph structures, so ANNs can be developed. This
technique also allows the obtaining of simplified networks that solve
the problem with a small group of neurons. In order to measure the
performance of the system and to compare the results with other
ANN development methods by means of Evolutionary Computation
(EC) techniques, several tests were performed with problems based
on some of the most used test databases. The results of those
comparisons show that the system achieves good results comparable
with the already existing techniques and, in most of the cases, they
worked better than those techniques.
Abstract: The objective of this paper is to develop a neural
network-based residual generator to detect the fault in the actuators
for a specific communication satellite in its attitude control system
(ACS). First, a dynamic multilayer perceptron network with dynamic
neurons is used, those neurons correspond a second order linear
Infinite Impulse Response (IIR) filter and a nonlinear activation
function with adjustable parameters. Second, the parameters from the
network are adjusted to minimize a performance index specified by
the output estimated error, with the given input-output data collected
from the specific ACS. Then, the proposed dynamic neural network
is trained and applied for detecting the faults injected to the wheel,
which is the main actuator in the normal mode for the communication
satellite. Then the performance and capabilities of the proposed
network were tested and compared with a conventional model-based
observer residual, showing the differences between these two
methods, and indicating the benefit of the proposed algorithm to
know the real status of the momentum wheel. Finally, the application
of the methods in a satellite ground station is discussed.
Abstract: This paper presents a general trainable framework
for fast and robust upright human face and non-human object
detection and verification in static images. To enhance the
performance of the detection process, the technique we develop is
based on the combination of fast neural network (FNN) and
classical neural network (CNN). In FNN, a useful correlation is
exploited to sustain high level of detection accuracy between input
image and the weight of the hidden neurons. This is to enable the
use of Fourier transform that significantly speed up the time
detection. The combination of CNN is responsible to verify the
face region. A bootstrap algorithm is used to collect non human
object, which adds the false detection to the training process of the
human and non-human object. Experimental results on test images
with both simple and complex background demonstrate that the
proposed method has obtained high detection rate and low false
positive rate in detecting both human face and non-human object.
Abstract: The neurogenic potential of many herbal extracts used
in Indian medicine is hitherto unknown. Extracts derived from
Clitoria ternatea Linn have been used in Indian Ayurvedic system of
medicine as an ingredient of “Medhya rasayana", consumed for
improving memory and longevity in humans and also in treatment of
various neurological disorders. Our earlier experimental studies with
oral intubation of Clitoria ternatea aqueous root extract (CTR) had
shown significant enhancement of learning and memory in postnatal
and young adult Wistar rats. The present study was designed to
elucidate the in vitro effects of 200ng/ml of CTR on proliferation,
differentiation and growth of anterior subventricular zone neural
stem cells (aSVZ NSC-s) derived from prenatal and postnatal rat
pups. Results show significant increase in proliferation and growth of
neurospheres and increase in the yield of differentiated neurons of
aSVZ neural precursor cells (aSVZNPC-s) at 7 days in vitro when
treated with 200ng/ml of CTR as compared to age matched control.
Results indicate that CTR has growth promoting neurogenic effect on
aSVZ neural stem cells and their survival similar to neurotrophic
factors like Survivin, Neuregulin 1, FGF-2, BDNF possibly the basis
for enhanced learning and memory.
Abstract: In this paper, a model of self-organizing spiking neural networks is introduced and applied to mobile robot environment representation and path planning problem. A network of spike-response-model neurons with a recurrent architecture is used to create robot-s internal representation from surrounding environment. The overall activity of network simulates a self-organizing system with unsupervised learning. A modified A* algorithm is used to find the best path using this internal representation between starting and goal points. This method can be used with good performance for both known and unknown environments.
Abstract: Security has been an important issue and concern in the
smart home systems. Smart home networks consist of a wide range of
wired or wireless devices, there is possibility that illegal access to
some restricted data or devices may happen. Password-based
authentication is widely used to identify authorize users, because this
method is cheap, easy and quite accurate. In this paper, a neural
network is trained to store the passwords instead of using verification
table. This method is useful in solving security problems that
happened in some authentication system. The conventional way to
train the network using Backpropagation (BPN) requires a long
training time. Hence, a faster training algorithm, Resilient
Backpropagation (RPROP) is embedded to the MLPs Neural
Network to accelerate the training process. For the Data Part, 200
sets of UserID and Passwords were created and encoded into binary
as the input. The simulation had been carried out to evaluate the
performance for different number of hidden neurons and combination
of transfer functions. Mean Square Error (MSE), training time and
number of epochs are used to determine the network performance.
From the results obtained, using Tansig and Purelin in hidden and
output layer and 250 hidden neurons gave the better performance. As
a result, a password-based user authentication system for smart home
by using neural network had been developed successfully.
Abstract: Renewable energy resources are inexhaustible, clean as compared with conventional resources. Also, it is used to supply regions with no grid, no telephone lines, and often with difficult accessibility by common transport. Satellite earth stations which located in remote areas are the most important application of renewable energy. Neural control is a branch of the general field of intelligent control, which is based on the concept of artificial intelligence. This paper presents the mathematical modeling of satellite earth station power system which is required for simulating the system.Aswan is selected to be the site under consideration because it is a rich region with solar energy. The complete power system is simulated using MATLAB–SIMULINK.An artificial neural network (ANN) based model has been developed for the optimum operation of earth station power system. An ANN is trained using a back propagation with Levenberg–Marquardt algorithm. The best validation performance is obtained for minimum mean square error. The regression between the network output and the corresponding target is equal to 96% which means a high accuracy. Neural network controller architecture gives satisfactory results with small number of neurons, hence better in terms of memory and time are required for NNC implementation. The results indicate that the proposed control unit using ANN can be successfully used for controlling the satellite earth station power system.
Abstract: The self-organizing map (SOM) model is a well-known neural network model with wide spread of applications. The main characteristics of SOM are two-fold, namely dimension reduction and topology preservation. Using SOM, a high-dimensional data space will be mapped to some low-dimensional space. Meanwhile, the topological relations among data will be preserved. With such characteristics, the SOM was usually applied on data clustering and visualization tasks. However, the SOM has main disadvantage of the need to know the number and structure of neurons prior to training, which are difficult to be determined. Several schemes have been proposed to tackle such deficiency. Examples are growing/expandable SOM, hierarchical SOM, and growing hierarchical SOM. These schemes could dynamically expand the map, even generate hierarchical maps, during training. Encouraging results were reported. Basically, these schemes adapt the size and structure of the map according to the distribution of training data. That is, they are data-driven or dataoriented SOM schemes. In this work, a topic-oriented SOM scheme which is suitable for document clustering and organization will be developed. The proposed SOM will automatically adapt the number as well as the structure of the map according to identified topics. Unlike other data-oriented SOMs, our approach expands the map and generates the hierarchies both according to the topics and their characteristics of the neurons. The preliminary experiments give promising result and demonstrate the plausibility of the method.
Abstract: The competitive learning is an adaptive process in
which the neurons in a neural network gradually become sensitive to
different input pattern clusters. The basic idea behind the Kohonen-s
Self-Organizing Feature Maps (SOFM) is competitive learning.
SOFM can generate mappings from high-dimensional signal spaces
to lower dimensional topological structures. The main features of this
kind of mappings are topology preserving, feature mappings and
probability distribution approximation of input patterns. To overcome
some limitations of SOFM, e.g., a fixed number of neural units and a
topology of fixed dimensionality, Growing Self-Organizing Neural
Network (GSONN) can be used. GSONN can change its topological
structure during learning. It grows by learning and shrinks by
forgetting. To speed up the training and convergence, a new variant
of GSONN, twin growing cell structures (TGCS) is presented here.
This paper first gives an introduction to competitive learning, SOFM
and its variants. Then, we discuss some GSONN with fixed
dimensionality, which include growing cell structures, its variants
and the author-s model: TGCS. It is ended with some testing results
comparison and conclusions.
Abstract: For over a decade, the Pulse Coupled Neural Network
(PCNN) based algorithms have been successfully used in image
interpretation applications including image segmentation. There are
several versions of the PCNN based image segmentation methods,
and the segmentation accuracy of all of them is very sensitive to the
values of the network parameters. Most methods treat PCNN
parameters like linking coefficient and primary firing threshold as
global parameters, and determine them by trial-and-error. The
automatic determination of appropriate values for linking coefficient,
and primary firing threshold is a challenging problem and deserves
further research. This paper presents a method for obtaining global as
well as local values for the linking coefficient and the primary firing
threshold for neurons directly from the image statistics. Extensive
simulation results show that the proposed approach achieves
excellent segmentation accuracy comparable to the best accuracy
obtainable by trial-and-error for a variety of images.
Abstract: Space Vector Modulation (SVM) is an optimum Pulse Width Modulation (PWM) technique for an inverter used in a variable frequency drive applications. It is computationally rigorous and hence limits the inverter switching frequency. Increase in switching frequency can be achieved using Neural Network (NN) based SVM, implemented on application specific chips. This paper proposes a neural network based SVM technique for a Voltage Source Inverter (VSI). The network proposed is independent of switching frequency. Different architectures are investigated keeping the total number of neurons constant. The performance of the inverter is compared for various switching frequencies for different architectures of NN based SVM. From the results obtained, the network with minimum resource and appropriate word length is identified. The bit precision required for this application is identified. The network with 8-bit precision is implemented in the IC XCV 400 and the results are presented. The performance of NN based general purpose SVM with higher bit precision is discussed.
Abstract: This research paper deals with the implementation of face recognition using neural network (recognition classifier) on low-resolution images. The proposed system contains two parts, preprocessing and face classification. The preprocessing part converts original images into blurry image using average filter and equalizes the histogram of those image (lighting normalization). The bi-cubic interpolation function is applied onto equalized image to get resized image. The resized image is actually low-resolution image providing faster processing for training and testing. The preprocessed image becomes the input to neural network classifier, which uses back-propagation algorithm to recognize the familiar faces. The crux of proposed algorithm is its beauty to use single neural network as classifier, which produces straightforward approach towards face recognition. The single neural network consists of three layers with Log sigmoid, Hyperbolic tangent sigmoid and Linear transfer function respectively. The training function, which is incorporated in our work, is Gradient descent with momentum (adaptive learning rate) back propagation. The proposed algorithm was trained on ORL (Olivetti Research Laboratory) database with 5 training images. The empirical results provide the accuracy of 94.50%, 93.00% and 90.25% for 20, 30 and 40 subjects respectively, with time delay of 0.0934 sec per image.
Abstract: This paper presents a critical study about the
application of Neural Networks to ion-exchange process. Ionexchange
is a complex non-linear process involving many factors
influencing the ions uptake mechanisms from the pregnant solution.
The following step includes the elution. Published data presents
empirical isotherm equations with definite shortcomings resulting in
unreliable predictions. Although Neural Network simulation
technique encounters a number of disadvantages including its “black
box", and a limited ability to explicitly identify possible causal
relationships, it has the advantage to implicitly handle complex
nonlinear relationships between dependent and independent
variables. In the present paper, the Neural Network model based on
the back-propagation algorithm Levenberg-Marquardt was developed
using a three layer approach with a tangent sigmoid transfer function
(tansig) at hidden layer with 11 neurons and linear transfer function
(purelin) at out layer. The above mentioned approach has been used
to test the effectiveness in simulating ion exchange processes. The
modeling results showed that there is an excellent agreement between
the experimental data and the predicted values of copper ions
removed from aqueous solutions.
Abstract: The present study focuses on the discussion over the
parameter of Artificial Neural Network (ANN). Sensitivity analysis is
applied to assess the effect of the parameters of ANN on the prediction
of turbidity of raw water in the water treatment plant. The result shows
that transfer function of hidden layer is a critical parameter of ANN.
When the transfer function changes, the reliability of prediction of
water turbidity is greatly different. Moreover, the estimated water
turbidity is less sensitive to training times and learning velocity than
the number of neurons in the hidden layer. Therefore, it is important to
select an appropriate transfer function and suitable number of neurons
in the hidden layer in the process of parameter training and validation.
Abstract: A neuron can emit spikes in an irregular time basis and by averaging over a certain time window one would ignore a lot of information. It is known that in the context of fast information processing there is no sufficient time to sample an average firing rate of the spiking neurons. The present work shows that the spiking neurons are capable of computing the radial basis functions by storing the relevant information in the neurons' delays. One of the fundamental findings of the this research also is that when using overlapping receptive fields to encode the data patterns it increases the network-s clustering capacity. The clustering algorithm that is discussed here is interesting from computer science and neuroscience point of view as well as from a perspective.
Abstract: Functional Magnetic Resonance Imaging(fMRI) is a
noninvasive imaging technique that measures the hemodynamic
response related to neural activity in the human brain. Event-related
functional magnetic resonance imaging (efMRI) is a form of
functional Magnetic Resonance Imaging (fMRI) in which a series of
fMRI images are time-locked to a stimulus presentation and averaged
together over many trials. Again an event related potential (ERP) is a
measured brain response that is directly the result of a thought or
perception. Here the neuronal response of human visual cortex in
normal healthy patients have been studied. The patients were asked
to perform a visual three choice reaction task; from the relative
response of each patient corresponding neuronal activity in visual
cortex was imaged. The average number of neurons in the adult
human primary visual cortex, in each hemisphere has been estimated
at around 140 million. Statistical analysis of this experiment was
done with SPM5(Statistical Parametric Mapping version 5) software.
The result shows a robust design of imaging the neuronal activity of
human visual cortex.